genmoai
/
mochi-1-lora-trainer
a-r-r-o-w/cogvideox-factory for Mochi-1 LoRA Training
Prediction
genmoai/mochi-1-lora-trainer:170ea99fb48a30fef98cb1c9fb403a2882ab9d60c2ba15ad9383ace33c3fa385IDw6kpq5g261rme0ckps0rad27hcStatusSucceededSourceWebHardwareH100Total durationCreatedInput
- seed
- 42
- steps
- 1000
- hf_token
- ████████████████████
This value was redacted after being sent to the model.
- optimizer
- adamw
- batch_size
- 1
- hf_repo_id
- lucataco/mochi-lora-disney
- compile_dit
- input_videos
- disney-30.zip
- learning_rate
- 0.0004
- trim_and_crop
- caption_dropout
- 0.1
{ "seed": 42, "steps": 1000, "hf_token": "[REDACTED]", "optimizer": "adamw", "batch_size": 1, "hf_repo_id": "lucataco/mochi-lora-disney", "compile_dit": true, "input_videos": "https://replicate.delivery/pbxt/M7sXFavLoOaCdC6eorODzcq0ap6vQC5P10Ai12OzyVYiDUja/disney-30.zip", "learning_rate": 0.0004, "trim_and_crop": true, "caption_dropout": 0.1 }
Install Replicate’s Node.js client library:npm install replicate
Import and set up the client:import Replicate from "replicate"; const replicate = new Replicate({ auth: process.env.REPLICATE_API_TOKEN, });
Run genmoai/mochi-1-lora-trainer using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "genmoai/mochi-1-lora-trainer:170ea99fb48a30fef98cb1c9fb403a2882ab9d60c2ba15ad9383ace33c3fa385", { input: { seed: 42, steps: 1000, hf_token: "[REDACTED]", optimizer: "adamw", batch_size: 1, hf_repo_id: "lucataco/mochi-lora-disney", compile_dit: true, input_videos: "https://replicate.delivery/pbxt/M7sXFavLoOaCdC6eorODzcq0ap6vQC5P10Ai12OzyVYiDUja/disney-30.zip", learning_rate: 0.0004, trim_and_crop: true, caption_dropout: 0.1 } } ); console.log(output);
To learn more, take a look at the guide on getting started with Node.js.
Install Replicate’s Python client library:pip install replicate
Import the client:import replicate
Run genmoai/mochi-1-lora-trainer using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "genmoai/mochi-1-lora-trainer:170ea99fb48a30fef98cb1c9fb403a2882ab9d60c2ba15ad9383ace33c3fa385", input={ "seed": 42, "steps": 1000, "hf_token": "[REDACTED]", "optimizer": "adamw", "batch_size": 1, "hf_repo_id": "lucataco/mochi-lora-disney", "compile_dit": True, "input_videos": "https://replicate.delivery/pbxt/M7sXFavLoOaCdC6eorODzcq0ap6vQC5P10Ai12OzyVYiDUja/disney-30.zip", "learning_rate": 0.0004, "trim_and_crop": True, "caption_dropout": 0.1 } ) print(output)
To learn more, take a look at the guide on getting started with Python.
Run genmoai/mochi-1-lora-trainer using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
curl -s -X POST \ -H "Authorization: Bearer $REPLICATE_API_TOKEN" \ -H "Content-Type: application/json" \ -H "Prefer: wait" \ -d $'{ "version": "170ea99fb48a30fef98cb1c9fb403a2882ab9d60c2ba15ad9383ace33c3fa385", "input": { "seed": 42, "steps": 1000, "hf_token": "[REDACTED]", "optimizer": "adamw", "batch_size": 1, "hf_repo_id": "lucataco/mochi-lora-disney", "compile_dit": true, "input_videos": "https://replicate.delivery/pbxt/M7sXFavLoOaCdC6eorODzcq0ap6vQC5P10Ai12OzyVYiDUja/disney-30.zip", "learning_rate": 0.0004, "trim_and_crop": true, "caption_dropout": 0.1 } }' \ https://api.replicate.com/v1/predictions
To learn more, take a look at Replicate’s HTTP API reference docs.
Output
{ "completed_at": "2024-12-11T15:46:26.026534Z", "created_at": "2024-12-11T15:05:34Z", "data_removed": false, "error": null, "id": "w6kpq5g261rme0ckps0rad27hc", "input": { "seed": 42, "steps": 1000, "hf_token": "[REDACTED]", "optimizer": "adamw", "batch_size": 1, "hf_repo_id": "lucataco/mochi-lora-disney", "compile_dit": true, "input_videos": "https://replicate.delivery/pbxt/M7sXFavLoOaCdC6eorODzcq0ap6vQC5P10Ai12OzyVYiDUja/disney-30.zip", "learning_rate": 0.0004, "trim_and_crop": true, "caption_dropout": 0.1 }, "logs": "Cleaning up previous runs\nExtracted 60 files from zip to videos_input\n---Starting to Trim input videos---\nProcessing: videos_input/0bb5f6dbf8ed2e0060f0ac4164b24847.mp4\nvideos_input/0bb5f6dbf8ed2e0060f0ac4164b24847.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/0bb5f6dbf8ed2e0060f0ac4164b24847.txt to videos_prepared/0bb5f6dbf8ed2e0060f0ac4164b24847.txt\nMoviepy - Building video videos_prepared/0bb5f6dbf8ed2e0060f0ac4164b24847.mp4.\n 0%| | 0/30 [00:00<?, ?it/s]\n0%| | 0/30 [00:00<?, ?it/s]\nMoviepy - Writing video videos_prepared/0bb5f6dbf8ed2e0060f0ac4164b24847.mp4\n 0%| | 0/30 [00:00<?, ?it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\n \u001b[A\n0%| | 0/30 [00:00<?, ?it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/0bb5f6dbf8ed2e0060f0ac4164b24847.mp4\n 0%| | 0/30 [00:00<?, ?it/s]\nProcessing: videos_input/1d50a3d9703f152758d5422c8b48010f.mp4\nvideos_input/1d50a3d9703f152758d5422c8b48010f.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/1d50a3d9703f152758d5422c8b48010f.txt to videos_prepared/1d50a3d9703f152758d5422c8b48010f.txt\nMoviepy - Building video videos_prepared/1d50a3d9703f152758d5422c8b48010f.mp4.\nMoviepy - Writing video videos_prepared/1d50a3d9703f152758d5422c8b48010f.mp4\n 3%|▎ | 1/30 [00:00<00:07, 3.78it/s]\n3%|▎ | 1/30 [00:00<00:07, 3.78it/s]\n 3%|▎ | 1/30 [00:00<00:07, 3.78it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 98%|█████████▊| 39/40 [00:00<00:00, 385.32it/s, now=None]\u001b[A\n \u001b[A\n3%|▎ | 1/30 [00:00<00:07, 3.78it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/1d50a3d9703f152758d5422c8b48010f.mp4\n 3%|▎ | 1/30 [00:00<00:07, 3.78it/s]\nProcessing: videos_input/2c1ed5408882479b06681f7cf372916a.mp4\nvideos_input/2c1ed5408882479b06681f7cf372916a.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/2c1ed5408882479b06681f7cf372916a.txt to videos_prepared/2c1ed5408882479b06681f7cf372916a.txt\n 7%|▋ | 2/30 [00:00<00:07, 3.53it/s]\n7%|▋ | 2/30 [00:00<00:07, 3.53it/s]\nMoviepy - Building video videos_prepared/2c1ed5408882479b06681f7cf372916a.mp4.\nMoviepy - Writing video videos_prepared/2c1ed5408882479b06681f7cf372916a.mp4\n 7%|▋ | 2/30 [00:00<00:07, 3.53it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 100%|██████████| 40/40 [00:00<00:00, 391.86it/s, now=None]\u001b[A\n \u001b[A\n7%|▋ | 2/30 [00:00<00:07, 3.53it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/2c1ed5408882479b06681f7cf372916a.mp4\n 7%|▋ | 2/30 [00:00<00:07, 3.53it/s]\nProcessing: videos_input/3f0979e6cae25447f416372c49ad5e07.mp4\nvideos_input/3f0979e6cae25447f416372c49ad5e07.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/3f0979e6cae25447f416372c49ad5e07.txt to videos_prepared/3f0979e6cae25447f416372c49ad5e07.txt\nMoviepy - Building video videos_prepared/3f0979e6cae25447f416372c49ad5e07.mp4.\nMoviepy - Writing video videos_prepared/3f0979e6cae25447f416372c49ad5e07.mp4\n 10%|█ | 3/30 [00:00<00:07, 3.53it/s]\n10%|█ | 3/30 [00:00<00:07, 3.53it/s]\n 10%|█ | 3/30 [00:00<00:07, 3.53it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 98%|█████████▊| 39/40 [00:00<00:00, 384.65it/s, now=None]\u001b[A\n \u001b[A\n10%|█ | 3/30 [00:01<00:07, 3.53it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/3f0979e6cae25447f416372c49ad5e07.mp4\n 10%|█ | 3/30 [00:01<00:07, 3.53it/s]\nProcessing: videos_input/4adbb3a2945c9edd78785daccfd23e80.mp4\nvideos_input/4adbb3a2945c9edd78785daccfd23e80.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/4adbb3a2945c9edd78785daccfd23e80.txt to videos_prepared/4adbb3a2945c9edd78785daccfd23e80.txt\nMoviepy - Building video videos_prepared/4adbb3a2945c9edd78785daccfd23e80.mp4.\nMoviepy - Writing video videos_prepared/4adbb3a2945c9edd78785daccfd23e80.mp4\n 13%|█▎ | 4/30 [00:01<00:07, 3.54it/s]\n13%|█▎ | 4/30 [00:01<00:07, 3.54it/s]\n 13%|█▎ | 4/30 [00:01<00:07, 3.54it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\n \u001b[A\n13%|█▎ | 4/30 [00:01<00:07, 3.54it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/4adbb3a2945c9edd78785daccfd23e80.mp4\n 13%|█▎ | 4/30 [00:01<00:07, 3.54it/s]\nProcessing: videos_input/4c918b917308ff03120e9e86650a2d3c.mp4\nvideos_input/4c918b917308ff03120e9e86650a2d3c.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/4c918b917308ff03120e9e86650a2d3c.txt to videos_prepared/4c918b917308ff03120e9e86650a2d3c.txt\nMoviepy - Building video videos_prepared/4c918b917308ff03120e9e86650a2d3c.mp4.\nMoviepy - Writing video videos_prepared/4c918b917308ff03120e9e86650a2d3c.mp4\n 17%|█▋ | 5/30 [00:01<00:06, 3.68it/s]\n17%|█▋ | 5/30 [00:01<00:06, 3.68it/s]\n 17%|█▋ | 5/30 [00:01<00:06, 3.68it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 98%|█████████▊| 39/40 [00:00<00:00, 388.06it/s, now=None]\u001b[A\n \u001b[A\n17%|█▋ | 5/30 [00:01<00:06, 3.68it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/4c918b917308ff03120e9e86650a2d3c.mp4\n 17%|█▋ | 5/30 [00:01<00:06, 3.68it/s]\nProcessing: videos_input/5a0229ffdb3bd9d8e81dca7988d7cdbb.mp4\nvideos_input/5a0229ffdb3bd9d8e81dca7988d7cdbb.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/5a0229ffdb3bd9d8e81dca7988d7cdbb.txt to videos_prepared/5a0229ffdb3bd9d8e81dca7988d7cdbb.txt\nMoviepy - Building video videos_prepared/5a0229ffdb3bd9d8e81dca7988d7cdbb.mp4.\nMoviepy - Writing video videos_prepared/5a0229ffdb3bd9d8e81dca7988d7cdbb.mp4\n 20%|██ | 6/30 [00:01<00:06, 3.60it/s]\n20%|██ | 6/30 [00:01<00:06, 3.60it/s]\n 20%|██ | 6/30 [00:01<00:06, 3.60it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 92%|█████████▎| 37/40 [00:00<00:00, 366.42it/s, now=None]\u001b[A\n \u001b[A\n20%|██ | 6/30 [00:01<00:06, 3.60it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/5a0229ffdb3bd9d8e81dca7988d7cdbb.mp4\n 20%|██ | 6/30 [00:01<00:06, 3.60it/s]\nProcessing: videos_input/05a234b0164d015d468f2f53e771b4cf.mp4\nvideos_input/05a234b0164d015d468f2f53e771b4cf.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/05a234b0164d015d468f2f53e771b4cf.txt to videos_prepared/05a234b0164d015d468f2f53e771b4cf.txt\nMoviepy - Building video videos_prepared/05a234b0164d015d468f2f53e771b4cf.mp4.\nMoviepy - Writing video videos_prepared/05a234b0164d015d468f2f53e771b4cf.mp4\n 23%|██▎ | 7/30 [00:01<00:06, 3.61it/s]\n23%|██▎ | 7/30 [00:01<00:06, 3.61it/s]\n 23%|██▎ | 7/30 [00:01<00:06, 3.61it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\n \u001b[A\n23%|██▎ | 7/30 [00:02<00:06, 3.61it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/05a234b0164d015d468f2f53e771b4cf.mp4\n 23%|██▎ | 7/30 [00:02<00:06, 3.61it/s]\nProcessing: videos_input/05ccfa61ece031e881d173289761cf91.mp4\nvideos_input/05ccfa61ece031e881d173289761cf91.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/05ccfa61ece031e881d173289761cf91.txt to videos_prepared/05ccfa61ece031e881d173289761cf91.txt\n 27%|██▋ | 8/30 [00:02<00:06, 3.63it/s]\nMoviepy - Building video videos_prepared/05ccfa61ece031e881d173289761cf91.mp4.\n27%|██▋ | 8/30 [00:02<00:06, 3.63it/s]\nMoviepy - Writing video videos_prepared/05ccfa61ece031e881d173289761cf91.mp4\n 27%|██▋ | 8/30 [00:02<00:06, 3.63it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 98%|█████████▊| 39/40 [00:00<00:00, 382.15it/s, now=None]\u001b[A\n \u001b[A\n27%|██▋ | 8/30 [00:02<00:06, 3.63it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/05ccfa61ece031e881d173289761cf91.mp4\n 27%|██▋ | 8/30 [00:02<00:06, 3.63it/s]\nProcessing: videos_input/7d6dcf13f5c3d45b85c5ea0544c429e4.mp4\nvideos_input/7d6dcf13f5c3d45b85c5ea0544c429e4.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/7d6dcf13f5c3d45b85c5ea0544c429e4.txt to videos_prepared/7d6dcf13f5c3d45b85c5ea0544c429e4.txt\nMoviepy - Building video videos_prepared/7d6dcf13f5c3d45b85c5ea0544c429e4.mp4.\nMoviepy - Writing video videos_prepared/7d6dcf13f5c3d45b85c5ea0544c429e4.mp4\n 30%|███ | 9/30 [00:02<00:05, 3.59it/s]\n30%|███ | 9/30 [00:02<00:05, 3.59it/s]\n 30%|███ | 9/30 [00:02<00:05, 3.59it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 98%|█████████▊| 39/40 [00:00<00:00, 387.57it/s, now=None]\u001b[A\n \u001b[A\n30%|███ | 9/30 [00:02<00:05, 3.59it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/7d6dcf13f5c3d45b85c5ea0544c429e4.mp4\n 30%|███ | 9/30 [00:02<00:05, 3.59it/s]\nProcessing: videos_input/7fe0c83572de828da1cab0c118dece14.mp4\nvideos_input/7fe0c83572de828da1cab0c118dece14.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/7fe0c83572de828da1cab0c118dece14.txt to videos_prepared/7fe0c83572de828da1cab0c118dece14.txt\nMoviepy - Building video videos_prepared/7fe0c83572de828da1cab0c118dece14.mp4.\nMoviepy - Writing video videos_prepared/7fe0c83572de828da1cab0c118dece14.mp4\n 33%|███▎ | 10/30 [00:02<00:05, 3.49it/s]\n33%|███▎ | 10/30 [00:02<00:05, 3.49it/s]\n 33%|███▎ | 10/30 [00:02<00:05, 3.49it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 98%|█████████▊| 39/40 [00:00<00:00, 385.04it/s, now=None]\u001b[A\n \u001b[A\n33%|███▎ | 10/30 [00:03<00:05, 3.49it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/7fe0c83572de828da1cab0c118dece14.mp4\n 33%|███▎ | 10/30 [00:03<00:05, 3.49it/s]\nProcessing: videos_input/8adfde998361b1d7c6f38a35481667fd.mp4\nvideos_input/8adfde998361b1d7c6f38a35481667fd.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/8adfde998361b1d7c6f38a35481667fd.txt to videos_prepared/8adfde998361b1d7c6f38a35481667fd.txt\nMoviepy - Building video videos_prepared/8adfde998361b1d7c6f38a35481667fd.mp4.\nMoviepy - Writing video videos_prepared/8adfde998361b1d7c6f38a35481667fd.mp4\n 37%|███▋ | 11/30 [00:03<00:05, 3.54it/s]\n37%|███▋ | 11/30 [00:03<00:05, 3.54it/s]\n 37%|███▋ | 11/30 [00:03<00:05, 3.54it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 98%|█████████▊| 39/40 [00:00<00:00, 387.19it/s, now=None]\u001b[A\n \u001b[A\n37%|███▋ | 11/30 [00:03<00:05, 3.54it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/8adfde998361b1d7c6f38a35481667fd.mp4\n 37%|███▋ | 11/30 [00:03<00:05, 3.54it/s]\nProcessing: videos_input/8ae679ab483ab344c881d4a813e0cb51.mp4\nvideos_input/8ae679ab483ab344c881d4a813e0cb51.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/8ae679ab483ab344c881d4a813e0cb51.txt to videos_prepared/8ae679ab483ab344c881d4a813e0cb51.txt\nMoviepy - Building video videos_prepared/8ae679ab483ab344c881d4a813e0cb51.mp4.\nMoviepy - Writing video videos_prepared/8ae679ab483ab344c881d4a813e0cb51.mp4\n 40%|████ | 12/30 [00:03<00:05, 3.50it/s]\n40%|████ | 12/30 [00:03<00:05, 3.50it/s]\n 40%|████ | 12/30 [00:03<00:05, 3.50it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\n \u001b[A\n40%|████ | 12/30 [00:03<00:05, 3.50it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/8ae679ab483ab344c881d4a813e0cb51.mp4\n 40%|████ | 12/30 [00:03<00:05, 3.50it/s]\nProcessing: videos_input/8d616fee8e0a280d2d87e478b948a729.mp4\nvideos_input/8d616fee8e0a280d2d87e478b948a729.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/8d616fee8e0a280d2d87e478b948a729.txt to videos_prepared/8d616fee8e0a280d2d87e478b948a729.txt\nMoviepy - Building video videos_prepared/8d616fee8e0a280d2d87e478b948a729.mp4.\nMoviepy - Writing video videos_prepared/8d616fee8e0a280d2d87e478b948a729.mp4\n 43%|████▎ | 13/30 [00:03<00:04, 3.49it/s]\n43%|████▎ | 13/30 [00:03<00:04, 3.49it/s]\n 43%|████▎ | 13/30 [00:03<00:04, 3.49it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 98%|█████████▊| 39/40 [00:00<00:00, 389.56it/s, now=None]\u001b[A\n \u001b[A\n43%|████▎ | 13/30 [00:03<00:04, 3.49it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/8d616fee8e0a280d2d87e478b948a729.mp4\n 43%|████▎ | 13/30 [00:03<00:04, 3.49it/s]\nProcessing: videos_input/8e7722634784cf969c15f4a597f3af4d.mp4\nvideos_input/8e7722634784cf969c15f4a597f3af4d.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/8e7722634784cf969c15f4a597f3af4d.txt to videos_prepared/8e7722634784cf969c15f4a597f3af4d.txt\nMoviepy - Building video videos_prepared/8e7722634784cf969c15f4a597f3af4d.mp4.\n 47%|████▋ | 14/30 [00:03<00:04, 3.46it/s]\n47%|████▋ | 14/30 [00:03<00:04, 3.46it/s]\nMoviepy - Writing video videos_prepared/8e7722634784cf969c15f4a597f3af4d.mp4\n 47%|████▋ | 14/30 [00:03<00:04, 3.46it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 100%|██████████| 40/40 [00:00<00:00, 395.87it/s, now=None]\u001b[A\n \u001b[A\n47%|████▋ | 14/30 [00:04<00:04, 3.46it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/8e7722634784cf969c15f4a597f3af4d.mp4\n 47%|████▋ | 14/30 [00:04<00:04, 3.46it/s]\nProcessing: videos_input/12e51adf1acbf7acbb703a96a464a39b.mp4\nvideos_input/12e51adf1acbf7acbb703a96a464a39b.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/12e51adf1acbf7acbb703a96a464a39b.txt to videos_prepared/12e51adf1acbf7acbb703a96a464a39b.txt\nMoviepy - Building video videos_prepared/12e51adf1acbf7acbb703a96a464a39b.mp4.\n 50%|█████ | 15/30 [00:04<00:04, 3.46it/s]\n50%|█████ | 15/30 [00:04<00:04, 3.46it/s]\nMoviepy - Writing video videos_prepared/12e51adf1acbf7acbb703a96a464a39b.mp4\n 50%|█████ | 15/30 [00:04<00:04, 3.46it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\n \u001b[A\n50%|█████ | 15/30 [00:04<00:04, 3.46it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/12e51adf1acbf7acbb703a96a464a39b.mp4\n 50%|█████ | 15/30 [00:04<00:04, 3.46it/s]\nProcessing: videos_input/46e9d133d051655c956c7089b672f519.mp4\nvideos_input/46e9d133d051655c956c7089b672f519.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/46e9d133d051655c956c7089b672f519.txt to videos_prepared/46e9d133d051655c956c7089b672f519.txt\nMoviepy - Building video videos_prepared/46e9d133d051655c956c7089b672f519.mp4.\nMoviepy - Writing video videos_prepared/46e9d133d051655c956c7089b672f519.mp4\n 53%|█████▎ | 16/30 [00:04<00:04, 3.50it/s]\n53%|█████▎ | 16/30 [00:04<00:04, 3.50it/s]\n 53%|█████▎ | 16/30 [00:04<00:04, 3.50it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 92%|█████████▎| 37/40 [00:00<00:00, 363.67it/s, now=None]\u001b[A\n \u001b[A\nMoviepy - Done !\n 53%|█████▎ | 16/30 [00:04<00:04, 3.50it/s]\nMoviepy - video ready videos_prepared/46e9d133d051655c956c7089b672f519.mp4\n 53%|█████▎ | 16/30 [00:04<00:04, 3.50it/s]\nProcessing: videos_input/46f4eee0864dd89c9225367d826a657f.mp4\nvideos_input/46f4eee0864dd89c9225367d826a657f.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/46f4eee0864dd89c9225367d826a657f.txt to videos_prepared/46f4eee0864dd89c9225367d826a657f.txt\nMoviepy - Building video videos_prepared/46f4eee0864dd89c9225367d826a657f.mp4.\nMoviepy - Writing video videos_prepared/46f4eee0864dd89c9225367d826a657f.mp4\n 57%|█████▋ | 17/30 [00:04<00:03, 3.47it/s]\n57%|█████▋ | 17/30 [00:04<00:03, 3.47it/s]\n 57%|█████▋ | 17/30 [00:04<00:03, 3.47it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\n \u001b[A\n57%|█████▋ | 17/30 [00:05<00:03, 3.47it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/46f4eee0864dd89c9225367d826a657f.mp4\n 57%|█████▋ | 17/30 [00:05<00:03, 3.47it/s]\nProcessing: videos_input/58b88d44575e945cd7dcd11b3aac6ff0.mp4\nvideos_input/58b88d44575e945cd7dcd11b3aac6ff0.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/58b88d44575e945cd7dcd11b3aac6ff0.txt to videos_prepared/58b88d44575e945cd7dcd11b3aac6ff0.txt\nMoviepy - Building video videos_prepared/58b88d44575e945cd7dcd11b3aac6ff0.mp4.\n 60%|██████ | 18/30 [00:05<00:03, 3.42it/s]\n60%|██████ | 18/30 [00:05<00:03, 3.42it/s]\nMoviepy - Writing video videos_prepared/58b88d44575e945cd7dcd11b3aac6ff0.mp4\n 60%|██████ | 18/30 [00:05<00:03, 3.42it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 100%|██████████| 40/40 [00:00<00:00, 399.20it/s, now=None]\u001b[A\n \u001b[A\n60%|██████ | 18/30 [00:05<00:03, 3.42it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/58b88d44575e945cd7dcd11b3aac6ff0.mp4\n 60%|██████ | 18/30 [00:05<00:03, 3.42it/s]\nProcessing: videos_input/81c5dab878d73e6c21181d18d83f2808.mp4\nvideos_input/81c5dab878d73e6c21181d18d83f2808.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/81c5dab878d73e6c21181d18d83f2808.txt to videos_prepared/81c5dab878d73e6c21181d18d83f2808.txt\nMoviepy - Building video videos_prepared/81c5dab878d73e6c21181d18d83f2808.mp4.\nMoviepy - Writing video videos_prepared/81c5dab878d73e6c21181d18d83f2808.mp4\n 63%|██████▎ | 19/30 [00:05<00:03, 3.49it/s]\n63%|██████▎ | 19/30 [00:05<00:03, 3.49it/s]\n 63%|██████▎ | 19/30 [00:05<00:03, 3.49it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 100%|██████████| 40/40 [00:00<00:00, 398.49it/s, now=None]\u001b[A\n \u001b[A\n63%|██████▎ | 19/30 [00:05<00:03, 3.49it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/81c5dab878d73e6c21181d18d83f2808.mp4\n 63%|██████▎ | 19/30 [00:05<00:03, 3.49it/s]\nProcessing: videos_input/96d342ea7c7cfddbe1106072bc34be5a.mp4\nvideos_input/96d342ea7c7cfddbe1106072bc34be5a.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/96d342ea7c7cfddbe1106072bc34be5a.txt to videos_prepared/96d342ea7c7cfddbe1106072bc34be5a.txt\nMoviepy - Building video videos_prepared/96d342ea7c7cfddbe1106072bc34be5a.mp4.\nMoviepy - Writing video videos_prepared/96d342ea7c7cfddbe1106072bc34be5a.mp4\n 67%|██████▋ | 20/30 [00:05<00:02, 3.53it/s]\n67%|██████▋ | 20/30 [00:05<00:02, 3.53it/s]\n 67%|██████▋ | 20/30 [00:05<00:02, 3.53it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 98%|█████████▊| 39/40 [00:00<00:00, 389.71it/s, now=None]\u001b[A\n \u001b[A\n67%|██████▋ | 20/30 [00:05<00:02, 3.53it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/96d342ea7c7cfddbe1106072bc34be5a.mp4\n 67%|██████▋ | 20/30 [00:05<00:02, 3.53it/s]\nProcessing: videos_input/0288f3d69c08e816d81b014da620db49.mp4\nvideos_input/0288f3d69c08e816d81b014da620db49.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/0288f3d69c08e816d81b014da620db49.txt to videos_prepared/0288f3d69c08e816d81b014da620db49.txt\nMoviepy - Building video videos_prepared/0288f3d69c08e816d81b014da620db49.mp4.\nMoviepy - Writing video videos_prepared/0288f3d69c08e816d81b014da620db49.mp4\n 70%|███████ | 21/30 [00:05<00:02, 3.53it/s]\n70%|███████ | 21/30 [00:05<00:02, 3.53it/s]\n 70%|███████ | 21/30 [00:05<00:02, 3.53it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 92%|█████████▎| 37/40 [00:00<00:00, 365.51it/s, now=None]\u001b[A\n \u001b[A\n70%|███████ | 21/30 [00:06<00:02, 3.53it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/0288f3d69c08e816d81b014da620db49.mp4\n 70%|███████ | 21/30 [00:06<00:02, 3.53it/s]\nProcessing: videos_input/328fc12cf9cf3d540e67efadeb893f61.mp4\nvideos_input/328fc12cf9cf3d540e67efadeb893f61.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/328fc12cf9cf3d540e67efadeb893f61.txt to videos_prepared/328fc12cf9cf3d540e67efadeb893f61.txt\nMoviepy - Building video videos_prepared/328fc12cf9cf3d540e67efadeb893f61.mp4.\nMoviepy - Writing video videos_prepared/328fc12cf9cf3d540e67efadeb893f61.mp4\n 73%|███████▎ | 22/30 [00:06<00:02, 3.48it/s]\n73%|███████▎ | 22/30 [00:06<00:02, 3.48it/s]\n 73%|███████▎ | 22/30 [00:06<00:02, 3.48it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\n \u001b[A\n73%|███████▎ | 22/30 [00:06<00:02, 3.48it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/328fc12cf9cf3d540e67efadeb893f61.mp4\n 73%|███████▎ | 22/30 [00:06<00:02, 3.48it/s]\nProcessing: videos_input/383cb4b496d17695554655f3ec79c587.mp4\nvideos_input/383cb4b496d17695554655f3ec79c587.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/383cb4b496d17695554655f3ec79c587.txt to videos_prepared/383cb4b496d17695554655f3ec79c587.txt\nMoviepy - Building video videos_prepared/383cb4b496d17695554655f3ec79c587.mp4.\nMoviepy - Writing video videos_prepared/383cb4b496d17695554655f3ec79c587.mp4\n 77%|███████▋ | 23/30 [00:06<00:02, 3.46it/s]\n77%|███████▋ | 23/30 [00:06<00:02, 3.46it/s]\n 77%|███████▋ | 23/30 [00:06<00:02, 3.46it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 98%|█████████▊| 39/40 [00:00<00:00, 388.65it/s, now=None]\u001b[A\n \u001b[A\n77%|███████▋ | 23/30 [00:06<00:02, 3.46it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/383cb4b496d17695554655f3ec79c587.mp4\n 77%|███████▋ | 23/30 [00:06<00:02, 3.46it/s]\nProcessing: videos_input/485b43aa4524327f3c7a40d28e1cf7bc.mp4\nvideos_input/485b43aa4524327f3c7a40d28e1cf7bc.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/485b43aa4524327f3c7a40d28e1cf7bc.txt to videos_prepared/485b43aa4524327f3c7a40d28e1cf7bc.txt\nMoviepy - Building video videos_prepared/485b43aa4524327f3c7a40d28e1cf7bc.mp4.\n 80%|████████ | 24/30 [00:06<00:01, 3.46it/s]\n80%|████████ | 24/30 [00:06<00:01, 3.46it/s]\nMoviepy - Writing video videos_prepared/485b43aa4524327f3c7a40d28e1cf7bc.mp4\n 80%|████████ | 24/30 [00:06<00:01, 3.46it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\n \u001b[A\n80%|████████ | 24/30 [00:07<00:01, 3.46it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/485b43aa4524327f3c7a40d28e1cf7bc.mp4\n 80%|████████ | 24/30 [00:07<00:01, 3.46it/s]\nProcessing: videos_input/560c6472660330638c2809d823d59be3.mp4\nvideos_input/560c6472660330638c2809d823d59be3.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/560c6472660330638c2809d823d59be3.txt to videos_prepared/560c6472660330638c2809d823d59be3.txt\nMoviepy - Building video videos_prepared/560c6472660330638c2809d823d59be3.mp4.\nMoviepy - Writing video videos_prepared/560c6472660330638c2809d823d59be3.mp4\n 83%|████████▎ | 25/30 [00:07<00:01, 3.47it/s]\n83%|████████▎ | 25/30 [00:07<00:01, 3.47it/s]\n 83%|████████▎ | 25/30 [00:07<00:01, 3.47it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 95%|█████████▌| 38/40 [00:00<00:00, 378.13it/s, now=None]\u001b[A\n \u001b[A\n83%|████████▎ | 25/30 [00:07<00:01, 3.47it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/560c6472660330638c2809d823d59be3.mp4\n 83%|████████▎ | 25/30 [00:07<00:01, 3.47it/s]\nProcessing: videos_input/614cf13ae1974436cf4072a5cc7d7c57.mp4\nvideos_input/614cf13ae1974436cf4072a5cc7d7c57.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/614cf13ae1974436cf4072a5cc7d7c57.txt to videos_prepared/614cf13ae1974436cf4072a5cc7d7c57.txt\nMoviepy - Building video videos_prepared/614cf13ae1974436cf4072a5cc7d7c57.mp4.\nMoviepy - Writing video videos_prepared/614cf13ae1974436cf4072a5cc7d7c57.mp4\n 87%|████████▋ | 26/30 [00:07<00:01, 3.47it/s]\n87%|████████▋ | 26/30 [00:07<00:01, 3.47it/s]\n 87%|████████▋ | 26/30 [00:07<00:01, 3.47it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 100%|██████████| 40/40 [00:00<00:00, 396.80it/s, now=None]\u001b[A\n \u001b[A\n87%|████████▋ | 26/30 [00:07<00:01, 3.47it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/614cf13ae1974436cf4072a5cc7d7c57.mp4\n 87%|████████▋ | 26/30 [00:07<00:01, 3.47it/s]\nProcessing: videos_input/1151c01bd77450dfc603a2eb7352822e.mp4\nvideos_input/1151c01bd77450dfc603a2eb7352822e.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/1151c01bd77450dfc603a2eb7352822e.txt to videos_prepared/1151c01bd77450dfc603a2eb7352822e.txt\n 90%|█████████ | 27/30 [00:07<00:00, 3.50it/s]\n90%|█████████ | 27/30 [00:07<00:00, 3.50it/s]\nMoviepy - Building video videos_prepared/1151c01bd77450dfc603a2eb7352822e.mp4.\nMoviepy - Writing video videos_prepared/1151c01bd77450dfc603a2eb7352822e.mp4\n 90%|█████████ | 27/30 [00:07<00:00, 3.50it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\n \u001b[A\n90%|█████████ | 27/30 [00:07<00:00, 3.50it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/1151c01bd77450dfc603a2eb7352822e.mp4\n 90%|█████████ | 27/30 [00:07<00:00, 3.50it/s]\nProcessing: videos_input/2325e5f8e287753e50e47ab2fc2e8241.mp4\nvideos_input/2325e5f8e287753e50e47ab2fc2e8241.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/2325e5f8e287753e50e47ab2fc2e8241.txt to videos_prepared/2325e5f8e287753e50e47ab2fc2e8241.txt\nMoviepy - Building video videos_prepared/2325e5f8e287753e50e47ab2fc2e8241.mp4.\n 93%|█████████▎| 28/30 [00:07<00:00, 3.48it/s]\n93%|█████████▎| 28/30 [00:07<00:00, 3.48it/s]\nMoviepy - Writing video videos_prepared/2325e5f8e287753e50e47ab2fc2e8241.mp4\n 93%|█████████▎| 28/30 [00:07<00:00, 3.48it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 98%|█████████▊| 39/40 [00:00<00:00, 388.53it/s, now=None]\u001b[A\n \u001b[A\n93%|█████████▎| 28/30 [00:08<00:00, 3.48it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/2325e5f8e287753e50e47ab2fc2e8241.mp4\n 93%|█████████▎| 28/30 [00:08<00:00, 3.48it/s]\nProcessing: videos_input/3108dd567bd8669967bc83e0bc50dab2.mp4\nvideos_input/3108dd567bd8669967bc83e0bc50dab2.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video.\nCopied videos_input/3108dd567bd8669967bc83e0bc50dab2.txt to videos_prepared/3108dd567bd8669967bc83e0bc50dab2.txt\nMoviepy - Building video videos_prepared/3108dd567bd8669967bc83e0bc50dab2.mp4.\nMoviepy - Writing video videos_prepared/3108dd567bd8669967bc83e0bc50dab2.mp4\n 97%|█████████▋| 29/30 [00:08<00:00, 3.53it/s]\n97%|█████████▋| 29/30 [00:08<00:00, 3.53it/s]\n 97%|█████████▋| 29/30 [00:08<00:00, 3.53it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\n \u001b[A\n97%|█████████▋| 29/30 [00:08<00:00, 3.53it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/3108dd567bd8669967bc83e0bc50dab2.mp4\n 97%|█████████▋| 29/30 [00:08<00:00, 3.53it/s]\n100%|██████████| 30/30 [00:08<00:00, 3.65it/s]\n100%|██████████| 30/30 [00:08<00:00, 3.53it/s]\n---Starting to Embed videos---\nLoading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s]\nLoading checkpoint shards: 50%|█████ | 1/2 [00:00<00:00, 1.78it/s]\nLoading checkpoint shards: 100%|██████████| 2/2 [00:01<00:00, 1.90it/s]\nLoading checkpoint shards: 100%|██████████| 2/2 [00:01<00:00, 1.88it/s]\nLoading pipeline components...: 0%| | 0/3 [00:00<?, ?it/s]\nLoading pipeline components...: 100%|██████████| 3/3 [00:00<00:00, 651.69it/s]\nProcessing videos_prepared/0288f3d69c08e816d81b014da620db49.mp4\nTrimmed video from 40 to first 37 frames\n0it [00:00, ?it/s]\nProcessing videos_prepared/05a234b0164d015d468f2f53e771b4cf.mp4\nTrimmed video from 40 to first 37 frames\n1it [00:01, 1.40s/it]\nProcessing videos_prepared/05ccfa61ece031e881d173289761cf91.mp4\nTrimmed video from 40 to first 37 frames\n2it [00:02, 1.14s/it]\nProcessing videos_prepared/0bb5f6dbf8ed2e0060f0ac4164b24847.mp4\nTrimmed video from 40 to first 37 frames\n3it [00:03, 1.05s/it]\nProcessing videos_prepared/1151c01bd77450dfc603a2eb7352822e.mp4\nTrimmed video from 40 to first 37 frames\n4it [00:04, 1.01s/it]\nProcessing videos_prepared/12e51adf1acbf7acbb703a96a464a39b.mp4\nTrimmed video from 40 to first 37 frames\n5it [00:05, 1.01it/s]\nProcessing videos_prepared/1d50a3d9703f152758d5422c8b48010f.mp4\nTrimmed video from 40 to first 37 frames\n6it [00:06, 1.02it/s]\nProcessing videos_prepared/2325e5f8e287753e50e47ab2fc2e8241.mp4\nTrimmed video from 40 to first 37 frames\n7it [00:07, 1.03it/s]\nProcessing videos_prepared/2c1ed5408882479b06681f7cf372916a.mp4\nTrimmed video from 40 to first 37 frames\n8it [00:08, 1.04it/s]\nProcessing videos_prepared/3108dd567bd8669967bc83e0bc50dab2.mp4\nTrimmed video from 40 to first 37 frames\n9it [00:09, 1.01it/s]\nProcessing videos_prepared/328fc12cf9cf3d540e67efadeb893f61.mp4\nTrimmed video from 40 to first 37 frames\n10it [00:10, 1.02it/s]\nProcessing videos_prepared/383cb4b496d17695554655f3ec79c587.mp4\nTrimmed video from 40 to first 37 frames\n11it [00:11, 1.00s/it]\nProcessing videos_prepared/3f0979e6cae25447f416372c49ad5e07.mp4\nTrimmed video from 40 to first 37 frames\n12it [00:12, 1.02it/s]\nProcessing videos_prepared/46e9d133d051655c956c7089b672f519.mp4\nTrimmed video from 40 to first 37 frames\n13it [00:12, 1.03it/s]\nProcessing videos_prepared/46f4eee0864dd89c9225367d826a657f.mp4\nTrimmed video from 40 to first 37 frames\n14it [00:13, 1.04it/s]\nProcessing videos_prepared/485b43aa4524327f3c7a40d28e1cf7bc.mp4\nTrimmed video from 40 to first 37 frames\n15it [00:14, 1.04it/s]\nProcessing videos_prepared/4adbb3a2945c9edd78785daccfd23e80.mp4\nTrimmed video from 40 to first 37 frames\n16it [00:15, 1.04it/s]\nProcessing videos_prepared/4c918b917308ff03120e9e86650a2d3c.mp4\nTrimmed video from 40 to first 37 frames\n17it [00:16, 1.05it/s]\nProcessing videos_prepared/560c6472660330638c2809d823d59be3.mp4\nTrimmed video from 40 to first 37 frames\n18it [00:17, 1.05it/s]\nProcessing videos_prepared/58b88d44575e945cd7dcd11b3aac6ff0.mp4\nTrimmed video from 40 to first 37 frames\n19it [00:18, 1.02it/s]\nProcessing videos_prepared/5a0229ffdb3bd9d8e81dca7988d7cdbb.mp4\nTrimmed video from 40 to first 37 frames\n20it [00:19, 1.03it/s]\nProcessing videos_prepared/614cf13ae1974436cf4072a5cc7d7c57.mp4\nTrimmed video from 40 to first 37 frames\n21it [00:20, 1.04it/s]\nProcessing videos_prepared/7d6dcf13f5c3d45b85c5ea0544c429e4.mp4\nTrimmed video from 40 to first 37 frames\n22it [00:21, 1.05it/s]\nProcessing videos_prepared/7fe0c83572de828da1cab0c118dece14.mp4\nTrimmed video from 40 to first 37 frames\n23it [00:22, 1.05it/s]\nProcessing videos_prepared/81c5dab878d73e6c21181d18d83f2808.mp4\nTrimmed video from 40 to first 37 frames\n24it [00:23, 1.05it/s]\nProcessing videos_prepared/8adfde998361b1d7c6f38a35481667fd.mp4\nTrimmed video from 40 to first 37 frames\n25it [00:24, 1.05it/s]\nProcessing videos_prepared/8ae679ab483ab344c881d4a813e0cb51.mp4\nTrimmed video from 40 to first 37 frames\n26it [00:25, 1.05it/s]\nProcessing videos_prepared/8d616fee8e0a280d2d87e478b948a729.mp4\nTrimmed video from 40 to first 37 frames\n27it [00:26, 1.06it/s]\nProcessing videos_prepared/8e7722634784cf969c15f4a597f3af4d.mp4\nTrimmed video from 40 to first 37 frames\n28it [00:27, 1.05it/s]\nProcessing videos_prepared/96d342ea7c7cfddbe1106072bc34be5a.mp4\nTrimmed video from 40 to first 37 frames\n29it [00:28, 1.02it/s]\n30it [00:29, 1.03it/s]\n30it [00:29, 1.02it/s]\n---Starting training---\nFound 30 training videos in videos_prepared\nLoaded 30/30 valid file pairs.\n===== Memory before training =====\nmemory_allocated=18.903 GB\nmax_memory_allocated=18.903 GB\nmax_memory_reserved=28.078 GB\n***** Running training *****\nNum trainable parameters = 19005440\nNum examples = 30\nNum batches each epoch = 30\nNum epochs = 34\nInstantaneous batch size per device = 1\nTotal train batch size (w. parallel, distributed & accumulation) = 1\nTotal optimization steps = 1000\nSteps: 0%| | 0/1000 [00:00<?, ?it/s]W1211 15:09:46.660000 135675630435840 torch/fx/experimental/symbolic_shapes.py:4449] [0/0] xindex is not in var_ranges, defaulting to unknown range.\nW1211 15:09:46.674000 135675630435840 torch/fx/experimental/symbolic_shapes.py:4449] [0/0] xindex is not in var_ranges, defaulting to unknown range.\nW1211 15:09:46.812000 135675630435840 torch/fx/experimental/symbolic_shapes.py:4449] [0/0] xindex is not in var_ranges, defaulting to unknown range.\nSteps: 0%| | 1/1000 [04:18<71:48:02, 258.74s/it]\nSteps: 0%| | 1/1000 [04:18<71:48:02, 258.74s/it, loss=1.07, lr=2e-6]\nSteps: 0%| | 2/1000 [04:20<29:50:05, 107.62s/it, loss=1.07, lr=2e-6]\nSteps: 0%| | 2/1000 [04:20<29:50:05, 107.62s/it, loss=0.666, lr=4e-6]\nSteps: 0%| | 3/1000 [04:22<16:25:48, 59.33s/it, loss=0.666, lr=4e-6] \nSteps: 0%| | 3/1000 [04:22<16:25:48, 59.33s/it, loss=0.335, lr=6e-6]\nSteps: 0%| | 4/1000 [04:24<10:08:13, 36.64s/it, loss=0.335, lr=6e-6]\nSteps: 0%| | 4/1000 [04:24<10:08:13, 36.64s/it, loss=0.362, lr=8e-6]\nSteps: 0%| | 5/1000 [04:26<6:39:34, 24.09s/it, loss=0.362, lr=8e-6] \nSteps: 0%| | 5/1000 [04:26<6:39:34, 24.09s/it, loss=0.905, lr=1e-5]\nSteps: 1%| | 6/1000 [04:28<4:33:55, 16.53s/it, loss=0.905, lr=1e-5]\nSteps: 1%| | 6/1000 [04:28<4:33:55, 16.53s/it, loss=0.767, lr=1.2e-5]\nSteps: 1%| | 7/1000 [04:29<3:14:14, 11.74s/it, loss=0.767, lr=1.2e-5]\nSteps: 1%| | 7/1000 [04:29<3:14:14, 11.74s/it, loss=0.973, lr=1.4e-5]\nSteps: 1%| | 8/1000 [04:31<2:22:04, 8.59s/it, loss=0.973, lr=1.4e-5]\nSteps: 1%| | 8/1000 [04:31<2:22:04, 8.59s/it, loss=0.821, lr=1.6e-5]\nSteps: 1%| | 9/1000 [04:33<1:47:09, 6.49s/it, loss=0.821, lr=1.6e-5]\nSteps: 1%| | 9/1000 [04:33<1:47:09, 6.49s/it, loss=0.472, lr=1.8e-5]\nSteps: 1%| | 10/1000 [04:35<1:23:29, 5.06s/it, loss=0.472, lr=1.8e-5]\nSteps: 1%| | 10/1000 [04:35<1:23:29, 5.06s/it, loss=0.358, lr=2e-5] \nSteps: 1%| | 11/1000 [04:37<1:07:16, 4.08s/it, loss=0.358, lr=2e-5]\nSteps: 1%| | 11/1000 [04:37<1:07:16, 4.08s/it, loss=0.332, lr=2.2e-5]\nSteps: 1%| | 12/1000 [04:39<56:04, 3.41s/it, loss=0.332, lr=2.2e-5] \nSteps: 1%| | 12/1000 [04:39<56:04, 3.41s/it, loss=0.353, lr=2.4e-5]\nSteps: 1%|▏ | 13/1000 [04:41<48:22, 2.94s/it, loss=0.353, lr=2.4e-5]\nSteps: 1%|▏ | 13/1000 [04:41<48:22, 2.94s/it, loss=0.346, lr=2.6e-5]\nSteps: 1%|▏ | 14/1000 [04:42<42:57, 2.61s/it, loss=0.346, lr=2.6e-5]\nSteps: 1%|▏ | 14/1000 [04:42<42:57, 2.61s/it, loss=0.499, lr=2.8e-5]\nSteps: 2%|▏ | 15/1000 [04:44<39:12, 2.39s/it, loss=0.499, lr=2.8e-5]\nSteps: 2%|▏ | 15/1000 [04:44<39:12, 2.39s/it, loss=1.07, lr=3e-5] \nSteps: 2%|▏ | 16/1000 [04:46<36:35, 2.23s/it, loss=1.07, lr=3e-5]\nSteps: 2%|▏ | 16/1000 [04:46<36:35, 2.23s/it, loss=0.448, lr=3.2e-5]\nSteps: 2%|▏ | 17/1000 [04:48<34:44, 2.12s/it, loss=0.448, lr=3.2e-5]\nSteps: 2%|▏ | 17/1000 [04:48<34:44, 2.12s/it, loss=0.752, lr=3.4e-5]\nSteps: 2%|▏ | 18/1000 [04:50<33:26, 2.04s/it, loss=0.752, lr=3.4e-5]\nSteps: 2%|▏ | 18/1000 [04:50<33:26, 2.04s/it, loss=0.33, lr=3.6e-5] \nSteps: 2%|▏ | 19/1000 [04:52<32:30, 1.99s/it, loss=0.33, lr=3.6e-5]\nSteps: 2%|▏ | 19/1000 [04:52<32:30, 1.99s/it, loss=0.873, lr=3.8e-5]\nSteps: 2%|▏ | 20/1000 [04:54<31:52, 1.95s/it, loss=0.873, lr=3.8e-5]\nSteps: 2%|▏ | 20/1000 [04:54<31:52, 1.95s/it, loss=0.499, lr=4e-5] \nSteps: 2%|▏ | 21/1000 [04:55<31:25, 1.93s/it, loss=0.499, lr=4e-5]\nSteps: 2%|▏ | 21/1000 [04:55<31:25, 1.93s/it, loss=0.55, lr=4.2e-5]\nSteps: 2%|▏ | 22/1000 [04:57<31:07, 1.91s/it, loss=0.55, lr=4.2e-5]\nSteps: 2%|▏ | 22/1000 [04:57<31:07, 1.91s/it, loss=0.304, lr=4.4e-5]\nSteps: 2%|▏ | 23/1000 [04:59<30:52, 1.90s/it, loss=0.304, lr=4.4e-5]\nSteps: 2%|▏ | 23/1000 [04:59<30:52, 1.90s/it, loss=0.42, lr=4.6e-5] \nSteps: 2%|▏ | 24/1000 [05:01<30:41, 1.89s/it, loss=0.42, lr=4.6e-5]\nSteps: 2%|▏ | 24/1000 [05:01<30:41, 1.89s/it, loss=0.442, lr=4.8e-5]\nSteps: 2%|▎ | 25/1000 [05:03<30:34, 1.88s/it, loss=0.442, lr=4.8e-5]\nSteps: 2%|▎ | 25/1000 [05:03<30:34, 1.88s/it, loss=0.386, lr=5e-5] \nSteps: 3%|▎ | 26/1000 [05:05<30:28, 1.88s/it, loss=0.386, lr=5e-5]\nSteps: 3%|▎ | 26/1000 [05:05<30:28, 1.88s/it, loss=0.453, lr=5.2e-5]\nSteps: 3%|▎ | 27/1000 [05:07<30:22, 1.87s/it, loss=0.453, lr=5.2e-5]\nSteps: 3%|▎ | 27/1000 [05:07<30:22, 1.87s/it, loss=0.524, lr=5.4e-5]\nSteps: 3%|▎ | 28/1000 [05:09<30:18, 1.87s/it, loss=0.524, lr=5.4e-5]\nSteps: 3%|▎ | 28/1000 [05:09<30:18, 1.87s/it, loss=0.853, lr=5.6e-5]\nSteps: 3%|▎ | 29/1000 [05:10<30:14, 1.87s/it, loss=0.853, lr=5.6e-5]\nSteps: 3%|▎ | 29/1000 [05:10<30:14, 1.87s/it, loss=0.383, lr=5.8e-5]\nSteps: 3%|▎ | 30/1000 [05:12<30:12, 1.87s/it, loss=0.383, lr=5.8e-5]\nSteps: 3%|▎ | 30/1000 [05:12<30:12, 1.87s/it, loss=0.674, lr=6e-5] \nSteps: 3%|▎ | 31/1000 [05:20<58:22, 3.61s/it, loss=0.674, lr=6e-5]\nSteps: 3%|▎ | 31/1000 [05:20<58:22, 3.61s/it, loss=0.638, lr=6.2e-5]\nSteps: 3%|▎ | 32/1000 [05:22<49:53, 3.09s/it, loss=0.638, lr=6.2e-5]\nSteps: 3%|▎ | 32/1000 [05:22<49:53, 3.09s/it, loss=1.04, lr=6.4e-5] \nSteps: 3%|▎ | 33/1000 [05:24<43:53, 2.72s/it, loss=1.04, lr=6.4e-5]\nSteps: 3%|▎ | 33/1000 [05:24<43:53, 2.72s/it, loss=0.504, lr=6.6e-5]\nSteps: 3%|▎ | 34/1000 [05:26<39:41, 2.47s/it, loss=0.504, lr=6.6e-5]\nSteps: 3%|▎ | 34/1000 [05:26<39:41, 2.47s/it, loss=0.638, lr=6.8e-5]\nSteps: 4%|▎ | 35/1000 [05:27<36:45, 2.29s/it, loss=0.638, lr=6.8e-5]\nSteps: 4%|▎ | 35/1000 [05:27<36:45, 2.29s/it, loss=1.01, lr=7e-5] \nSteps: 4%|▎ | 36/1000 [05:29<34:41, 2.16s/it, loss=1.01, lr=7e-5]\nSteps: 4%|▎ | 36/1000 [05:29<34:41, 2.16s/it, loss=1.03, lr=7.2e-5]\nSteps: 4%|▎ | 37/1000 [05:31<33:16, 2.07s/it, loss=1.03, lr=7.2e-5]\nSteps: 4%|▎ | 37/1000 [05:31<33:16, 2.07s/it, loss=0.447, lr=7.4e-5]\nSteps: 4%|▍ | 38/1000 [05:33<32:15, 2.01s/it, loss=0.447, lr=7.4e-5]\nSteps: 4%|▍ | 38/1000 [05:33<32:15, 2.01s/it, loss=0.56, lr=7.6e-5] \nSteps: 4%|▍ | 39/1000 [05:35<31:30, 1.97s/it, loss=0.56, lr=7.6e-5]\nSteps: 4%|▍ | 39/1000 [05:35<31:30, 1.97s/it, loss=0.317, lr=7.8e-5]\nSteps: 4%|▍ | 40/1000 [05:37<30:58, 1.94s/it, loss=0.317, lr=7.8e-5]\nSteps: 4%|▍ | 40/1000 [05:37<30:58, 1.94s/it, loss=0.787, lr=8e-5] \nSteps: 4%|▍ | 41/1000 [05:39<30:38, 1.92s/it, loss=0.787, lr=8e-5]\nSteps: 4%|▍ | 41/1000 [05:39<30:38, 1.92s/it, loss=0.309, lr=8.2e-5]\nSteps: 4%|▍ | 42/1000 [05:40<30:21, 1.90s/it, loss=0.309, lr=8.2e-5]\nSteps: 4%|▍ | 42/1000 [05:40<30:21, 1.90s/it, loss=0.805, lr=8.4e-5]\nSteps: 4%|▍ | 43/1000 [05:42<30:09, 1.89s/it, loss=0.805, lr=8.4e-5]\nSteps: 4%|▍ | 43/1000 [05:42<30:09, 1.89s/it, loss=1.1, lr=8.6e-5] \nSteps: 4%|▍ | 44/1000 [05:44<30:00, 1.88s/it, loss=1.1, lr=8.6e-5]\nSteps: 4%|▍ | 44/1000 [05:44<30:00, 1.88s/it, loss=0.307, lr=8.8e-5]\nSteps: 4%|▍ | 45/1000 [05:46<29:54, 1.88s/it, loss=0.307, lr=8.8e-5]\nSteps: 4%|▍ | 45/1000 [05:46<29:54, 1.88s/it, loss=0.991, lr=9e-5] \nSteps: 5%|▍ | 46/1000 [05:48<29:50, 1.88s/it, loss=0.991, lr=9e-5]\nSteps: 5%|▍ | 46/1000 [05:48<29:50, 1.88s/it, loss=0.431, lr=9.2e-5]\nSteps: 5%|▍ | 47/1000 [05:50<29:45, 1.87s/it, loss=0.431, lr=9.2e-5]\nSteps: 5%|▍ | 47/1000 [05:50<29:45, 1.87s/it, loss=0.301, lr=9.4e-5]\nSteps: 5%|▍ | 48/1000 [05:52<29:42, 1.87s/it, loss=0.301, lr=9.4e-5]\nSteps: 5%|▍ | 48/1000 [05:52<29:42, 1.87s/it, loss=0.78, lr=9.6e-5] \nSteps: 5%|▍ | 49/1000 [05:54<29:37, 1.87s/it, loss=0.78, lr=9.6e-5]\nSteps: 5%|▍ | 49/1000 [05:54<29:37, 1.87s/it, loss=0.699, lr=9.8e-5]\nSteps: 5%|▌ | 50/1000 [05:55<29:36, 1.87s/it, loss=0.699, lr=9.8e-5]\nSteps: 5%|▌ | 50/1000 [05:55<29:36, 1.87s/it, loss=0.784, lr=0.0001]\nSteps: 5%|▌ | 51/1000 [05:57<29:33, 1.87s/it, loss=0.784, lr=0.0001]\nSteps: 5%|▌ | 51/1000 [05:57<29:33, 1.87s/it, loss=0.487, lr=0.000102]\nSteps: 5%|▌ | 52/1000 [05:59<29:31, 1.87s/it, loss=0.487, lr=0.000102]\nSteps: 5%|▌ | 52/1000 [05:59<29:31, 1.87s/it, loss=0.608, lr=0.000104]\nSteps: 5%|▌ | 53/1000 [06:01<29:30, 1.87s/it, loss=0.608, lr=0.000104]\nSteps: 5%|▌ | 53/1000 [06:01<29:30, 1.87s/it, loss=0.371, lr=0.000106]\nSteps: 5%|▌ | 54/1000 [06:03<29:29, 1.87s/it, loss=0.371, lr=0.000106]\nSteps: 5%|▌ | 54/1000 [06:03<29:29, 1.87s/it, loss=0.302, lr=0.000108]\nSteps: 6%|▌ | 55/1000 [06:05<29:26, 1.87s/it, loss=0.302, lr=0.000108]\nSteps: 6%|▌ | 55/1000 [06:05<29:26, 1.87s/it, loss=0.568, lr=0.00011] \nSteps: 6%|▌ | 56/1000 [06:07<29:25, 1.87s/it, loss=0.568, lr=0.00011]\nSteps: 6%|▌ | 56/1000 [06:07<29:25, 1.87s/it, loss=0.316, lr=0.000112]\nSteps: 6%|▌ | 57/1000 [06:08<29:22, 1.87s/it, loss=0.316, lr=0.000112]\nSteps: 6%|▌ | 57/1000 [06:09<29:22, 1.87s/it, loss=0.611, lr=0.000114]\nSteps: 6%|▌ | 58/1000 [06:10<29:21, 1.87s/it, loss=0.611, lr=0.000114]\nSteps: 6%|▌ | 58/1000 [06:10<29:21, 1.87s/it, loss=0.531, lr=0.000116]\nSteps: 6%|▌ | 59/1000 [06:12<29:20, 1.87s/it, loss=0.531, lr=0.000116]\nSteps: 6%|▌ | 59/1000 [06:12<29:20, 1.87s/it, loss=0.451, lr=0.000118]\nSteps: 6%|▌ | 60/1000 [06:14<29:18, 1.87s/it, loss=0.451, lr=0.000118]\nSteps: 6%|▌ | 60/1000 [06:14<29:18, 1.87s/it, loss=0.353, lr=0.00012] \nSteps: 6%|▌ | 61/1000 [06:22<56:51, 3.63s/it, loss=0.353, lr=0.00012]\nSteps: 6%|▌ | 61/1000 [06:22<56:51, 3.63s/it, loss=0.44, lr=0.000122]\nSteps: 6%|▌ | 62/1000 [06:24<48:31, 3.10s/it, loss=0.44, lr=0.000122]\nSteps: 6%|▌ | 62/1000 [06:24<48:31, 3.10s/it, loss=0.314, lr=0.000124]\nSteps: 6%|▋ | 63/1000 [06:26<42:42, 2.73s/it, loss=0.314, lr=0.000124]\nSteps: 6%|▋ | 63/1000 [06:26<42:42, 2.73s/it, loss=0.364, lr=0.000126]\nSteps: 6%|▋ | 64/1000 [06:27<38:36, 2.47s/it, loss=0.364, lr=0.000126]\nSteps: 6%|▋ | 64/1000 [06:27<38:36, 2.47s/it, loss=0.35, lr=0.000128] \nSteps: 6%|▋ | 65/1000 [06:29<35:42, 2.29s/it, loss=0.35, lr=0.000128]\nSteps: 6%|▋ | 65/1000 [06:29<35:42, 2.29s/it, loss=0.293, lr=0.00013]\nSteps: 7%|▋ | 66/1000 [06:31<33:40, 2.16s/it, loss=0.293, lr=0.00013]\nSteps: 7%|▋ | 66/1000 [06:31<33:40, 2.16s/it, loss=0.978, lr=0.000132]\nSteps: 7%|▋ | 67/1000 [06:33<32:14, 2.07s/it, loss=0.978, lr=0.000132]\nSteps: 7%|▋ | 67/1000 [06:33<32:14, 2.07s/it, loss=0.847, lr=0.000134]\nSteps: 7%|▋ | 68/1000 [06:35<31:15, 2.01s/it, loss=0.847, lr=0.000134]\nSteps: 7%|▋ | 68/1000 [06:35<31:15, 2.01s/it, loss=0.442, lr=0.000136]\nSteps: 7%|▋ | 69/1000 [06:37<30:34, 1.97s/it, loss=0.442, lr=0.000136]\nSteps: 7%|▋ | 69/1000 [06:37<30:34, 1.97s/it, loss=0.295, lr=0.000138]\nSteps: 7%|▋ | 70/1000 [06:39<30:03, 1.94s/it, loss=0.295, lr=0.000138]\nSteps: 7%|▋ | 70/1000 [06:39<30:03, 1.94s/it, loss=0.314, lr=0.00014] \nSteps: 7%|▋ | 71/1000 [06:41<29:42, 1.92s/it, loss=0.314, lr=0.00014]\nSteps: 7%|▋ | 71/1000 [06:41<29:42, 1.92s/it, loss=1.03, lr=0.000142]\nSteps: 7%|▋ | 72/1000 [06:42<29:27, 1.90s/it, loss=1.03, lr=0.000142]\nSteps: 7%|▋ | 72/1000 [06:42<29:27, 1.90s/it, loss=0.524, lr=0.000144]\nSteps: 7%|▋ | 73/1000 [06:44<29:15, 1.89s/it, loss=0.524, lr=0.000144]\nSteps: 7%|▋ | 73/1000 [06:44<29:15, 1.89s/it, loss=0.3, lr=0.000146] \nSteps: 7%|▋ | 74/1000 [06:46<29:06, 1.89s/it, loss=0.3, lr=0.000146]\nSteps: 7%|▋ | 74/1000 [06:46<29:06, 1.89s/it, loss=0.374, lr=0.000148]\nSteps: 8%|▊ | 75/1000 [06:48<28:59, 1.88s/it, loss=0.374, lr=0.000148]\nSteps: 8%|▊ | 75/1000 [06:48<28:59, 1.88s/it, loss=0.328, lr=0.00015] \nSteps: 8%|▊ | 76/1000 [06:50<28:53, 1.88s/it, loss=0.328, lr=0.00015]\nSteps: 8%|▊ | 76/1000 [06:50<28:53, 1.88s/it, loss=0.547, lr=0.000152]\nSteps: 8%|▊ | 77/1000 [06:52<28:50, 1.88s/it, loss=0.547, lr=0.000152]\nSteps: 8%|▊ | 77/1000 [06:52<28:50, 1.88s/it, loss=0.301, lr=0.000154]\nSteps: 8%|▊ | 78/1000 [06:54<28:46, 1.87s/it, loss=0.301, lr=0.000154]\nSteps: 8%|▊ | 78/1000 [06:54<28:46, 1.87s/it, loss=1.02, lr=0.000156] \nSteps: 8%|▊ | 79/1000 [06:55<28:44, 1.87s/it, loss=1.02, lr=0.000156]\nSteps: 8%|▊ | 79/1000 [06:56<28:44, 1.87s/it, loss=0.303, lr=0.000158]\nSteps: 8%|▊ | 80/1000 [06:57<28:41, 1.87s/it, loss=0.303, lr=0.000158]\nSteps: 8%|▊ | 80/1000 [06:57<28:41, 1.87s/it, loss=0.386, lr=0.00016] \nSteps: 8%|▊ | 81/1000 [06:59<28:40, 1.87s/it, loss=0.386, lr=0.00016]\nSteps: 8%|▊ | 81/1000 [06:59<28:40, 1.87s/it, loss=0.399, lr=0.000162]\nSteps: 8%|▊ | 82/1000 [07:01<28:36, 1.87s/it, loss=0.399, lr=0.000162]\nSteps: 8%|▊ | 82/1000 [07:01<28:36, 1.87s/it, loss=0.47, lr=0.000164] \nSteps: 8%|▊ | 83/1000 [07:03<28:34, 1.87s/it, loss=0.47, lr=0.000164]\nSteps: 8%|▊ | 83/1000 [07:03<28:34, 1.87s/it, loss=0.909, lr=0.000166]\nSteps: 8%|▊ | 84/1000 [07:05<28:33, 1.87s/it, loss=0.909, lr=0.000166]\nSteps: 8%|▊ | 84/1000 [07:05<28:33, 1.87s/it, loss=0.284, lr=0.000168]\nSteps: 8%|▊ | 85/1000 [07:07<28:32, 1.87s/it, loss=0.284, lr=0.000168]\nSteps: 8%|▊ | 85/1000 [07:07<28:32, 1.87s/it, loss=0.52, lr=0.00017] \nSteps: 9%|▊ | 86/1000 [07:09<28:30, 1.87s/it, loss=0.52, lr=0.00017]\nSteps: 9%|▊ | 86/1000 [07:09<28:30, 1.87s/it, loss=0.286, lr=0.000172]\nSteps: 9%|▊ | 87/1000 [07:10<28:27, 1.87s/it, loss=0.286, lr=0.000172]\nSteps: 9%|▊ | 87/1000 [07:10<28:27, 1.87s/it, loss=0.642, lr=0.000174]\nSteps: 9%|▉ | 88/1000 [07:12<28:24, 1.87s/it, loss=0.642, lr=0.000174]\nSteps: 9%|▉ | 88/1000 [07:12<28:24, 1.87s/it, loss=0.305, lr=0.000176]\nSteps: 9%|▉ | 89/1000 [07:14<28:23, 1.87s/it, loss=0.305, lr=0.000176]\nSteps: 9%|▉ | 89/1000 [07:14<28:23, 1.87s/it, loss=1.01, lr=0.000178] \nSteps: 9%|▉ | 90/1000 [07:16<28:21, 1.87s/it, loss=1.01, lr=0.000178]\nSteps: 9%|▉ | 90/1000 [07:16<28:21, 1.87s/it, loss=0.287, lr=0.00018]\nSteps: 9%|▉ | 91/1000 [07:24<54:57, 3.63s/it, loss=0.287, lr=0.00018]\nSteps: 9%|▉ | 91/1000 [07:24<54:57, 3.63s/it, loss=0.731, lr=0.000182]\nSteps: 9%|▉ | 92/1000 [07:26<46:54, 3.10s/it, loss=0.731, lr=0.000182]\nSteps: 9%|▉ | 92/1000 [07:26<46:54, 3.10s/it, loss=0.585, lr=0.000184]\nSteps: 9%|▉ | 93/1000 [07:28<41:15, 2.73s/it, loss=0.585, lr=0.000184]\nSteps: 9%|▉ | 93/1000 [07:28<41:15, 2.73s/it, loss=0.737, lr=0.000186]\nSteps: 9%|▉ | 94/1000 [07:29<37:18, 2.47s/it, loss=0.737, lr=0.000186]\nSteps: 9%|▉ | 94/1000 [07:29<37:18, 2.47s/it, loss=0.679, lr=0.000188]\nSteps: 10%|▉ | 95/1000 [07:31<34:32, 2.29s/it, loss=0.679, lr=0.000188]\nSteps: 10%|▉ | 95/1000 [07:31<34:32, 2.29s/it, loss=0.305, lr=0.00019] \nSteps: 10%|▉ | 96/1000 [07:33<32:34, 2.16s/it, loss=0.305, lr=0.00019]\nSteps: 10%|▉ | 96/1000 [07:33<32:34, 2.16s/it, loss=0.355, lr=0.000192]\nSteps: 10%|▉ | 97/1000 [07:35<31:12, 2.07s/it, loss=0.355, lr=0.000192]\nSteps: 10%|▉ | 97/1000 [07:35<31:12, 2.07s/it, loss=0.331, lr=0.000194]\nSteps: 10%|▉ | 98/1000 [07:37<30:16, 2.01s/it, loss=0.331, lr=0.000194]\nSteps: 10%|▉ | 98/1000 [07:37<30:16, 2.01s/it, loss=0.954, lr=0.000196]\nSteps: 10%|▉ | 99/1000 [07:39<29:34, 1.97s/it, loss=0.954, lr=0.000196]\nSteps: 10%|▉ | 99/1000 [07:39<29:34, 1.97s/it, loss=0.692, lr=0.000198]\nSteps: 10%|█ | 100/1000 [07:41<29:05, 1.94s/it, loss=0.692, lr=0.000198]\nSteps: 10%|█ | 100/1000 [07:41<29:05, 1.94s/it, loss=0.329, lr=0.0002] \nSteps: 10%|█ | 101/1000 [07:42<28:44, 1.92s/it, loss=0.329, lr=0.0002]\nSteps: 10%|█ | 101/1000 [07:42<28:44, 1.92s/it, loss=0.283, lr=0.000202]\nSteps: 10%|█ | 102/1000 [07:44<28:30, 1.90s/it, loss=0.283, lr=0.000202]\nSteps: 10%|█ | 102/1000 [07:44<28:30, 1.90s/it, loss=0.633, lr=0.000204]\nSteps: 10%|█ | 103/1000 [07:46<28:19, 1.89s/it, loss=0.633, lr=0.000204]\nSteps: 10%|█ | 103/1000 [07:46<28:19, 1.89s/it, loss=0.355, lr=0.000206]\nSteps: 10%|█ | 104/1000 [07:48<28:10, 1.89s/it, loss=0.355, lr=0.000206]\nSteps: 10%|█ | 104/1000 [07:48<28:10, 1.89s/it, loss=1.03, lr=0.000208] \nSteps: 10%|█ | 105/1000 [07:50<28:04, 1.88s/it, loss=1.03, lr=0.000208]\nSteps: 10%|█ | 105/1000 [07:50<28:04, 1.88s/it, loss=0.62, lr=0.00021] \nSteps: 11%|█ | 106/1000 [07:52<27:59, 1.88s/it, loss=0.62, lr=0.00021]\nSteps: 11%|█ | 106/1000 [07:52<27:59, 1.88s/it, loss=0.404, lr=0.000212]\nSteps: 11%|█ | 107/1000 [07:54<27:54, 1.87s/it, loss=0.404, lr=0.000212]\nSteps: 11%|█ | 107/1000 [07:54<27:54, 1.87s/it, loss=0.22, lr=0.000214] \nSteps: 11%|█ | 108/1000 [07:56<27:52, 1.87s/it, loss=0.22, lr=0.000214]\nSteps: 11%|█ | 108/1000 [07:56<27:52, 1.87s/it, loss=0.314, lr=0.000216]\nSteps: 11%|█ | 109/1000 [07:57<27:49, 1.87s/it, loss=0.314, lr=0.000216]\nSteps: 11%|█ | 109/1000 [07:57<27:49, 1.87s/it, loss=0.704, lr=0.000218]\nSteps: 11%|█ | 110/1000 [07:59<27:45, 1.87s/it, loss=0.704, lr=0.000218]\nSteps: 11%|█ | 110/1000 [07:59<27:45, 1.87s/it, loss=0.539, lr=0.00022] \nSteps: 11%|█ | 111/1000 [08:01<27:43, 1.87s/it, loss=0.539, lr=0.00022]\nSteps: 11%|█ | 111/1000 [08:01<27:43, 1.87s/it, loss=0.569, lr=0.000222]\nSteps: 11%|█ | 112/1000 [08:03<27:40, 1.87s/it, loss=0.569, lr=0.000222]\nSteps: 11%|█ | 112/1000 [08:03<27:40, 1.87s/it, loss=0.591, lr=0.000224]\nSteps: 11%|█▏ | 113/1000 [08:05<27:39, 1.87s/it, loss=0.591, lr=0.000224]\nSteps: 11%|█▏ | 113/1000 [08:05<27:39, 1.87s/it, loss=0.32, lr=0.000226] \nSteps: 11%|█▏ | 114/1000 [08:07<27:36, 1.87s/it, loss=0.32, lr=0.000226]\nSteps: 11%|█▏ | 114/1000 [08:07<27:36, 1.87s/it, loss=0.462, lr=0.000228]\nSteps: 12%|█▏ | 115/1000 [08:09<27:34, 1.87s/it, loss=0.462, lr=0.000228]\nSteps: 12%|█▏ | 115/1000 [08:09<27:34, 1.87s/it, loss=0.409, lr=0.00023] \nSteps: 12%|█▏ | 116/1000 [08:11<27:32, 1.87s/it, loss=0.409, lr=0.00023]\nSteps: 12%|█▏ | 116/1000 [08:11<27:32, 1.87s/it, loss=0.943, lr=0.000232]\nSteps: 12%|█▏ | 117/1000 [08:12<27:30, 1.87s/it, loss=0.943, lr=0.000232]\nSteps: 12%|█▏ | 117/1000 [08:12<27:30, 1.87s/it, loss=0.33, lr=0.000234] \nSteps: 12%|█▏ | 118/1000 [08:14<27:28, 1.87s/it, loss=0.33, lr=0.000234]\nSteps: 12%|█▏ | 118/1000 [08:14<27:28, 1.87s/it, loss=0.447, lr=0.000236]\nSteps: 12%|█▏ | 119/1000 [08:16<27:27, 1.87s/it, loss=0.447, lr=0.000236]\nSteps: 12%|█▏ | 119/1000 [08:16<27:27, 1.87s/it, loss=0.929, lr=0.000238]\nSteps: 12%|█▏ | 120/1000 [08:18<27:24, 1.87s/it, loss=0.929, lr=0.000238]\nSteps: 12%|█▏ | 120/1000 [08:18<27:24, 1.87s/it, loss=0.908, lr=0.00024] \nSteps: 12%|█▏ | 121/1000 [08:26<53:04, 3.62s/it, loss=0.908, lr=0.00024]\nSteps: 12%|█▏ | 121/1000 [08:26<53:04, 3.62s/it, loss=0.81, lr=0.000242]\nSteps: 12%|█▏ | 122/1000 [08:28<45:18, 3.10s/it, loss=0.81, lr=0.000242]\nSteps: 12%|█▏ | 122/1000 [08:28<45:18, 3.10s/it, loss=0.315, lr=0.000244]\nSteps: 12%|█▏ | 123/1000 [08:29<39:51, 2.73s/it, loss=0.315, lr=0.000244]\nSteps: 12%|█▏ | 123/1000 [08:29<39:51, 2.73s/it, loss=0.311, lr=0.000246]\nSteps: 12%|█▏ | 124/1000 [08:31<36:03, 2.47s/it, loss=0.311, lr=0.000246]\nSteps: 12%|█▏ | 124/1000 [08:31<36:03, 2.47s/it, loss=0.634, lr=0.000248]\nSteps: 12%|█▎ | 125/1000 [08:33<33:20, 2.29s/it, loss=0.634, lr=0.000248]\nSteps: 12%|█▎ | 125/1000 [08:33<33:20, 2.29s/it, loss=0.728, lr=0.00025] \nSteps: 13%|█▎ | 126/1000 [08:35<31:28, 2.16s/it, loss=0.728, lr=0.00025]\nSteps: 13%|█▎ | 126/1000 [08:35<31:28, 2.16s/it, loss=0.38, lr=0.000252]\nSteps: 13%|█▎ | 127/1000 [08:37<30:09, 2.07s/it, loss=0.38, lr=0.000252]\nSteps: 13%|█▎ | 127/1000 [08:37<30:09, 2.07s/it, loss=0.335, lr=0.000254]\nSteps: 13%|█▎ | 128/1000 [08:39<29:14, 2.01s/it, loss=0.335, lr=0.000254]\nSteps: 13%|█▎ | 128/1000 [08:39<29:14, 2.01s/it, loss=0.41, lr=0.000256] \nSteps: 13%|█▎ | 129/1000 [08:41<28:36, 1.97s/it, loss=0.41, lr=0.000256]\nSteps: 13%|█▎ | 129/1000 [08:41<28:36, 1.97s/it, loss=0.336, lr=0.000258]\nSteps: 13%|█▎ | 130/1000 [08:43<28:06, 1.94s/it, loss=0.336, lr=0.000258]\nSteps: 13%|█▎ | 130/1000 [08:43<28:06, 1.94s/it, loss=0.8, lr=0.00026] \nSteps: 13%|█▎ | 131/1000 [08:44<27:46, 1.92s/it, loss=0.8, lr=0.00026]\nSteps: 13%|█▎ | 131/1000 [08:44<27:46, 1.92s/it, loss=0.97, lr=0.000262]\nSteps: 13%|█▎ | 132/1000 [08:46<27:31, 1.90s/it, loss=0.97, lr=0.000262]\nSteps: 13%|█▎ | 132/1000 [08:46<27:31, 1.90s/it, loss=0.688, lr=0.000264]\nSteps: 13%|█▎ | 133/1000 [08:48<27:21, 1.89s/it, loss=0.688, lr=0.000264]\nSteps: 13%|█▎ | 133/1000 [08:48<27:21, 1.89s/it, loss=0.557, lr=0.000266]\nSteps: 13%|█▎ | 134/1000 [08:50<27:13, 1.89s/it, loss=0.557, lr=0.000266]\nSteps: 13%|█▎ | 134/1000 [08:50<27:13, 1.89s/it, loss=0.548, lr=0.000268]\nSteps: 14%|█▎ | 135/1000 [08:52<27:07, 1.88s/it, loss=0.548, lr=0.000268]\nSteps: 14%|█▎ | 135/1000 [08:52<27:07, 1.88s/it, loss=0.355, lr=0.00027] \nSteps: 14%|█▎ | 136/1000 [08:54<27:04, 1.88s/it, loss=0.355, lr=0.00027]\nSteps: 14%|█▎ | 136/1000 [08:54<27:04, 1.88s/it, loss=0.873, lr=0.000272]\nSteps: 14%|█▎ | 137/1000 [08:56<26:59, 1.88s/it, loss=0.873, lr=0.000272]\nSteps: 14%|█▎ | 137/1000 [08:56<26:59, 1.88s/it, loss=0.217, lr=0.000274]\nSteps: 14%|█▍ | 138/1000 [08:57<26:55, 1.87s/it, loss=0.217, lr=0.000274]\nSteps: 14%|█▍ | 138/1000 [08:57<26:55, 1.87s/it, loss=0.332, lr=0.000276]\nSteps: 14%|█▍ | 139/1000 [08:59<26:52, 1.87s/it, loss=0.332, lr=0.000276]\nSteps: 14%|█▍ | 139/1000 [08:59<26:52, 1.87s/it, loss=0.547, lr=0.000278]\nSteps: 14%|█▍ | 140/1000 [09:01<26:49, 1.87s/it, loss=0.547, lr=0.000278]\nSteps: 14%|█▍ | 140/1000 [09:01<26:49, 1.87s/it, loss=0.644, lr=0.00028] \nSteps: 14%|█▍ | 141/1000 [09:03<26:47, 1.87s/it, loss=0.644, lr=0.00028]\nSteps: 14%|█▍ | 141/1000 [09:03<26:47, 1.87s/it, loss=0.493, lr=0.000282]\nSteps: 14%|█▍ | 142/1000 [09:05<26:45, 1.87s/it, loss=0.493, lr=0.000282]\nSteps: 14%|█▍ | 142/1000 [09:05<26:45, 1.87s/it, loss=0.339, lr=0.000284]\nSteps: 14%|█▍ | 143/1000 [09:07<26:42, 1.87s/it, loss=0.339, lr=0.000284]\nSteps: 14%|█▍ | 143/1000 [09:07<26:42, 1.87s/it, loss=0.47, lr=0.000286] \nSteps: 14%|█▍ | 144/1000 [09:09<26:41, 1.87s/it, loss=0.47, lr=0.000286]\nSteps: 14%|█▍ | 144/1000 [09:09<26:41, 1.87s/it, loss=0.236, lr=0.000288]\nSteps: 14%|█▍ | 145/1000 [09:11<26:39, 1.87s/it, loss=0.236, lr=0.000288]\nSteps: 14%|█▍ | 145/1000 [09:11<26:39, 1.87s/it, loss=0.722, lr=0.00029] \nSteps: 15%|█▍ | 146/1000 [09:12<26:36, 1.87s/it, loss=0.722, lr=0.00029]\nSteps: 15%|█▍ | 146/1000 [09:12<26:36, 1.87s/it, loss=0.636, lr=0.000292]\nSteps: 15%|█▍ | 147/1000 [09:14<26:34, 1.87s/it, loss=0.636, lr=0.000292]\nSteps: 15%|█▍ | 147/1000 [09:14<26:34, 1.87s/it, loss=0.563, lr=0.000294]\nSteps: 15%|█▍ | 148/1000 [09:16<26:32, 1.87s/it, loss=0.563, lr=0.000294]\nSteps: 15%|█▍ | 148/1000 [09:16<26:32, 1.87s/it, loss=0.534, lr=0.000296]\nSteps: 15%|█▍ | 149/1000 [09:18<26:31, 1.87s/it, loss=0.534, lr=0.000296]\nSteps: 15%|█▍ | 149/1000 [09:18<26:31, 1.87s/it, loss=0.71, lr=0.000298] \nSteps: 15%|█▌ | 150/1000 [09:20<26:30, 1.87s/it, loss=0.71, lr=0.000298]\nSteps: 15%|█▌ | 150/1000 [09:20<26:30, 1.87s/it, loss=0.825, lr=0.0003] \nSteps: 15%|█▌ | 151/1000 [09:28<51:04, 3.61s/it, loss=0.825, lr=0.0003]\nSteps: 15%|█▌ | 151/1000 [09:28<51:04, 3.61s/it, loss=0.336, lr=0.000302]\nSteps: 15%|█▌ | 152/1000 [09:29<43:37, 3.09s/it, loss=0.336, lr=0.000302]\nSteps: 15%|█▌ | 152/1000 [09:29<43:37, 3.09s/it, loss=0.331, lr=0.000304]\nSteps: 15%|█▌ | 153/1000 [09:31<38:24, 2.72s/it, loss=0.331, lr=0.000304]\nSteps: 15%|█▌ | 153/1000 [09:31<38:24, 2.72s/it, loss=0.313, lr=0.000306]\nSteps: 15%|█▌ | 154/1000 [09:33<34:45, 2.46s/it, loss=0.313, lr=0.000306]\nSteps: 15%|█▌ | 154/1000 [09:33<34:45, 2.46s/it, loss=0.345, lr=0.000308]\nSteps: 16%|█▌ | 155/1000 [09:35<32:11, 2.29s/it, loss=0.345, lr=0.000308]\nSteps: 16%|█▌ | 155/1000 [09:35<32:11, 2.29s/it, loss=0.606, lr=0.00031] \nSteps: 16%|█▌ | 156/1000 [09:37<30:24, 2.16s/it, loss=0.606, lr=0.00031]\nSteps: 16%|█▌ | 156/1000 [09:37<30:24, 2.16s/it, loss=0.288, lr=0.000312]\nSteps: 16%|█▌ | 157/1000 [09:39<29:08, 2.07s/it, loss=0.288, lr=0.000312]\nSteps: 16%|█▌ | 157/1000 [09:39<29:08, 2.07s/it, loss=0.866, lr=0.000314]\nSteps: 16%|█▌ | 158/1000 [09:41<28:14, 2.01s/it, loss=0.866, lr=0.000314]\nSteps: 16%|█▌ | 158/1000 [09:41<28:14, 2.01s/it, loss=0.418, lr=0.000316]\nSteps: 16%|█▌ | 159/1000 [09:43<27:36, 1.97s/it, loss=0.418, lr=0.000316]\nSteps: 16%|█▌ | 159/1000 [09:43<27:36, 1.97s/it, loss=0.55, lr=0.000318] \nSteps: 16%|█▌ | 160/1000 [09:44<27:09, 1.94s/it, loss=0.55, lr=0.000318]\nSteps: 16%|█▌ | 160/1000 [09:44<27:09, 1.94s/it, loss=0.516, lr=0.00032]\nSteps: 16%|█▌ | 161/1000 [09:46<26:50, 1.92s/it, loss=0.516, lr=0.00032]\nSteps: 16%|█▌ | 161/1000 [09:46<26:50, 1.92s/it, loss=0.978, lr=0.000322]\nSteps: 16%|█▌ | 162/1000 [09:48<26:35, 1.90s/it, loss=0.978, lr=0.000322]\nSteps: 16%|█▌ | 162/1000 [09:48<26:35, 1.90s/it, loss=0.323, lr=0.000324]\nSteps: 16%|█▋ | 163/1000 [09:50<26:25, 1.89s/it, loss=0.323, lr=0.000324]\nSteps: 16%|█▋ | 163/1000 [09:50<26:25, 1.89s/it, loss=0.346, lr=0.000326]\nSteps: 16%|█▋ | 164/1000 [09:52<26:16, 1.89s/it, loss=0.346, lr=0.000326]\nSteps: 16%|█▋ | 164/1000 [09:52<26:16, 1.89s/it, loss=0.55, lr=0.000328] \nSteps: 16%|█▋ | 165/1000 [09:54<26:11, 1.88s/it, loss=0.55, lr=0.000328]\nSteps: 16%|█▋ | 165/1000 [09:54<26:11, 1.88s/it, loss=0.918, lr=0.00033]\nSteps: 17%|█▋ | 166/1000 [09:56<26:06, 1.88s/it, loss=0.918, lr=0.00033]\nSteps: 17%|█▋ | 166/1000 [09:56<26:06, 1.88s/it, loss=0.73, lr=0.000332]\nSteps: 17%|█▋ | 167/1000 [09:57<26:02, 1.88s/it, loss=0.73, lr=0.000332]\nSteps: 17%|█▋ | 167/1000 [09:58<26:02, 1.88s/it, loss=0.521, lr=0.000334]\nSteps: 17%|█▋ | 168/1000 [09:59<25:59, 1.87s/it, loss=0.521, lr=0.000334]\nSteps: 17%|█▋ | 168/1000 [09:59<25:59, 1.87s/it, loss=0.319, lr=0.000336]\nSteps: 17%|█▋ | 169/1000 [10:01<25:56, 1.87s/it, loss=0.319, lr=0.000336]\nSteps: 17%|█▋ | 169/1000 [10:01<25:56, 1.87s/it, loss=0.307, lr=0.000338]\nSteps: 17%|█▋ | 170/1000 [10:03<25:51, 1.87s/it, loss=0.307, lr=0.000338]\nSteps: 17%|█▋ | 170/1000 [10:03<25:51, 1.87s/it, loss=0.336, lr=0.00034] \nSteps: 17%|█▋ | 171/1000 [10:05<25:48, 1.87s/it, loss=0.336, lr=0.00034]\nSteps: 17%|█▋ | 171/1000 [10:05<25:48, 1.87s/it, loss=0.472, lr=0.000342]\nSteps: 17%|█▋ | 172/1000 [10:07<25:48, 1.87s/it, loss=0.472, lr=0.000342]\nSteps: 17%|█▋ | 172/1000 [10:07<25:48, 1.87s/it, loss=0.364, lr=0.000344]\nSteps: 17%|█▋ | 173/1000 [10:09<25:46, 1.87s/it, loss=0.364, lr=0.000344]\nSteps: 17%|█▋ | 173/1000 [10:09<25:46, 1.87s/it, loss=0.311, lr=0.000346]\nSteps: 17%|█▋ | 174/1000 [10:11<25:44, 1.87s/it, loss=0.311, lr=0.000346]\nSteps: 17%|█▋ | 174/1000 [10:11<25:44, 1.87s/it, loss=0.228, lr=0.000348]\nSteps: 18%|█▊ | 175/1000 [10:12<25:41, 1.87s/it, loss=0.228, lr=0.000348]\nSteps: 18%|█▊ | 175/1000 [10:12<25:41, 1.87s/it, loss=0.406, lr=0.00035] \nSteps: 18%|█▊ | 176/1000 [10:14<25:38, 1.87s/it, loss=0.406, lr=0.00035]\nSteps: 18%|█▊ | 176/1000 [10:14<25:38, 1.87s/it, loss=0.322, lr=0.000352]\nSteps: 18%|█▊ | 177/1000 [10:16<25:36, 1.87s/it, loss=0.322, lr=0.000352]\nSteps: 18%|█▊ | 177/1000 [10:16<25:36, 1.87s/it, loss=0.417, lr=0.000354]\nSteps: 18%|█▊ | 178/1000 [10:18<25:35, 1.87s/it, loss=0.417, lr=0.000354]\nSteps: 18%|█▊ | 178/1000 [10:18<25:35, 1.87s/it, loss=0.71, lr=0.000356] \nSteps: 18%|█▊ | 179/1000 [10:20<25:34, 1.87s/it, loss=0.71, lr=0.000356]\nSteps: 18%|█▊ | 179/1000 [10:20<25:34, 1.87s/it, loss=0.443, lr=0.000358]\nSteps: 18%|█▊ | 180/1000 [10:22<25:33, 1.87s/it, loss=0.443, lr=0.000358]\nSteps: 18%|█▊ | 180/1000 [10:22<25:33, 1.87s/it, loss=0.893, lr=0.00036] \nSteps: 18%|█▊ | 181/1000 [10:29<49:23, 3.62s/it, loss=0.893, lr=0.00036]\nSteps: 18%|█▊ | 181/1000 [10:29<49:23, 3.62s/it, loss=0.798, lr=0.000362]\nSteps: 18%|█▊ | 182/1000 [10:31<42:10, 3.09s/it, loss=0.798, lr=0.000362]\nSteps: 18%|█▊ | 182/1000 [10:31<42:10, 3.09s/it, loss=1.03, lr=0.000364] \nSteps: 18%|█▊ | 183/1000 [10:33<37:06, 2.73s/it, loss=1.03, lr=0.000364]\nSteps: 18%|█▊ | 183/1000 [10:33<37:06, 2.73s/it, loss=0.711, lr=0.000366]\nSteps: 18%|█▊ | 184/1000 [10:35<33:33, 2.47s/it, loss=0.711, lr=0.000366]\nSteps: 18%|█▊ | 184/1000 [10:35<33:33, 2.47s/it, loss=0.311, lr=0.000368]\nSteps: 18%|█▊ | 185/1000 [10:37<31:05, 2.29s/it, loss=0.311, lr=0.000368]\nSteps: 18%|█▊ | 185/1000 [10:37<31:05, 2.29s/it, loss=1.05, lr=0.00037] \nSteps: 19%|█▊ | 186/1000 [10:39<29:20, 2.16s/it, loss=1.05, lr=0.00037]\nSteps: 19%|█▊ | 186/1000 [10:39<29:20, 2.16s/it, loss=0.781, lr=0.000372]\nSteps: 19%|█▊ | 187/1000 [10:41<28:07, 2.08s/it, loss=0.781, lr=0.000372]\nSteps: 19%|█▊ | 187/1000 [10:41<28:07, 2.08s/it, loss=0.506, lr=0.000374]\nSteps: 19%|█▉ | 188/1000 [10:43<27:14, 2.01s/it, loss=0.506, lr=0.000374]\nSteps: 19%|█▉ | 188/1000 [10:43<27:14, 2.01s/it, loss=0.415, lr=0.000376]\nSteps: 19%|█▉ | 189/1000 [10:44<26:38, 1.97s/it, loss=0.415, lr=0.000376]\nSteps: 19%|█▉ | 189/1000 [10:44<26:38, 1.97s/it, loss=0.37, lr=0.000378] \nSteps: 19%|█▉ | 190/1000 [10:46<26:10, 1.94s/it, loss=0.37, lr=0.000378]\nSteps: 19%|█▉ | 190/1000 [10:46<26:10, 1.94s/it, loss=0.327, lr=0.00038]\nSteps: 19%|█▉ | 191/1000 [10:48<25:51, 1.92s/it, loss=0.327, lr=0.00038]\nSteps: 19%|█▉ | 191/1000 [10:48<25:51, 1.92s/it, loss=0.883, lr=0.000382]\nSteps: 19%|█▉ | 192/1000 [10:50<25:38, 1.90s/it, loss=0.883, lr=0.000382]\nSteps: 19%|█▉ | 192/1000 [10:50<25:38, 1.90s/it, loss=0.868, lr=0.000384]\nSteps: 19%|█▉ | 193/1000 [10:52<25:29, 1.89s/it, loss=0.868, lr=0.000384]\nSteps: 19%|█▉ | 193/1000 [10:52<25:29, 1.89s/it, loss=0.294, lr=0.000386]\nSteps: 19%|█▉ | 194/1000 [10:54<25:21, 1.89s/it, loss=0.294, lr=0.000386]\nSteps: 19%|█▉ | 194/1000 [10:54<25:21, 1.89s/it, loss=0.529, lr=0.000388]\nSteps: 20%|█▉ | 195/1000 [10:56<25:15, 1.88s/it, loss=0.529, lr=0.000388]\nSteps: 20%|█▉ | 195/1000 [10:56<25:15, 1.88s/it, loss=0.343, lr=0.00039] \nSteps: 20%|█▉ | 196/1000 [10:58<25:10, 1.88s/it, loss=0.343, lr=0.00039]\nSteps: 20%|█▉ | 196/1000 [10:58<25:10, 1.88s/it, loss=0.996, lr=0.000392]\nSteps: 20%|█▉ | 197/1000 [10:59<25:06, 1.88s/it, loss=0.996, lr=0.000392]\nSteps: 20%|█▉ | 197/1000 [10:59<25:06, 1.88s/it, loss=0.36, lr=0.000394] \nSteps: 20%|█▉ | 198/1000 [11:01<25:02, 1.87s/it, loss=0.36, lr=0.000394]\nSteps: 20%|█▉ | 198/1000 [11:01<25:02, 1.87s/it, loss=0.869, lr=0.000396]\nSteps: 20%|█▉ | 199/1000 [11:03<25:00, 1.87s/it, loss=0.869, lr=0.000396]\nSteps: 20%|█▉ | 199/1000 [11:03<25:00, 1.87s/it, loss=1.02, lr=0.000398] \nSteps: 20%|██ | 200/1000 [11:05<24:58, 1.87s/it, loss=1.02, lr=0.000398]\nSteps: 20%|██ | 200/1000 [11:05<24:58, 1.87s/it, loss=0.336, lr=0.0004] \nSteps: 20%|██ | 201/1000 [11:07<24:54, 1.87s/it, loss=0.336, lr=0.0004]\nSteps: 20%|██ | 201/1000 [11:07<24:54, 1.87s/it, loss=0.51, lr=0.0004] \nSteps: 20%|██ | 202/1000 [11:09<24:52, 1.87s/it, loss=0.51, lr=0.0004]\nSteps: 20%|██ | 202/1000 [11:09<24:52, 1.87s/it, loss=0.543, lr=0.0004]\nSteps: 20%|██ | 203/1000 [11:11<24:51, 1.87s/it, loss=0.543, lr=0.0004]\nSteps: 20%|██ | 203/1000 [11:11<24:51, 1.87s/it, loss=1.08, lr=0.0004] \nSteps: 20%|██ | 204/1000 [11:12<24:48, 1.87s/it, loss=1.08, lr=0.0004]\nSteps: 20%|██ | 204/1000 [11:12<24:48, 1.87s/it, loss=0.29, lr=0.0004]\nSteps: 20%|██ | 205/1000 [11:14<24:47, 1.87s/it, loss=0.29, lr=0.0004]\nSteps: 20%|██ | 205/1000 [11:14<24:47, 1.87s/it, loss=0.432, lr=0.0004]\nSteps: 21%|██ | 206/1000 [11:16<24:45, 1.87s/it, loss=0.432, lr=0.0004]\nSteps: 21%|██ | 206/1000 [11:16<24:45, 1.87s/it, loss=0.486, lr=0.0004]\nSteps: 21%|██ | 207/1000 [11:18<24:44, 1.87s/it, loss=0.486, lr=0.0004]\nSteps: 21%|██ | 207/1000 [11:18<24:44, 1.87s/it, loss=0.376, lr=0.0004]\nSteps: 21%|██ | 208/1000 [11:20<24:43, 1.87s/it, loss=0.376, lr=0.0004]\nSteps: 21%|██ | 208/1000 [11:20<24:43, 1.87s/it, loss=1.03, lr=0.0004] \nSteps: 21%|██ | 209/1000 [11:22<24:40, 1.87s/it, loss=1.03, lr=0.0004]\nSteps: 21%|██ | 209/1000 [11:22<24:40, 1.87s/it, loss=0.757, lr=0.0004]\nSteps: 21%|██ | 210/1000 [11:24<24:37, 1.87s/it, loss=0.757, lr=0.0004]\nSteps: 21%|██ | 210/1000 [11:24<24:37, 1.87s/it, loss=0.469, lr=0.0004]\nSteps: 21%|██ | 211/1000 [11:31<47:37, 3.62s/it, loss=0.469, lr=0.0004]\nSteps: 21%|██ | 211/1000 [11:31<47:37, 3.62s/it, loss=0.361, lr=0.0004]\nSteps: 21%|██ | 212/1000 [11:33<40:39, 3.10s/it, loss=0.361, lr=0.0004]\nSteps: 21%|██ | 212/1000 [11:33<40:39, 3.10s/it, loss=0.325, lr=0.0004]\nSteps: 21%|██▏ | 213/1000 [11:35<35:45, 2.73s/it, loss=0.325, lr=0.0004]\nSteps: 21%|██▏ | 213/1000 [11:35<35:45, 2.73s/it, loss=0.449, lr=0.0004]\nSteps: 21%|██▏ | 214/1000 [11:37<32:20, 2.47s/it, loss=0.449, lr=0.0004]\nSteps: 21%|██▏ | 214/1000 [11:37<32:20, 2.47s/it, loss=0.918, lr=0.0004]\nSteps: 22%|██▏ | 215/1000 [11:39<29:56, 2.29s/it, loss=0.918, lr=0.0004]\nSteps: 22%|██▏ | 215/1000 [11:39<29:56, 2.29s/it, loss=0.51, lr=0.0004] \nSteps: 22%|██▏ | 216/1000 [11:41<28:15, 2.16s/it, loss=0.51, lr=0.0004]\nSteps: 22%|██▏ | 216/1000 [11:41<28:15, 2.16s/it, loss=0.909, lr=0.0004]\nSteps: 22%|██▏ | 217/1000 [11:43<27:05, 2.08s/it, loss=0.909, lr=0.0004]\nSteps: 22%|██▏ | 217/1000 [11:43<27:05, 2.08s/it, loss=0.676, lr=0.0004]\nSteps: 22%|██▏ | 218/1000 [11:45<26:14, 2.01s/it, loss=0.676, lr=0.0004]\nSteps: 22%|██▏ | 218/1000 [11:45<26:14, 2.01s/it, loss=0.345, lr=0.0004]\nSteps: 22%|██▏ | 219/1000 [11:46<25:40, 1.97s/it, loss=0.345, lr=0.0004]\nSteps: 22%|██▏ | 219/1000 [11:46<25:40, 1.97s/it, loss=0.619, lr=0.000399]\nSteps: 22%|██▏ | 220/1000 [11:48<25:14, 1.94s/it, loss=0.619, lr=0.000399]\nSteps: 22%|██▏ | 220/1000 [11:48<25:14, 1.94s/it, loss=0.333, lr=0.000399]\nSteps: 22%|██▏ | 221/1000 [11:50<24:55, 1.92s/it, loss=0.333, lr=0.000399]\nSteps: 22%|██▏ | 221/1000 [11:50<24:55, 1.92s/it, loss=0.915, lr=0.000399]\nSteps: 22%|██▏ | 222/1000 [11:52<24:41, 1.90s/it, loss=0.915, lr=0.000399]\nSteps: 22%|██▏ | 222/1000 [11:52<24:41, 1.90s/it, loss=0.36, lr=0.000399] \nSteps: 22%|██▏ | 223/1000 [11:54<24:31, 1.89s/it, loss=0.36, lr=0.000399]\nSteps: 22%|██▏ | 223/1000 [11:54<24:31, 1.89s/it, loss=0.39, lr=0.000399]\nSteps: 22%|██▏ | 224/1000 [11:56<24:24, 1.89s/it, loss=0.39, lr=0.000399]\nSteps: 22%|██▏ | 224/1000 [11:56<24:24, 1.89s/it, loss=1, lr=0.000399] \nSteps: 22%|██▎ | 225/1000 [11:58<24:19, 1.88s/it, loss=1, lr=0.000399]\nSteps: 22%|██▎ | 225/1000 [11:58<24:19, 1.88s/it, loss=0.49, lr=0.000399]\nSteps: 23%|██▎ | 226/1000 [11:59<24:13, 1.88s/it, loss=0.49, lr=0.000399]\nSteps: 23%|██▎ | 226/1000 [11:59<24:13, 1.88s/it, loss=0.729, lr=0.000399]\nSteps: 23%|██▎ | 227/1000 [12:01<24:10, 1.88s/it, loss=0.729, lr=0.000399]\nSteps: 23%|██▎ | 227/1000 [12:01<24:10, 1.88s/it, loss=0.512, lr=0.000399]\nSteps: 23%|██▎ | 228/1000 [12:03<24:07, 1.87s/it, loss=0.512, lr=0.000399]\nSteps: 23%|██▎ | 228/1000 [12:03<24:07, 1.87s/it, loss=0.311, lr=0.000399]\nSteps: 23%|██▎ | 229/1000 [12:05<24:04, 1.87s/it, loss=0.311, lr=0.000399]\nSteps: 23%|██▎ | 229/1000 [12:05<24:04, 1.87s/it, loss=0.6, lr=0.000399] \nSteps: 23%|██▎ | 230/1000 [12:07<24:00, 1.87s/it, loss=0.6, lr=0.000399]\nSteps: 23%|██▎ | 230/1000 [12:07<24:00, 1.87s/it, loss=0.635, lr=0.000399]\nSteps: 23%|██▎ | 231/1000 [12:09<24:00, 1.87s/it, loss=0.635, lr=0.000399]\nSteps: 23%|██▎ | 231/1000 [12:09<24:00, 1.87s/it, loss=0.945, lr=0.000399]\nSteps: 23%|██▎ | 232/1000 [12:11<23:59, 1.87s/it, loss=0.945, lr=0.000399]\nSteps: 23%|██▎ | 232/1000 [12:11<23:59, 1.87s/it, loss=0.644, lr=0.000398]\nSteps: 23%|██▎ | 233/1000 [12:13<23:56, 1.87s/it, loss=0.644, lr=0.000398]\nSteps: 23%|██▎ | 233/1000 [12:13<23:56, 1.87s/it, loss=0.553, lr=0.000398]\nSteps: 23%|██▎ | 234/1000 [12:14<23:54, 1.87s/it, loss=0.553, lr=0.000398]\nSteps: 23%|██▎ | 234/1000 [12:14<23:54, 1.87s/it, loss=0.975, lr=0.000398]\nSteps: 24%|██▎ | 235/1000 [12:16<23:51, 1.87s/it, loss=0.975, lr=0.000398]\nSteps: 24%|██▎ | 235/1000 [12:16<23:51, 1.87s/it, loss=0.839, lr=0.000398]\nSteps: 24%|██▎ | 236/1000 [12:18<23:50, 1.87s/it, loss=0.839, lr=0.000398]\nSteps: 24%|██▎ | 236/1000 [12:18<23:50, 1.87s/it, loss=0.346, lr=0.000398]\nSteps: 24%|██▎ | 237/1000 [12:20<23:48, 1.87s/it, loss=0.346, lr=0.000398]\nSteps: 24%|██▎ | 237/1000 [12:20<23:48, 1.87s/it, loss=0.325, lr=0.000398]\nSteps: 24%|██▍ | 238/1000 [12:22<23:46, 1.87s/it, loss=0.325, lr=0.000398]\nSteps: 24%|██▍ | 238/1000 [12:22<23:46, 1.87s/it, loss=0.562, lr=0.000398]\nSteps: 24%|██▍ | 239/1000 [12:24<23:44, 1.87s/it, loss=0.562, lr=0.000398]\nSteps: 24%|██▍ | 239/1000 [12:24<23:44, 1.87s/it, loss=0.508, lr=0.000398]\nSteps: 24%|██▍ | 240/1000 [12:26<23:42, 1.87s/it, loss=0.508, lr=0.000398]\nSteps: 24%|██▍ | 240/1000 [12:26<23:42, 1.87s/it, loss=0.486, lr=0.000398]\nSteps: 24%|██▍ | 241/1000 [12:33<45:44, 3.62s/it, loss=0.486, lr=0.000398]\nSteps: 24%|██▍ | 241/1000 [12:33<45:44, 3.62s/it, loss=0.593, lr=0.000397]\nSteps: 24%|██▍ | 242/1000 [12:35<39:01, 3.09s/it, loss=0.593, lr=0.000397]\nSteps: 24%|██▍ | 242/1000 [12:35<39:01, 3.09s/it, loss=0.567, lr=0.000397]\nSteps: 24%|██▍ | 243/1000 [12:37<34:20, 2.72s/it, loss=0.567, lr=0.000397]\nSteps: 24%|██▍ | 243/1000 [12:37<34:20, 2.72s/it, loss=0.515, lr=0.000397]\nSteps: 24%|██▍ | 244/1000 [12:39<31:03, 2.46s/it, loss=0.515, lr=0.000397]\nSteps: 24%|██▍ | 244/1000 [12:39<31:03, 2.46s/it, loss=0.465, lr=0.000397]\nSteps: 24%|██▍ | 245/1000 [12:41<28:46, 2.29s/it, loss=0.465, lr=0.000397]\nSteps: 24%|██▍ | 245/1000 [12:41<28:46, 2.29s/it, loss=1.02, lr=0.000397] \nSteps: 25%|██▍ | 246/1000 [12:43<27:09, 2.16s/it, loss=1.02, lr=0.000397]\nSteps: 25%|██▍ | 246/1000 [12:43<27:09, 2.16s/it, loss=0.31, lr=0.000397]\nSteps: 25%|██▍ | 247/1000 [12:45<26:01, 2.07s/it, loss=0.31, lr=0.000397]\nSteps: 25%|██▍ | 247/1000 [12:45<26:01, 2.07s/it, loss=0.84, lr=0.000397]\nSteps: 25%|██▍ | 248/1000 [12:46<25:13, 2.01s/it, loss=0.84, lr=0.000397]\nSteps: 25%|██▍ | 248/1000 [12:46<25:13, 2.01s/it, loss=0.425, lr=0.000396]\nSteps: 25%|██▍ | 249/1000 [12:48<24:39, 1.97s/it, loss=0.425, lr=0.000396]\nSteps: 25%|██▍ | 249/1000 [12:48<24:39, 1.97s/it, loss=0.586, lr=0.000396]\nSteps: 25%|██▌ | 250/1000 [12:50<24:14, 1.94s/it, loss=0.586, lr=0.000396]\nSteps: 25%|██▌ | 250/1000 [12:50<24:14, 1.94s/it, loss=0.319, lr=0.000396]\nSteps: 25%|██▌ | 251/1000 [12:52<23:55, 1.92s/it, loss=0.319, lr=0.000396]\nSteps: 25%|██▌ | 251/1000 [12:52<23:55, 1.92s/it, loss=0.498, lr=0.000396]\nSteps: 25%|██▌ | 252/1000 [12:54<23:41, 1.90s/it, loss=0.498, lr=0.000396]\nSteps: 25%|██▌ | 252/1000 [12:54<23:41, 1.90s/it, loss=0.296, lr=0.000396]\nSteps: 25%|██▌ | 253/1000 [12:56<23:30, 1.89s/it, loss=0.296, lr=0.000396]\nSteps: 25%|██▌ | 253/1000 [12:56<23:30, 1.89s/it, loss=0.635, lr=0.000396]\nSteps: 25%|██▌ | 254/1000 [12:58<23:23, 1.88s/it, loss=0.635, lr=0.000396]\nSteps: 25%|██▌ | 254/1000 [12:58<23:23, 1.88s/it, loss=0.294, lr=0.000396]\nSteps: 26%|██▌ | 255/1000 [12:59<23:18, 1.88s/it, loss=0.294, lr=0.000396]\nSteps: 26%|██▌ | 255/1000 [12:59<23:18, 1.88s/it, loss=1.02, lr=0.000395] \nSteps: 26%|██▌ | 256/1000 [13:01<23:13, 1.87s/it, loss=1.02, lr=0.000395]\nSteps: 26%|██▌ | 256/1000 [13:01<23:13, 1.87s/it, loss=0.376, lr=0.000395]\nSteps: 26%|██▌ | 257/1000 [13:03<23:09, 1.87s/it, loss=0.376, lr=0.000395]\nSteps: 26%|██▌ | 257/1000 [13:03<23:09, 1.87s/it, loss=0.251, lr=0.000395]\nSteps: 26%|██▌ | 258/1000 [13:05<23:06, 1.87s/it, loss=0.251, lr=0.000395]\nSteps: 26%|██▌ | 258/1000 [13:05<23:06, 1.87s/it, loss=0.311, lr=0.000395]\nSteps: 26%|██▌ | 259/1000 [13:07<23:02, 1.87s/it, loss=0.311, lr=0.000395]\nSteps: 26%|██▌ | 259/1000 [13:07<23:02, 1.87s/it, loss=0.36, lr=0.000395] \nSteps: 26%|██▌ | 260/1000 [13:09<22:59, 1.86s/it, loss=0.36, lr=0.000395]\nSteps: 26%|██▌ | 260/1000 [13:09<22:59, 1.86s/it, loss=0.892, lr=0.000394]\nSteps: 26%|██▌ | 261/1000 [13:11<22:56, 1.86s/it, loss=0.892, lr=0.000394]\nSteps: 26%|██▌ | 261/1000 [13:11<22:56, 1.86s/it, loss=1.02, lr=0.000394] \nSteps: 26%|██▌ | 262/1000 [13:13<22:56, 1.87s/it, loss=1.02, lr=0.000394]\nSteps: 26%|██▌ | 262/1000 [13:13<22:56, 1.87s/it, loss=0.481, lr=0.000394]\nSteps: 26%|██▋ | 263/1000 [13:14<22:53, 1.86s/it, loss=0.481, lr=0.000394]\nSteps: 26%|██▋ | 263/1000 [13:14<22:53, 1.86s/it, loss=1.03, lr=0.000394] \nSteps: 26%|██▋ | 264/1000 [13:16<22:50, 1.86s/it, loss=1.03, lr=0.000394]\nSteps: 26%|██▋ | 264/1000 [13:16<22:50, 1.86s/it, loss=0.393, lr=0.000394]\nSteps: 26%|██▋ | 265/1000 [13:18<22:49, 1.86s/it, loss=0.393, lr=0.000394]\nSteps: 26%|██▋ | 265/1000 [13:18<22:49, 1.86s/it, loss=0.546, lr=0.000394]\nSteps: 27%|██▋ | 266/1000 [13:20<22:47, 1.86s/it, loss=0.546, lr=0.000394]\nSteps: 27%|██▋ | 266/1000 [13:20<22:47, 1.86s/it, loss=0.786, lr=0.000393]\nSteps: 27%|██▋ | 267/1000 [13:22<22:45, 1.86s/it, loss=0.786, lr=0.000393]\nSteps: 27%|██▋ | 267/1000 [13:22<22:45, 1.86s/it, loss=0.431, lr=0.000393]\nSteps: 27%|██▋ | 268/1000 [13:24<22:43, 1.86s/it, loss=0.431, lr=0.000393]\nSteps: 27%|██▋ | 268/1000 [13:24<22:43, 1.86s/it, loss=0.815, lr=0.000393]\nSteps: 27%|██▋ | 269/1000 [13:26<22:41, 1.86s/it, loss=0.815, lr=0.000393]\nSteps: 27%|██▋ | 269/1000 [13:26<22:41, 1.86s/it, loss=0.551, lr=0.000393]\nSteps: 27%|██▋ | 270/1000 [13:27<22:40, 1.86s/it, loss=0.551, lr=0.000393]\nSteps: 27%|██▋ | 270/1000 [13:27<22:40, 1.86s/it, loss=0.948, lr=0.000392]\nSteps: 27%|██▋ | 271/1000 [13:35<43:52, 3.61s/it, loss=0.948, lr=0.000392]\nSteps: 27%|██▋ | 271/1000 [13:35<43:52, 3.61s/it, loss=0.387, lr=0.000392]\nSteps: 27%|██▋ | 272/1000 [13:37<37:28, 3.09s/it, loss=0.387, lr=0.000392]\nSteps: 27%|██▋ | 272/1000 [13:37<37:28, 3.09s/it, loss=0.634, lr=0.000392]\nSteps: 27%|██▋ | 273/1000 [13:39<32:58, 2.72s/it, loss=0.634, lr=0.000392]\nSteps: 27%|██▋ | 273/1000 [13:39<32:58, 2.72s/it, loss=0.463, lr=0.000392]\nSteps: 27%|██▋ | 274/1000 [13:41<29:50, 2.47s/it, loss=0.463, lr=0.000392]\nSteps: 27%|██▋ | 274/1000 [13:41<29:50, 2.47s/it, loss=0.27, lr=0.000392] \nSteps: 28%|██▊ | 275/1000 [13:43<27:38, 2.29s/it, loss=0.27, lr=0.000392]\nSteps: 28%|██▊ | 275/1000 [13:43<27:38, 2.29s/it, loss=0.49, lr=0.000391]\nSteps: 28%|██▊ | 276/1000 [13:44<26:05, 2.16s/it, loss=0.49, lr=0.000391]\nSteps: 28%|██▊ | 276/1000 [13:44<26:05, 2.16s/it, loss=0.532, lr=0.000391]\nSteps: 28%|██▊ | 277/1000 [13:46<24:59, 2.07s/it, loss=0.532, lr=0.000391]\nSteps: 28%|██▊ | 277/1000 [13:46<24:59, 2.07s/it, loss=0.567, lr=0.000391]\nSteps: 28%|██▊ | 278/1000 [13:48<24:13, 2.01s/it, loss=0.567, lr=0.000391]\nSteps: 28%|██▊ | 278/1000 [13:48<24:13, 2.01s/it, loss=0.58, lr=0.000391] \nSteps: 28%|██▊ | 279/1000 [13:50<23:40, 1.97s/it, loss=0.58, lr=0.000391]\nSteps: 28%|██▊ | 279/1000 [13:50<23:40, 1.97s/it, loss=0.46, lr=0.00039] \nSteps: 28%|██▊ | 280/1000 [13:52<23:17, 1.94s/it, loss=0.46, lr=0.00039]\nSteps: 28%|██▊ | 280/1000 [13:52<23:17, 1.94s/it, loss=0.31, lr=0.00039]\nSteps: 28%|██▊ | 281/1000 [13:54<23:00, 1.92s/it, loss=0.31, lr=0.00039]\nSteps: 28%|██▊ | 281/1000 [13:54<23:00, 1.92s/it, loss=0.328, lr=0.00039]\nSteps: 28%|██▊ | 282/1000 [13:56<22:47, 1.90s/it, loss=0.328, lr=0.00039]\nSteps: 28%|██▊ | 282/1000 [13:56<22:47, 1.90s/it, loss=0.712, lr=0.00039]\nSteps: 28%|██▊ | 283/1000 [13:58<22:37, 1.89s/it, loss=0.712, lr=0.00039]\nSteps: 28%|██▊ | 283/1000 [13:58<22:37, 1.89s/it, loss=0.335, lr=0.000389]\nSteps: 28%|██▊ | 284/1000 [13:59<22:31, 1.89s/it, loss=0.335, lr=0.000389]\nSteps: 28%|██▊ | 284/1000 [13:59<22:31, 1.89s/it, loss=0.621, lr=0.000389]\nSteps: 28%|██▊ | 285/1000 [14:01<22:25, 1.88s/it, loss=0.621, lr=0.000389]\nSteps: 28%|██▊ | 285/1000 [14:01<22:25, 1.88s/it, loss=0.368, lr=0.000389]\nSteps: 29%|██▊ | 286/1000 [14:03<22:21, 1.88s/it, loss=0.368, lr=0.000389]\nSteps: 29%|██▊ | 286/1000 [14:03<22:21, 1.88s/it, loss=0.709, lr=0.000389]\nSteps: 29%|██▊ | 287/1000 [14:05<22:17, 1.88s/it, loss=0.709, lr=0.000389]\nSteps: 29%|██▊ | 287/1000 [14:05<22:17, 1.88s/it, loss=0.947, lr=0.000388]\nSteps: 29%|██▉ | 288/1000 [14:07<22:13, 1.87s/it, loss=0.947, lr=0.000388]\nSteps: 29%|██▉ | 288/1000 [14:07<22:13, 1.87s/it, loss=0.336, lr=0.000388]\nSteps: 29%|██▉ | 289/1000 [14:09<22:10, 1.87s/it, loss=0.336, lr=0.000388]\nSteps: 29%|██▉ | 289/1000 [14:09<22:10, 1.87s/it, loss=1.03, lr=0.000388] \nSteps: 29%|██▉ | 290/1000 [14:11<22:08, 1.87s/it, loss=1.03, lr=0.000388]\nSteps: 29%|██▉ | 290/1000 [14:11<22:08, 1.87s/it, loss=0.524, lr=0.000388]\nSteps: 29%|██▉ | 291/1000 [14:13<22:06, 1.87s/it, loss=0.524, lr=0.000388]\nSteps: 29%|██▉ | 291/1000 [14:13<22:06, 1.87s/it, loss=0.304, lr=0.000387]\nSteps: 29%|██▉ | 292/1000 [14:14<22:05, 1.87s/it, loss=0.304, lr=0.000387]\nSteps: 29%|██▉ | 292/1000 [14:14<22:05, 1.87s/it, loss=0.303, lr=0.000387]\nSteps: 29%|██▉ | 293/1000 [14:16<22:03, 1.87s/it, loss=0.303, lr=0.000387]\nSteps: 29%|██▉ | 293/1000 [14:16<22:03, 1.87s/it, loss=0.492, lr=0.000387]\nSteps: 29%|██▉ | 294/1000 [14:18<22:00, 1.87s/it, loss=0.492, lr=0.000387]\nSteps: 29%|██▉ | 294/1000 [14:18<22:00, 1.87s/it, loss=0.545, lr=0.000387]\nSteps: 30%|██▉ | 295/1000 [14:20<21:58, 1.87s/it, loss=0.545, lr=0.000387]\nSteps: 30%|██▉ | 295/1000 [14:20<21:58, 1.87s/it, loss=0.984, lr=0.000386]\nSteps: 30%|██▉ | 296/1000 [14:22<21:56, 1.87s/it, loss=0.984, lr=0.000386]\nSteps: 30%|██▉ | 296/1000 [14:22<21:56, 1.87s/it, loss=0.821, lr=0.000386]\nSteps: 30%|██▉ | 297/1000 [14:24<21:55, 1.87s/it, loss=0.821, lr=0.000386]\nSteps: 30%|██▉ | 297/1000 [14:24<21:55, 1.87s/it, loss=0.346, lr=0.000386]\nSteps: 30%|██▉ | 298/1000 [14:26<21:55, 1.87s/it, loss=0.346, lr=0.000386]\nSteps: 30%|██▉ | 298/1000 [14:26<21:55, 1.87s/it, loss=0.297, lr=0.000385]\nSteps: 30%|██▉ | 299/1000 [14:27<21:51, 1.87s/it, loss=0.297, lr=0.000385]\nSteps: 30%|██▉ | 299/1000 [14:27<21:51, 1.87s/it, loss=0.665, lr=0.000385]\nSteps: 30%|███ | 300/1000 [14:29<21:49, 1.87s/it, loss=0.665, lr=0.000385]\nSteps: 30%|███ | 300/1000 [14:29<21:49, 1.87s/it, loss=0.433, lr=0.000385]\nSteps: 30%|███ | 301/1000 [14:37<42:03, 3.61s/it, loss=0.433, lr=0.000385]\nSteps: 30%|███ | 301/1000 [14:37<42:03, 3.61s/it, loss=0.369, lr=0.000384]\nSteps: 30%|███ | 302/1000 [14:39<35:55, 3.09s/it, loss=0.369, lr=0.000384]\nSteps: 30%|███ | 302/1000 [14:39<35:55, 3.09s/it, loss=0.543, lr=0.000384]\nSteps: 30%|███ | 303/1000 [14:41<31:37, 2.72s/it, loss=0.543, lr=0.000384]\nSteps: 30%|███ | 303/1000 [14:41<31:37, 2.72s/it, loss=0.327, lr=0.000384]\nSteps: 30%|███ | 304/1000 [14:43<28:35, 2.46s/it, loss=0.327, lr=0.000384]\nSteps: 30%|███ | 304/1000 [14:43<28:35, 2.46s/it, loss=0.959, lr=0.000384]\nSteps: 30%|███ | 305/1000 [14:44<26:28, 2.29s/it, loss=0.959, lr=0.000384]\nSteps: 30%|███ | 305/1000 [14:44<26:28, 2.29s/it, loss=0.281, lr=0.000383]\nSteps: 31%|███ | 306/1000 [14:46<24:59, 2.16s/it, loss=0.281, lr=0.000383]\nSteps: 31%|███ | 306/1000 [14:46<24:59, 2.16s/it, loss=0.432, lr=0.000383]\nSteps: 31%|███ | 307/1000 [14:48<23:56, 2.07s/it, loss=0.432, lr=0.000383]\nSteps: 31%|███ | 307/1000 [14:48<23:56, 2.07s/it, loss=0.563, lr=0.000383]\nSteps: 31%|███ | 308/1000 [14:50<23:10, 2.01s/it, loss=0.563, lr=0.000383]\nSteps: 31%|███ | 308/1000 [14:50<23:10, 2.01s/it, loss=0.529, lr=0.000382]\nSteps: 31%|███ | 309/1000 [14:52<22:40, 1.97s/it, loss=0.529, lr=0.000382]\nSteps: 31%|███ | 309/1000 [14:52<22:40, 1.97s/it, loss=0.73, lr=0.000382] \nSteps: 31%|███ | 310/1000 [14:54<22:17, 1.94s/it, loss=0.73, lr=0.000382]\nSteps: 31%|███ | 310/1000 [14:54<22:17, 1.94s/it, loss=0.317, lr=0.000382]\nSteps: 31%|███ | 311/1000 [14:56<22:01, 1.92s/it, loss=0.317, lr=0.000382]\nSteps: 31%|███ | 311/1000 [14:56<22:01, 1.92s/it, loss=0.406, lr=0.000381]\nSteps: 31%|███ | 312/1000 [14:58<21:50, 1.90s/it, loss=0.406, lr=0.000381]\nSteps: 31%|███ | 312/1000 [14:58<21:50, 1.90s/it, loss=0.944, lr=0.000381]\nSteps: 31%|███▏ | 313/1000 [14:59<21:41, 1.89s/it, loss=0.944, lr=0.000381]\nSteps: 31%|███▏ | 313/1000 [14:59<21:41, 1.89s/it, loss=1.06, lr=0.000381] \nSteps: 31%|███▏ | 314/1000 [15:01<21:33, 1.89s/it, loss=1.06, lr=0.000381]\nSteps: 31%|███▏ | 314/1000 [15:01<21:33, 1.89s/it, loss=0.557, lr=0.00038]\nSteps: 32%|███▏ | 315/1000 [15:03<21:28, 1.88s/it, loss=0.557, lr=0.00038]\nSteps: 32%|███▏ | 315/1000 [15:03<21:28, 1.88s/it, loss=0.632, lr=0.00038]\nSteps: 32%|███▏ | 316/1000 [15:05<21:24, 1.88s/it, loss=0.632, lr=0.00038]\nSteps: 32%|███▏ | 316/1000 [15:05<21:24, 1.88s/it, loss=0.384, lr=0.00038]\nSteps: 32%|███▏ | 317/1000 [15:07<21:21, 1.88s/it, loss=0.384, lr=0.00038]\nSteps: 32%|███▏ | 317/1000 [15:07<21:21, 1.88s/it, loss=0.725, lr=0.000379]\nSteps: 32%|███▏ | 318/1000 [15:09<21:18, 1.87s/it, loss=0.725, lr=0.000379]\nSteps: 32%|███▏ | 318/1000 [15:09<21:18, 1.87s/it, loss=1.03, lr=0.000379] \nSteps: 32%|███▏ | 319/1000 [15:11<21:16, 1.87s/it, loss=1.03, lr=0.000379]\nSteps: 32%|███▏ | 319/1000 [15:11<21:16, 1.87s/it, loss=0.48, lr=0.000379]\nSteps: 32%|███▏ | 320/1000 [15:13<21:14, 1.87s/it, loss=0.48, lr=0.000379]\nSteps: 32%|███▏ | 320/1000 [15:13<21:14, 1.87s/it, loss=0.702, lr=0.000378]\nSteps: 32%|███▏ | 321/1000 [15:14<21:11, 1.87s/it, loss=0.702, lr=0.000378]\nSteps: 32%|███▏ | 321/1000 [15:14<21:11, 1.87s/it, loss=0.453, lr=0.000378]\nSteps: 32%|███▏ | 322/1000 [15:16<21:10, 1.87s/it, loss=0.453, lr=0.000378]\nSteps: 32%|███▏ | 322/1000 [15:16<21:10, 1.87s/it, loss=0.384, lr=0.000377]\nSteps: 32%|███▏ | 323/1000 [15:18<21:06, 1.87s/it, loss=0.384, lr=0.000377]\nSteps: 32%|███▏ | 323/1000 [15:18<21:06, 1.87s/it, loss=0.349, lr=0.000377]\nSteps: 32%|███▏ | 324/1000 [15:20<21:02, 1.87s/it, loss=0.349, lr=0.000377]\nSteps: 32%|███▏ | 324/1000 [15:20<21:02, 1.87s/it, loss=0.612, lr=0.000377]\nSteps: 32%|███▎ | 325/1000 [15:22<21:00, 1.87s/it, loss=0.612, lr=0.000377]\nSteps: 32%|███▎ | 325/1000 [15:22<21:00, 1.87s/it, loss=0.6, lr=0.000376] \nSteps: 33%|███▎ | 326/1000 [15:24<20:59, 1.87s/it, loss=0.6, lr=0.000376]\nSteps: 33%|███▎ | 326/1000 [15:24<20:59, 1.87s/it, loss=0.39, lr=0.000376]\nSteps: 33%|███▎ | 327/1000 [15:26<20:56, 1.87s/it, loss=0.39, lr=0.000376]\nSteps: 33%|███▎ | 327/1000 [15:26<20:56, 1.87s/it, loss=0.709, lr=0.000376]\nSteps: 33%|███▎ | 328/1000 [15:27<20:55, 1.87s/it, loss=0.709, lr=0.000376]\nSteps: 33%|███▎ | 328/1000 [15:27<20:55, 1.87s/it, loss=0.313, lr=0.000375]\nSteps: 33%|███▎ | 329/1000 [15:29<20:51, 1.86s/it, loss=0.313, lr=0.000375]\nSteps: 33%|███▎ | 329/1000 [15:29<20:51, 1.86s/it, loss=0.695, lr=0.000375]\nSteps: 33%|███▎ | 330/1000 [15:31<20:49, 1.86s/it, loss=0.695, lr=0.000375]\nSteps: 33%|███▎ | 330/1000 [15:31<20:49, 1.86s/it, loss=0.548, lr=0.000374]\nSteps: 33%|███▎ | 331/1000 [15:39<40:19, 3.62s/it, loss=0.548, lr=0.000374]\nSteps: 33%|███▎ | 331/1000 [15:39<40:19, 3.62s/it, loss=0.915, lr=0.000374]\nSteps: 33%|███▎ | 332/1000 [15:41<34:25, 3.09s/it, loss=0.915, lr=0.000374]\nSteps: 33%|███▎ | 332/1000 [15:41<34:25, 3.09s/it, loss=0.617, lr=0.000374]\nSteps: 33%|███▎ | 333/1000 [15:43<30:18, 2.73s/it, loss=0.617, lr=0.000374]\nSteps: 33%|███▎ | 333/1000 [15:43<30:18, 2.73s/it, loss=0.328, lr=0.000373]\nSteps: 33%|███▎ | 334/1000 [15:45<27:25, 2.47s/it, loss=0.328, lr=0.000373]\nSteps: 33%|███▎ | 334/1000 [15:45<27:25, 2.47s/it, loss=0.745, lr=0.000373]\nSteps: 34%|███▎ | 335/1000 [15:46<25:22, 2.29s/it, loss=0.745, lr=0.000373]\nSteps: 34%|███▎ | 335/1000 [15:46<25:22, 2.29s/it, loss=0.752, lr=0.000373]\nSteps: 34%|███▎ | 336/1000 [15:48<23:56, 2.16s/it, loss=0.752, lr=0.000373]\nSteps: 34%|███▎ | 336/1000 [15:48<23:56, 2.16s/it, loss=0.307, lr=0.000372]\nSteps: 34%|███▎ | 337/1000 [15:50<22:55, 2.08s/it, loss=0.307, lr=0.000372]\nSteps: 34%|███▎ | 337/1000 [15:50<22:55, 2.08s/it, loss=0.995, lr=0.000372]\nSteps: 34%|███▍ | 338/1000 [15:52<22:13, 2.01s/it, loss=0.995, lr=0.000372]\nSteps: 34%|███▍ | 338/1000 [15:52<22:13, 2.01s/it, loss=0.637, lr=0.000371]\nSteps: 34%|███▍ | 339/1000 [15:54<21:42, 1.97s/it, loss=0.637, lr=0.000371]\nSteps: 34%|███▍ | 339/1000 [15:54<21:42, 1.97s/it, loss=1.02, lr=0.000371] \nSteps: 34%|███▍ | 340/1000 [15:56<21:20, 1.94s/it, loss=1.02, lr=0.000371]\nSteps: 34%|███▍ | 340/1000 [15:56<21:20, 1.94s/it, loss=0.464, lr=0.000371]\nSteps: 34%|███▍ | 341/1000 [15:58<21:05, 1.92s/it, loss=0.464, lr=0.000371]\nSteps: 34%|███▍ | 341/1000 [15:58<21:05, 1.92s/it, loss=0.321, lr=0.00037] \nSteps: 34%|███▍ | 342/1000 [15:59<20:53, 1.91s/it, loss=0.321, lr=0.00037]\nSteps: 34%|███▍ | 342/1000 [15:59<20:53, 1.91s/it, loss=0.649, lr=0.00037]\nSteps: 34%|███▍ | 343/1000 [16:01<20:45, 1.90s/it, loss=0.649, lr=0.00037]\nSteps: 34%|███▍ | 343/1000 [16:01<20:45, 1.90s/it, loss=0.569, lr=0.000369]\nSteps: 34%|███▍ | 344/1000 [16:03<20:38, 1.89s/it, loss=0.569, lr=0.000369]\nSteps: 34%|███▍ | 344/1000 [16:03<20:38, 1.89s/it, loss=0.286, lr=0.000369]\nSteps: 34%|███▍ | 345/1000 [16:05<20:32, 1.88s/it, loss=0.286, lr=0.000369]\nSteps: 34%|███▍ | 345/1000 [16:05<20:32, 1.88s/it, loss=0.714, lr=0.000368]\nSteps: 35%|███▍ | 346/1000 [16:07<20:27, 1.88s/it, loss=0.714, lr=0.000368]\nSteps: 35%|███▍ | 346/1000 [16:07<20:27, 1.88s/it, loss=0.395, lr=0.000368]\nSteps: 35%|███▍ | 347/1000 [16:09<20:23, 1.87s/it, loss=0.395, lr=0.000368]\nSteps: 35%|███▍ | 347/1000 [16:09<20:23, 1.87s/it, loss=0.835, lr=0.000368]\nSteps: 35%|███▍ | 348/1000 [16:11<20:19, 1.87s/it, loss=0.835, lr=0.000368]\nSteps: 35%|███▍ | 348/1000 [16:11<20:19, 1.87s/it, loss=0.386, lr=0.000367]\nSteps: 35%|███▍ | 349/1000 [16:13<20:17, 1.87s/it, loss=0.386, lr=0.000367]\nSteps: 35%|███▍ | 349/1000 [16:13<20:17, 1.87s/it, loss=0.482, lr=0.000367]\nSteps: 35%|███▌ | 350/1000 [16:14<20:14, 1.87s/it, loss=0.482, lr=0.000367]\nSteps: 35%|███▌ | 350/1000 [16:14<20:14, 1.87s/it, loss=1.06, lr=0.000366] \nSteps: 35%|███▌ | 351/1000 [16:16<20:11, 1.87s/it, loss=1.06, lr=0.000366]\nSteps: 35%|███▌ | 351/1000 [16:16<20:11, 1.87s/it, loss=0.54, lr=0.000366]\nSteps: 35%|███▌ | 352/1000 [16:18<20:10, 1.87s/it, loss=0.54, lr=0.000366]\nSteps: 35%|███▌ | 352/1000 [16:18<20:10, 1.87s/it, loss=1.04, lr=0.000365]\nSteps: 35%|███▌ | 353/1000 [16:20<20:08, 1.87s/it, loss=1.04, lr=0.000365]\nSteps: 35%|███▌ | 353/1000 [16:20<20:08, 1.87s/it, loss=0.389, lr=0.000365]\nSteps: 35%|███▌ | 354/1000 [16:22<20:05, 1.87s/it, loss=0.389, lr=0.000365]\nSteps: 35%|███▌ | 354/1000 [16:22<20:05, 1.87s/it, loss=0.695, lr=0.000365]\nSteps: 36%|███▌ | 355/1000 [16:24<20:03, 1.87s/it, loss=0.695, lr=0.000365]\nSteps: 36%|███▌ | 355/1000 [16:24<20:03, 1.87s/it, loss=0.45, lr=0.000364] \nSteps: 36%|███▌ | 356/1000 [16:26<20:01, 1.87s/it, loss=0.45, lr=0.000364]\nSteps: 36%|███▌ | 356/1000 [16:26<20:01, 1.87s/it, loss=0.875, lr=0.000364]\nSteps: 36%|███▌ | 357/1000 [16:27<20:00, 1.87s/it, loss=0.875, lr=0.000364]\nSteps: 36%|███▌ | 357/1000 [16:27<20:00, 1.87s/it, loss=0.711, lr=0.000363]\nSteps: 36%|███▌ | 358/1000 [16:29<19:58, 1.87s/it, loss=0.711, lr=0.000363]\nSteps: 36%|███▌ | 358/1000 [16:29<19:58, 1.87s/it, loss=0.635, lr=0.000363]\nSteps: 36%|███▌ | 359/1000 [16:31<19:54, 1.86s/it, loss=0.635, lr=0.000363]\nSteps: 36%|███▌ | 359/1000 [16:31<19:54, 1.86s/it, loss=0.983, lr=0.000362]\nSteps: 36%|███▌ | 360/1000 [16:33<19:53, 1.86s/it, loss=0.983, lr=0.000362]\nSteps: 36%|███▌ | 360/1000 [16:33<19:53, 1.86s/it, loss=0.776, lr=0.000362]\nSteps: 36%|███▌ | 361/1000 [16:41<38:48, 3.64s/it, loss=0.776, lr=0.000362]\nSteps: 36%|███▌ | 361/1000 [16:41<38:48, 3.64s/it, loss=0.335, lr=0.000361]\nSteps: 36%|███▌ | 362/1000 [16:43<33:05, 3.11s/it, loss=0.335, lr=0.000361]\nSteps: 36%|███▌ | 362/1000 [16:43<33:05, 3.11s/it, loss=0.319, lr=0.000361]\nSteps: 36%|███▋ | 363/1000 [16:45<29:05, 2.74s/it, loss=0.319, lr=0.000361]\nSteps: 36%|███▋ | 363/1000 [16:45<29:05, 2.74s/it, loss=0.497, lr=0.00036] \nSteps: 36%|███▋ | 364/1000 [16:46<26:16, 2.48s/it, loss=0.497, lr=0.00036]\nSteps: 36%|███▋ | 364/1000 [16:46<26:16, 2.48s/it, loss=0.38, lr=0.00036] \nSteps: 36%|███▋ | 365/1000 [16:48<24:18, 2.30s/it, loss=0.38, lr=0.00036]\nSteps: 36%|███▋ | 365/1000 [16:48<24:18, 2.30s/it, loss=0.281, lr=0.000359]\nSteps: 37%|███▋ | 366/1000 [16:50<22:54, 2.17s/it, loss=0.281, lr=0.000359]\nSteps: 37%|███▋ | 366/1000 [16:50<22:54, 2.17s/it, loss=0.668, lr=0.000359]\nSteps: 37%|███▋ | 367/1000 [16:52<21:56, 2.08s/it, loss=0.668, lr=0.000359]\nSteps: 37%|███▋ | 367/1000 [16:52<21:56, 2.08s/it, loss=0.576, lr=0.000359]\nSteps: 37%|███▋ | 368/1000 [16:54<21:13, 2.02s/it, loss=0.576, lr=0.000359]\nSteps: 37%|███▋ | 368/1000 [16:54<21:13, 2.02s/it, loss=0.352, lr=0.000358]\nSteps: 37%|███▋ | 369/1000 [16:56<20:45, 1.97s/it, loss=0.352, lr=0.000358]\nSteps: 37%|███▋ | 369/1000 [16:56<20:45, 1.97s/it, loss=0.295, lr=0.000358]\nSteps: 37%|███▋ | 370/1000 [16:58<20:23, 1.94s/it, loss=0.295, lr=0.000358]\nSteps: 37%|███▋ | 370/1000 [16:58<20:23, 1.94s/it, loss=0.324, lr=0.000357]\nSteps: 37%|███▋ | 371/1000 [17:00<20:07, 1.92s/it, loss=0.324, lr=0.000357]\nSteps: 37%|███▋ | 371/1000 [17:00<20:07, 1.92s/it, loss=0.819, lr=0.000357]\nSteps: 37%|███▋ | 372/1000 [17:01<19:56, 1.90s/it, loss=0.819, lr=0.000357]\nSteps: 37%|███▋ | 372/1000 [17:01<19:56, 1.90s/it, loss=0.616, lr=0.000356]\nSteps: 37%|███▋ | 373/1000 [17:03<19:47, 1.89s/it, loss=0.616, lr=0.000356]\nSteps: 37%|███▋ | 373/1000 [17:03<19:47, 1.89s/it, loss=0.496, lr=0.000356]\nSteps: 37%|███▋ | 374/1000 [17:05<19:42, 1.89s/it, loss=0.496, lr=0.000356]\nSteps: 37%|███▋ | 374/1000 [17:05<19:42, 1.89s/it, loss=1.04, lr=0.000355] \nSteps: 38%|███▊ | 375/1000 [17:07<19:35, 1.88s/it, loss=1.04, lr=0.000355]\nSteps: 38%|███▊ | 375/1000 [17:07<19:35, 1.88s/it, loss=0.9, lr=0.000355] \nSteps: 38%|███▊ | 376/1000 [17:09<19:31, 1.88s/it, loss=0.9, lr=0.000355]\nSteps: 38%|███▊ | 376/1000 [17:09<19:31, 1.88s/it, loss=0.34, lr=0.000354]\nSteps: 38%|███▊ | 377/1000 [17:11<19:26, 1.87s/it, loss=0.34, lr=0.000354]\nSteps: 38%|███▊ | 377/1000 [17:11<19:26, 1.87s/it, loss=0.779, lr=0.000354]\nSteps: 38%|███▊ | 378/1000 [17:13<19:23, 1.87s/it, loss=0.779, lr=0.000354]\nSteps: 38%|███▊ | 378/1000 [17:13<19:23, 1.87s/it, loss=0.889, lr=0.000353]\nSteps: 38%|███▊ | 379/1000 [17:15<19:21, 1.87s/it, loss=0.889, lr=0.000353]\nSteps: 38%|███▊ | 379/1000 [17:15<19:21, 1.87s/it, loss=0.66, lr=0.000353] \nSteps: 38%|███▊ | 380/1000 [17:16<19:18, 1.87s/it, loss=0.66, lr=0.000353]\nSteps: 38%|███▊ | 380/1000 [17:16<19:18, 1.87s/it, loss=1.02, lr=0.000352]\nSteps: 38%|███▊ | 381/1000 [17:18<19:16, 1.87s/it, loss=1.02, lr=0.000352]\nSteps: 38%|███▊ | 381/1000 [17:18<19:16, 1.87s/it, loss=0.313, lr=0.000352]\nSteps: 38%|███▊ | 382/1000 [17:20<19:14, 1.87s/it, loss=0.313, lr=0.000352]\nSteps: 38%|███▊ | 382/1000 [17:20<19:14, 1.87s/it, loss=0.447, lr=0.000351]\nSteps: 38%|███▊ | 383/1000 [17:22<19:12, 1.87s/it, loss=0.447, lr=0.000351]\nSteps: 38%|███▊ | 383/1000 [17:22<19:12, 1.87s/it, loss=0.36, lr=0.000351] \nSteps: 38%|███▊ | 384/1000 [17:24<19:10, 1.87s/it, loss=0.36, lr=0.000351]\nSteps: 38%|███▊ | 384/1000 [17:24<19:10, 1.87s/it, loss=0.428, lr=0.00035]\nSteps: 38%|███▊ | 385/1000 [17:26<19:08, 1.87s/it, loss=0.428, lr=0.00035]\nSteps: 38%|███▊ | 385/1000 [17:26<19:08, 1.87s/it, loss=0.344, lr=0.00035]\nSteps: 39%|███▊ | 386/1000 [17:28<19:06, 1.87s/it, loss=0.344, lr=0.00035]\nSteps: 39%|███▊ | 386/1000 [17:28<19:06, 1.87s/it, loss=0.449, lr=0.000349]\nSteps: 39%|███▊ | 387/1000 [17:29<19:03, 1.87s/it, loss=0.449, lr=0.000349]\nSteps: 39%|███▊ | 387/1000 [17:29<19:03, 1.87s/it, loss=0.58, lr=0.000348] \nSteps: 39%|███▉ | 388/1000 [17:31<19:02, 1.87s/it, loss=0.58, lr=0.000348]\nSteps: 39%|███▉ | 388/1000 [17:31<19:02, 1.87s/it, loss=0.29, lr=0.000348]\nSteps: 39%|███▉ | 389/1000 [17:33<19:00, 1.87s/it, loss=0.29, lr=0.000348]\nSteps: 39%|███▉ | 389/1000 [17:33<19:00, 1.87s/it, loss=0.411, lr=0.000347]\nSteps: 39%|███▉ | 390/1000 [17:35<18:58, 1.87s/it, loss=0.411, lr=0.000347]\nSteps: 39%|███▉ | 390/1000 [17:35<18:58, 1.87s/it, loss=0.536, lr=0.000347]\nSteps: 39%|███▉ | 391/1000 [17:43<36:41, 3.61s/it, loss=0.536, lr=0.000347]\nSteps: 39%|███▉ | 391/1000 [17:43<36:41, 3.61s/it, loss=0.541, lr=0.000346]\nSteps: 39%|███▉ | 392/1000 [17:45<31:18, 3.09s/it, loss=0.541, lr=0.000346]\nSteps: 39%|███▉ | 392/1000 [17:45<31:18, 3.09s/it, loss=0.529, lr=0.000346]\nSteps: 39%|███▉ | 393/1000 [17:46<27:33, 2.72s/it, loss=0.529, lr=0.000346]\nSteps: 39%|███▉ | 393/1000 [17:46<27:33, 2.72s/it, loss=0.554, lr=0.000345]\nSteps: 39%|███▉ | 394/1000 [17:48<24:56, 2.47s/it, loss=0.554, lr=0.000345]\nSteps: 39%|███▉ | 394/1000 [17:48<24:56, 2.47s/it, loss=1.02, lr=0.000345] \nSteps: 40%|███▉ | 395/1000 [17:50<23:04, 2.29s/it, loss=1.02, lr=0.000345]\nSteps: 40%|███▉ | 395/1000 [17:50<23:04, 2.29s/it, loss=0.525, lr=0.000344]\nSteps: 40%|███▉ | 396/1000 [17:52<21:46, 2.16s/it, loss=0.525, lr=0.000344]\nSteps: 40%|███▉ | 396/1000 [17:52<21:46, 2.16s/it, loss=0.454, lr=0.000344]\nSteps: 40%|███▉ | 397/1000 [17:54<20:52, 2.08s/it, loss=0.454, lr=0.000344]\nSteps: 40%|███▉ | 397/1000 [17:54<20:52, 2.08s/it, loss=0.691, lr=0.000343]\nSteps: 40%|███▉ | 398/1000 [17:56<20:13, 2.02s/it, loss=0.691, lr=0.000343]\nSteps: 40%|███▉ | 398/1000 [17:56<20:13, 2.02s/it, loss=0.404, lr=0.000343]\nSteps: 40%|███▉ | 399/1000 [17:58<19:45, 1.97s/it, loss=0.404, lr=0.000343]\nSteps: 40%|███▉ | 399/1000 [17:58<19:45, 1.97s/it, loss=0.413, lr=0.000342]\nSteps: 40%|████ | 400/1000 [18:00<19:25, 1.94s/it, loss=0.413, lr=0.000342]\nSteps: 40%|████ | 400/1000 [18:00<19:25, 1.94s/it, loss=0.82, lr=0.000341] \nSteps: 40%|████ | 401/1000 [18:01<19:11, 1.92s/it, loss=0.82, lr=0.000341]\nSteps: 40%|████ | 401/1000 [18:01<19:11, 1.92s/it, loss=0.883, lr=0.000341]\nSteps: 40%|████ | 402/1000 [18:03<19:00, 1.91s/it, loss=0.883, lr=0.000341]\nSteps: 40%|████ | 402/1000 [18:03<19:00, 1.91s/it, loss=0.634, lr=0.00034] \nSteps: 40%|████ | 403/1000 [18:05<18:53, 1.90s/it, loss=0.634, lr=0.00034]\nSteps: 40%|████ | 403/1000 [18:05<18:53, 1.90s/it, loss=1.03, lr=0.00034] \nSteps: 40%|████ | 404/1000 [18:07<18:46, 1.89s/it, loss=1.03, lr=0.00034]\nSteps: 40%|████ | 404/1000 [18:07<18:46, 1.89s/it, loss=0.291, lr=0.000339]\nSteps: 40%|████ | 405/1000 [18:09<18:41, 1.89s/it, loss=0.291, lr=0.000339]\nSteps: 40%|████ | 405/1000 [18:09<18:41, 1.89s/it, loss=0.596, lr=0.000339]\nSteps: 41%|████ | 406/1000 [18:11<18:37, 1.88s/it, loss=0.596, lr=0.000339]\nSteps: 41%|████ | 406/1000 [18:11<18:37, 1.88s/it, loss=1.03, lr=0.000338] \nSteps: 41%|████ | 407/1000 [18:13<18:33, 1.88s/it, loss=1.03, lr=0.000338]\nSteps: 41%|████ | 407/1000 [18:13<18:33, 1.88s/it, loss=0.419, lr=0.000337]\nSteps: 41%|████ | 408/1000 [18:15<18:30, 1.88s/it, loss=0.419, lr=0.000337]\nSteps: 41%|████ | 408/1000 [18:15<18:30, 1.88s/it, loss=0.664, lr=0.000337]\nSteps: 41%|████ | 409/1000 [18:16<18:27, 1.87s/it, loss=0.664, lr=0.000337]\nSteps: 41%|████ | 409/1000 [18:16<18:27, 1.87s/it, loss=0.341, lr=0.000336]\nSteps: 41%|████ | 410/1000 [18:18<18:25, 1.87s/it, loss=0.341, lr=0.000336]\nSteps: 41%|████ | 410/1000 [18:18<18:25, 1.87s/it, loss=0.517, lr=0.000336]\nSteps: 41%|████ | 411/1000 [18:20<18:23, 1.87s/it, loss=0.517, lr=0.000336]\nSteps: 41%|████ | 411/1000 [18:20<18:23, 1.87s/it, loss=0.818, lr=0.000335]\nSteps: 41%|████ | 412/1000 [18:22<18:21, 1.87s/it, loss=0.818, lr=0.000335]\nSteps: 41%|████ | 412/1000 [18:22<18:21, 1.87s/it, loss=0.305, lr=0.000335]\nSteps: 41%|████▏ | 413/1000 [18:24<18:19, 1.87s/it, loss=0.305, lr=0.000335]\nSteps: 41%|████▏ | 413/1000 [18:24<18:19, 1.87s/it, loss=0.62, lr=0.000334] \nSteps: 41%|████▏ | 414/1000 [18:26<18:16, 1.87s/it, loss=0.62, lr=0.000334]\nSteps: 41%|████▏ | 414/1000 [18:26<18:16, 1.87s/it, loss=0.43, lr=0.000333]\nSteps: 42%|████▏ | 415/1000 [18:28<18:14, 1.87s/it, loss=0.43, lr=0.000333]\nSteps: 42%|████▏ | 415/1000 [18:28<18:14, 1.87s/it, loss=0.332, lr=0.000333]\nSteps: 42%|████▏ | 416/1000 [18:30<18:13, 1.87s/it, loss=0.332, lr=0.000333]\nSteps: 42%|████▏ | 416/1000 [18:30<18:13, 1.87s/it, loss=0.773, lr=0.000332]\nSteps: 42%|████▏ | 417/1000 [18:31<18:11, 1.87s/it, loss=0.773, lr=0.000332]\nSteps: 42%|████▏ | 417/1000 [18:31<18:11, 1.87s/it, loss=0.324, lr=0.000332]\nSteps: 42%|████▏ | 418/1000 [18:33<18:09, 1.87s/it, loss=0.324, lr=0.000332]\nSteps: 42%|████▏ | 418/1000 [18:33<18:09, 1.87s/it, loss=0.291, lr=0.000331]\nSteps: 42%|████▏ | 419/1000 [18:35<18:07, 1.87s/it, loss=0.291, lr=0.000331]\nSteps: 42%|████▏ | 419/1000 [18:35<18:07, 1.87s/it, loss=0.362, lr=0.00033] \nSteps: 42%|████▏ | 420/1000 [18:37<18:06, 1.87s/it, loss=0.362, lr=0.00033]\nSteps: 42%|████▏ | 420/1000 [18:37<18:06, 1.87s/it, loss=0.663, lr=0.00033]\nSteps: 42%|████▏ | 421/1000 [18:45<35:04, 3.64s/it, loss=0.663, lr=0.00033]\nSteps: 42%|████▏ | 421/1000 [18:45<35:04, 3.64s/it, loss=0.36, lr=0.000329]\nSteps: 42%|████▏ | 422/1000 [18:47<29:54, 3.11s/it, loss=0.36, lr=0.000329]\nSteps: 42%|████▏ | 422/1000 [18:47<29:54, 3.11s/it, loss=0.394, lr=0.000329]\nSteps: 42%|████▏ | 423/1000 [18:49<26:18, 2.74s/it, loss=0.394, lr=0.000329]\nSteps: 42%|████▏ | 423/1000 [18:49<26:18, 2.74s/it, loss=0.761, lr=0.000328]\nSteps: 42%|████▏ | 424/1000 [18:50<23:45, 2.48s/it, loss=0.761, lr=0.000328]\nSteps: 42%|████▏ | 424/1000 [18:50<23:45, 2.48s/it, loss=0.279, lr=0.000327]\nSteps: 42%|████▎ | 425/1000 [18:52<21:58, 2.29s/it, loss=0.279, lr=0.000327]\nSteps: 42%|████▎ | 425/1000 [18:52<21:58, 2.29s/it, loss=0.701, lr=0.000327]\nSteps: 43%|████▎ | 426/1000 [18:54<20:43, 2.17s/it, loss=0.701, lr=0.000327]\nSteps: 43%|████▎ | 426/1000 [18:54<20:43, 2.17s/it, loss=0.773, lr=0.000326]\nSteps: 43%|████▎ | 427/1000 [18:56<19:50, 2.08s/it, loss=0.773, lr=0.000326]\nSteps: 43%|████▎ | 427/1000 [18:56<19:50, 2.08s/it, loss=0.868, lr=0.000326]\nSteps: 43%|████▎ | 428/1000 [18:58<19:12, 2.01s/it, loss=0.868, lr=0.000326]\nSteps: 43%|████▎ | 428/1000 [18:58<19:12, 2.01s/it, loss=0.979, lr=0.000325]\nSteps: 43%|████▎ | 429/1000 [19:00<18:45, 1.97s/it, loss=0.979, lr=0.000325]\nSteps: 43%|████▎ | 429/1000 [19:00<18:45, 1.97s/it, loss=0.295, lr=0.000324]\nSteps: 43%|████▎ | 430/1000 [19:02<18:26, 1.94s/it, loss=0.295, lr=0.000324]\nSteps: 43%|████▎ | 430/1000 [19:02<18:26, 1.94s/it, loss=0.541, lr=0.000324]\nSteps: 43%|████▎ | 431/1000 [19:03<18:12, 1.92s/it, loss=0.541, lr=0.000324]\nSteps: 43%|████▎ | 431/1000 [19:03<18:12, 1.92s/it, loss=0.57, lr=0.000323] \nSteps: 43%|████▎ | 432/1000 [19:05<18:02, 1.91s/it, loss=0.57, lr=0.000323]\nSteps: 43%|████▎ | 432/1000 [19:05<18:02, 1.91s/it, loss=0.794, lr=0.000323]\nSteps: 43%|████▎ | 433/1000 [19:07<17:53, 1.89s/it, loss=0.794, lr=0.000323]\nSteps: 43%|████▎ | 433/1000 [19:07<17:53, 1.89s/it, loss=0.327, lr=0.000322]\nSteps: 43%|████▎ | 434/1000 [19:09<17:48, 1.89s/it, loss=0.327, lr=0.000322]\nSteps: 43%|████▎ | 434/1000 [19:09<17:48, 1.89s/it, loss=0.489, lr=0.000321]\nSteps: 44%|████▎ | 435/1000 [19:11<17:44, 1.88s/it, loss=0.489, lr=0.000321]\nSteps: 44%|████▎ | 435/1000 [19:11<17:44, 1.88s/it, loss=0.361, lr=0.000321]\nSteps: 44%|████▎ | 436/1000 [19:13<17:40, 1.88s/it, loss=0.361, lr=0.000321]\nSteps: 44%|████▎ | 436/1000 [19:13<17:40, 1.88s/it, loss=0.355, lr=0.00032] \nSteps: 44%|████▎ | 437/1000 [19:15<17:37, 1.88s/it, loss=0.355, lr=0.00032]\nSteps: 44%|████▎ | 437/1000 [19:15<17:37, 1.88s/it, loss=0.725, lr=0.000319]\nSteps: 44%|████▍ | 438/1000 [19:17<17:34, 1.88s/it, loss=0.725, lr=0.000319]\nSteps: 44%|████▍ | 438/1000 [19:17<17:34, 1.88s/it, loss=0.472, lr=0.000319]\nSteps: 44%|████▍ | 439/1000 [19:18<17:32, 1.88s/it, loss=0.472, lr=0.000319]\nSteps: 44%|████▍ | 439/1000 [19:18<17:32, 1.88s/it, loss=0.376, lr=0.000318]\nSteps: 44%|████▍ | 440/1000 [19:20<17:29, 1.87s/it, loss=0.376, lr=0.000318]\nSteps: 44%|████▍ | 440/1000 [19:20<17:29, 1.87s/it, loss=0.329, lr=0.000318]\nSteps: 44%|████▍ | 441/1000 [19:22<17:27, 1.87s/it, loss=0.329, lr=0.000318]\nSteps: 44%|████▍ | 441/1000 [19:22<17:27, 1.87s/it, loss=0.439, lr=0.000317]\nSteps: 44%|████▍ | 442/1000 [19:24<17:26, 1.87s/it, loss=0.439, lr=0.000317]\nSteps: 44%|████▍ | 442/1000 [19:24<17:26, 1.87s/it, loss=0.386, lr=0.000316]\nSteps: 44%|████▍ | 443/1000 [19:26<17:23, 1.87s/it, loss=0.386, lr=0.000316]\nSteps: 44%|████▍ | 443/1000 [19:26<17:23, 1.87s/it, loss=0.462, lr=0.000316]\nSteps: 44%|████▍ | 444/1000 [19:28<17:21, 1.87s/it, loss=0.462, lr=0.000316]\nSteps: 44%|████▍ | 444/1000 [19:28<17:21, 1.87s/it, loss=0.255, lr=0.000315]\nSteps: 44%|████▍ | 445/1000 [19:30<17:19, 1.87s/it, loss=0.255, lr=0.000315]\nSteps: 44%|████▍ | 445/1000 [19:30<17:19, 1.87s/it, loss=0.503, lr=0.000314]\nSteps: 45%|████▍ | 446/1000 [19:32<17:17, 1.87s/it, loss=0.503, lr=0.000314]\nSteps: 45%|████▍ | 446/1000 [19:32<17:17, 1.87s/it, loss=0.824, lr=0.000314]\nSteps: 45%|████▍ | 447/1000 [19:33<17:15, 1.87s/it, loss=0.824, lr=0.000314]\nSteps: 45%|████▍ | 447/1000 [19:33<17:15, 1.87s/it, loss=0.623, lr=0.000313]\nSteps: 45%|████▍ | 448/1000 [19:35<17:13, 1.87s/it, loss=0.623, lr=0.000313]\nSteps: 45%|████▍ | 448/1000 [19:35<17:13, 1.87s/it, loss=0.3, lr=0.000312] \nSteps: 45%|████▍ | 449/1000 [19:37<17:12, 1.87s/it, loss=0.3, lr=0.000312]\nSteps: 45%|████▍ | 449/1000 [19:37<17:12, 1.87s/it, loss=0.368, lr=0.000312]\nSteps: 45%|████▌ | 450/1000 [19:39<17:09, 1.87s/it, loss=0.368, lr=0.000312]\nSteps: 45%|████▌ | 450/1000 [19:39<17:09, 1.87s/it, loss=0.449, lr=0.000311]\nSteps: 45%|████▌ | 451/1000 [19:47<33:20, 3.64s/it, loss=0.449, lr=0.000311]\nSteps: 45%|████▌ | 451/1000 [19:47<33:20, 3.64s/it, loss=0.314, lr=0.00031] \nSteps: 45%|████▌ | 452/1000 [19:49<28:25, 3.11s/it, loss=0.314, lr=0.00031]\nSteps: 45%|████▌ | 452/1000 [19:49<28:25, 3.11s/it, loss=0.31, lr=0.00031] \nSteps: 45%|████▌ | 453/1000 [19:51<24:58, 2.74s/it, loss=0.31, lr=0.00031]\nSteps: 45%|████▌ | 453/1000 [19:51<24:58, 2.74s/it, loss=0.85, lr=0.000309]\nSteps: 45%|████▌ | 454/1000 [19:52<22:33, 2.48s/it, loss=0.85, lr=0.000309]\nSteps: 45%|████▌ | 454/1000 [19:52<22:33, 2.48s/it, loss=0.582, lr=0.000308]\nSteps: 46%|████▌ | 455/1000 [19:54<20:51, 2.30s/it, loss=0.582, lr=0.000308]\nSteps: 46%|████▌ | 455/1000 [19:54<20:51, 2.30s/it, loss=0.394, lr=0.000308]\nSteps: 46%|████▌ | 456/1000 [19:56<19:39, 2.17s/it, loss=0.394, lr=0.000308]\nSteps: 46%|████▌ | 456/1000 [19:56<19:39, 2.17s/it, loss=0.563, lr=0.000307]\nSteps: 46%|████▌ | 457/1000 [19:58<18:48, 2.08s/it, loss=0.563, lr=0.000307]\nSteps: 46%|████▌ | 457/1000 [19:58<18:48, 2.08s/it, loss=0.714, lr=0.000307]\nSteps: 46%|████▌ | 458/1000 [20:00<18:13, 2.02s/it, loss=0.714, lr=0.000307]\nSteps: 46%|████▌ | 458/1000 [20:00<18:13, 2.02s/it, loss=0.468, lr=0.000306]\nSteps: 46%|████▌ | 459/1000 [20:02<17:47, 1.97s/it, loss=0.468, lr=0.000306]\nSteps: 46%|████▌ | 459/1000 [20:02<17:47, 1.97s/it, loss=0.883, lr=0.000305]\nSteps: 46%|████▌ | 460/1000 [20:04<17:28, 1.94s/it, loss=0.883, lr=0.000305]\nSteps: 46%|████▌ | 460/1000 [20:04<17:28, 1.94s/it, loss=0.721, lr=0.000304]\nSteps: 46%|████▌ | 461/1000 [20:06<17:14, 1.92s/it, loss=0.721, lr=0.000304]\nSteps: 46%|████▌ | 461/1000 [20:06<17:14, 1.92s/it, loss=0.321, lr=0.000304]\nSteps: 46%|████▌ | 462/1000 [20:07<17:05, 1.91s/it, loss=0.321, lr=0.000304]\nSteps: 46%|████▌ | 462/1000 [20:07<17:05, 1.91s/it, loss=0.527, lr=0.000303]\nSteps: 46%|████▋ | 463/1000 [20:09<16:57, 1.90s/it, loss=0.527, lr=0.000303]\nSteps: 46%|████▋ | 463/1000 [20:09<16:57, 1.90s/it, loss=0.29, lr=0.000302] \nSteps: 46%|████▋ | 464/1000 [20:11<16:52, 1.89s/it, loss=0.29, lr=0.000302]\nSteps: 46%|████▋ | 464/1000 [20:11<16:52, 1.89s/it, loss=0.279, lr=0.000302]\nSteps: 46%|████▋ | 465/1000 [20:13<16:48, 1.88s/it, loss=0.279, lr=0.000302]\nSteps: 46%|████▋ | 465/1000 [20:13<16:48, 1.88s/it, loss=0.475, lr=0.000301]\nSteps: 47%|████▋ | 466/1000 [20:15<16:44, 1.88s/it, loss=0.475, lr=0.000301]\nSteps: 47%|████▋ | 466/1000 [20:15<16:44, 1.88s/it, loss=0.343, lr=0.0003] \nSteps: 47%|████▋ | 467/1000 [20:17<16:41, 1.88s/it, loss=0.343, lr=0.0003]\nSteps: 47%|████▋ | 467/1000 [20:17<16:41, 1.88s/it, loss=0.299, lr=0.0003]\nSteps: 47%|████▋ | 468/1000 [20:19<16:38, 1.88s/it, loss=0.299, lr=0.0003]\nSteps: 47%|████▋ | 468/1000 [20:19<16:38, 1.88s/it, loss=0.336, lr=0.000299]\nSteps: 47%|████▋ | 469/1000 [20:21<16:35, 1.87s/it, loss=0.336, lr=0.000299]\nSteps: 47%|████▋ | 469/1000 [20:21<16:35, 1.87s/it, loss=1.01, lr=0.000298] \nSteps: 47%|████▋ | 470/1000 [20:22<16:33, 1.87s/it, loss=1.01, lr=0.000298]\nSteps: 47%|████▋ | 470/1000 [20:22<16:33, 1.87s/it, loss=0.577, lr=0.000298]\nSteps: 47%|████▋ | 471/1000 [20:24<16:30, 1.87s/it, loss=0.577, lr=0.000298]\nSteps: 47%|████▋ | 471/1000 [20:24<16:30, 1.87s/it, loss=0.366, lr=0.000297]\nSteps: 47%|████▋ | 472/1000 [20:26<16:28, 1.87s/it, loss=0.366, lr=0.000297]\nSteps: 47%|████▋ | 472/1000 [20:26<16:28, 1.87s/it, loss=0.912, lr=0.000296]\nSteps: 47%|████▋ | 473/1000 [20:28<16:26, 1.87s/it, loss=0.912, lr=0.000296]\nSteps: 47%|████▋ | 473/1000 [20:28<16:26, 1.87s/it, loss=0.422, lr=0.000296]\nSteps: 47%|████▋ | 474/1000 [20:30<16:24, 1.87s/it, loss=0.422, lr=0.000296]\nSteps: 47%|████▋ | 474/1000 [20:30<16:24, 1.87s/it, loss=0.437, lr=0.000295]\nSteps: 48%|████▊ | 475/1000 [20:32<16:21, 1.87s/it, loss=0.437, lr=0.000295]\nSteps: 48%|████▊ | 475/1000 [20:32<16:21, 1.87s/it, loss=0.517, lr=0.000294]\nSteps: 48%|████▊ | 476/1000 [20:34<16:20, 1.87s/it, loss=0.517, lr=0.000294]\nSteps: 48%|████▊ | 476/1000 [20:34<16:20, 1.87s/it, loss=0.304, lr=0.000294]\nSteps: 48%|████▊ | 477/1000 [20:35<16:18, 1.87s/it, loss=0.304, lr=0.000294]\nSteps: 48%|████▊ | 477/1000 [20:35<16:18, 1.87s/it, loss=0.668, lr=0.000293]\nSteps: 48%|████▊ | 478/1000 [20:37<16:17, 1.87s/it, loss=0.668, lr=0.000293]\nSteps: 48%|████▊ | 478/1000 [20:37<16:17, 1.87s/it, loss=0.745, lr=0.000292]\nSteps: 48%|████▊ | 479/1000 [20:39<16:15, 1.87s/it, loss=0.745, lr=0.000292]\nSteps: 48%|████▊ | 479/1000 [20:39<16:15, 1.87s/it, loss=0.335, lr=0.000291]\nSteps: 48%|████▊ | 480/1000 [20:41<16:13, 1.87s/it, loss=0.335, lr=0.000291]\nSteps: 48%|████▊ | 480/1000 [20:41<16:13, 1.87s/it, loss=0.358, lr=0.000291]\nSteps: 48%|████▊ | 481/1000 [20:49<31:21, 3.62s/it, loss=0.358, lr=0.000291]\nSteps: 48%|████▊ | 481/1000 [20:49<31:21, 3.62s/it, loss=0.715, lr=0.00029] \nSteps: 48%|████▊ | 482/1000 [20:51<26:44, 3.10s/it, loss=0.715, lr=0.00029]\nSteps: 48%|████▊ | 482/1000 [20:51<26:44, 3.10s/it, loss=1.03, lr=0.000289]\nSteps: 48%|████▊ | 483/1000 [20:53<23:31, 2.73s/it, loss=1.03, lr=0.000289]\nSteps: 48%|████▊ | 483/1000 [20:53<23:31, 2.73s/it, loss=0.355, lr=0.000289]\nSteps: 48%|████▊ | 484/1000 [20:54<21:16, 2.47s/it, loss=0.355, lr=0.000289]\nSteps: 48%|████▊ | 484/1000 [20:54<21:16, 2.47s/it, loss=0.276, lr=0.000288]\nSteps: 48%|████▊ | 485/1000 [20:56<19:40, 2.29s/it, loss=0.276, lr=0.000288]\nSteps: 48%|████▊ | 485/1000 [20:56<19:40, 2.29s/it, loss=0.664, lr=0.000287]\nSteps: 49%|████▊ | 486/1000 [20:58<18:33, 2.17s/it, loss=0.664, lr=0.000287]\nSteps: 49%|████▊ | 486/1000 [20:58<18:33, 2.17s/it, loss=0.294, lr=0.000287]\nSteps: 49%|████▊ | 487/1000 [21:00<17:46, 2.08s/it, loss=0.294, lr=0.000287]\nSteps: 49%|████▊ | 487/1000 [21:00<17:46, 2.08s/it, loss=0.327, lr=0.000286]\nSteps: 49%|████▉ | 488/1000 [21:02<17:13, 2.02s/it, loss=0.327, lr=0.000286]\nSteps: 49%|████▉ | 488/1000 [21:02<17:13, 2.02s/it, loss=0.493, lr=0.000285]\nSteps: 49%|████▉ | 489/1000 [21:04<16:48, 1.97s/it, loss=0.493, lr=0.000285]\nSteps: 49%|████▉ | 489/1000 [21:04<16:48, 1.97s/it, loss=0.294, lr=0.000284]\nSteps: 49%|████▉ | 490/1000 [21:06<16:31, 1.94s/it, loss=0.294, lr=0.000284]\nSteps: 49%|████▉ | 490/1000 [21:06<16:31, 1.94s/it, loss=0.385, lr=0.000284]\nSteps: 49%|████▉ | 491/1000 [21:08<16:18, 1.92s/it, loss=0.385, lr=0.000284]\nSteps: 49%|████▉ | 491/1000 [21:08<16:18, 1.92s/it, loss=0.769, lr=0.000283]\nSteps: 49%|████▉ | 492/1000 [21:09<16:08, 1.91s/it, loss=0.769, lr=0.000283]\nSteps: 49%|████▉ | 492/1000 [21:09<16:08, 1.91s/it, loss=0.481, lr=0.000282]\nSteps: 49%|████▉ | 493/1000 [21:11<16:00, 1.89s/it, loss=0.481, lr=0.000282]\nSteps: 49%|████▉ | 493/1000 [21:11<16:00, 1.89s/it, loss=0.504, lr=0.000282]\nSteps: 49%|████▉ | 494/1000 [21:13<15:55, 1.89s/it, loss=0.504, lr=0.000282]\nSteps: 49%|████▉ | 494/1000 [21:13<15:55, 1.89s/it, loss=0.78, lr=0.000281] \nSteps: 50%|████▉ | 495/1000 [21:15<15:51, 1.88s/it, loss=0.78, lr=0.000281]\nSteps: 50%|████▉ | 495/1000 [21:15<15:51, 1.88s/it, loss=0.375, lr=0.00028]\nSteps: 50%|████▉ | 496/1000 [21:17<15:47, 1.88s/it, loss=0.375, lr=0.00028]\nSteps: 50%|████▉ | 496/1000 [21:17<15:47, 1.88s/it, loss=0.553, lr=0.000279]\nSteps: 50%|████▉ | 497/1000 [21:19<15:45, 1.88s/it, loss=0.553, lr=0.000279]\nSteps: 50%|████▉ | 497/1000 [21:19<15:45, 1.88s/it, loss=0.602, lr=0.000279]\nSteps: 50%|████▉ | 498/1000 [21:21<15:41, 1.88s/it, loss=0.602, lr=0.000279]\nSteps: 50%|████▉ | 498/1000 [21:21<15:41, 1.88s/it, loss=0.305, lr=0.000278]\nSteps: 50%|████▉ | 499/1000 [21:23<15:38, 1.87s/it, loss=0.305, lr=0.000278]\nSteps: 50%|████▉ | 499/1000 [21:23<15:38, 1.87s/it, loss=0.806, lr=0.000277]\nSteps: 50%|█████ | 500/1000 [21:24<15:37, 1.87s/it, loss=0.806, lr=0.000277]\nSteps: 50%|█████ | 500/1000 [21:24<15:37, 1.87s/it, loss=0.926, lr=0.000277]\nSteps: 50%|█████ | 501/1000 [21:26<15:35, 1.87s/it, loss=0.926, lr=0.000277]\nSteps: 50%|█████ | 501/1000 [21:26<15:35, 1.87s/it, loss=0.813, lr=0.000276]\nSteps: 50%|█████ | 502/1000 [21:28<15:33, 1.87s/it, loss=0.813, lr=0.000276]\nSteps: 50%|█████ | 502/1000 [21:28<15:33, 1.87s/it, loss=0.582, lr=0.000275]\nSteps: 50%|█████ | 503/1000 [21:30<15:30, 1.87s/it, loss=0.582, lr=0.000275]\nSteps: 50%|█████ | 503/1000 [21:30<15:30, 1.87s/it, loss=0.995, lr=0.000274]\nSteps: 50%|█████ | 504/1000 [21:32<15:27, 1.87s/it, loss=0.995, lr=0.000274]\nSteps: 50%|█████ | 504/1000 [21:32<15:27, 1.87s/it, loss=0.305, lr=0.000274]\nSteps: 50%|█████ | 505/1000 [21:34<15:26, 1.87s/it, loss=0.305, lr=0.000274]\nSteps: 50%|█████ | 505/1000 [21:34<15:26, 1.87s/it, loss=0.632, lr=0.000273]\nSteps: 51%|█████ | 506/1000 [21:36<15:23, 1.87s/it, loss=0.632, lr=0.000273]\nSteps: 51%|█████ | 506/1000 [21:36<15:23, 1.87s/it, loss=0.632, lr=0.000272]\nSteps: 51%|█████ | 507/1000 [21:37<15:21, 1.87s/it, loss=0.632, lr=0.000272]\nSteps: 51%|█████ | 507/1000 [21:37<15:21, 1.87s/it, loss=0.711, lr=0.000271]\nSteps: 51%|█████ | 508/1000 [21:39<15:18, 1.87s/it, loss=0.711, lr=0.000271]\nSteps: 51%|█████ | 508/1000 [21:39<15:18, 1.87s/it, loss=0.43, lr=0.000271] \nSteps: 51%|█████ | 509/1000 [21:41<15:17, 1.87s/it, loss=0.43, lr=0.000271]\nSteps: 51%|█████ | 509/1000 [21:41<15:17, 1.87s/it, loss=0.368, lr=0.00027]\nSteps: 51%|█████ | 510/1000 [21:43<15:15, 1.87s/it, loss=0.368, lr=0.00027]\nSteps: 51%|█████ | 510/1000 [21:43<15:15, 1.87s/it, loss=0.375, lr=0.000269]\nSteps: 51%|█████ | 511/1000 [21:51<29:36, 3.63s/it, loss=0.375, lr=0.000269]\nSteps: 51%|█████ | 511/1000 [21:51<29:36, 3.63s/it, loss=1.01, lr=0.000268] \nSteps: 51%|█████ | 512/1000 [21:53<25:14, 3.10s/it, loss=1.01, lr=0.000268]\nSteps: 51%|█████ | 512/1000 [21:53<25:14, 3.10s/it, loss=0.322, lr=0.000268]\nSteps: 51%|█████▏ | 513/1000 [21:55<22:10, 2.73s/it, loss=0.322, lr=0.000268]\nSteps: 51%|█████▏ | 513/1000 [21:55<22:10, 2.73s/it, loss=0.47, lr=0.000267] \nSteps: 51%|█████▏ | 514/1000 [21:56<20:03, 2.48s/it, loss=0.47, lr=0.000267]\nSteps: 51%|█████▏ | 514/1000 [21:56<20:03, 2.48s/it, loss=0.292, lr=0.000266]\nSteps: 52%|█████▏ | 515/1000 [21:58<18:32, 2.29s/it, loss=0.292, lr=0.000266]\nSteps: 52%|█████▏ | 515/1000 [21:58<18:32, 2.29s/it, loss=0.704, lr=0.000266]\nSteps: 52%|█████▏ | 516/1000 [22:00<17:28, 2.17s/it, loss=0.704, lr=0.000266]\nSteps: 52%|█████▏ | 516/1000 [22:00<17:28, 2.17s/it, loss=0.439, lr=0.000265]\nSteps: 52%|█████▏ | 517/1000 [22:02<16:44, 2.08s/it, loss=0.439, lr=0.000265]\nSteps: 52%|█████▏ | 517/1000 [22:02<16:44, 2.08s/it, loss=0.626, lr=0.000264]\nSteps: 52%|█████▏ | 518/1000 [22:04<16:12, 2.02s/it, loss=0.626, lr=0.000264]\nSteps: 52%|█████▏ | 518/1000 [22:04<16:12, 2.02s/it, loss=0.579, lr=0.000263]\nSteps: 52%|█████▏ | 519/1000 [22:06<15:49, 1.97s/it, loss=0.579, lr=0.000263]\nSteps: 52%|█████▏ | 519/1000 [22:06<15:49, 1.97s/it, loss=0.284, lr=0.000263]\nSteps: 52%|█████▏ | 520/1000 [22:08<15:31, 1.94s/it, loss=0.284, lr=0.000263]\nSteps: 52%|█████▏ | 520/1000 [22:08<15:31, 1.94s/it, loss=0.961, lr=0.000262]\nSteps: 52%|█████▏ | 521/1000 [22:10<15:20, 1.92s/it, loss=0.961, lr=0.000262]\nSteps: 52%|█████▏ | 521/1000 [22:10<15:20, 1.92s/it, loss=1.02, lr=0.000261] \nSteps: 52%|█████▏ | 522/1000 [22:11<15:11, 1.91s/it, loss=1.02, lr=0.000261]\nSteps: 52%|█████▏ | 522/1000 [22:11<15:11, 1.91s/it, loss=0.494, lr=0.00026]\nSteps: 52%|█████▏ | 523/1000 [22:13<15:04, 1.90s/it, loss=0.494, lr=0.00026]\nSteps: 52%|█████▏ | 523/1000 [22:13<15:04, 1.90s/it, loss=0.594, lr=0.00026]\nSteps: 52%|█████▏ | 524/1000 [22:15<14:59, 1.89s/it, loss=0.594, lr=0.00026]\nSteps: 52%|█████▏ | 524/1000 [22:15<14:59, 1.89s/it, loss=0.322, lr=0.000259]\nSteps: 52%|█████▎ | 525/1000 [22:17<14:55, 1.88s/it, loss=0.322, lr=0.000259]\nSteps: 52%|█████▎ | 525/1000 [22:17<14:55, 1.88s/it, loss=0.674, lr=0.000258]\nSteps: 53%|█████▎ | 526/1000 [22:19<14:51, 1.88s/it, loss=0.674, lr=0.000258]\nSteps: 53%|█████▎ | 526/1000 [22:19<14:51, 1.88s/it, loss=0.353, lr=0.000257]\nSteps: 53%|█████▎ | 527/1000 [22:21<14:48, 1.88s/it, loss=0.353, lr=0.000257]\nSteps: 53%|█████▎ | 527/1000 [22:21<14:48, 1.88s/it, loss=0.218, lr=0.000257]\nSteps: 53%|█████▎ | 528/1000 [22:23<14:45, 1.88s/it, loss=0.218, lr=0.000257]\nSteps: 53%|█████▎ | 528/1000 [22:23<14:45, 1.88s/it, loss=0.551, lr=0.000256]\nSteps: 53%|█████▎ | 529/1000 [22:25<14:42, 1.87s/it, loss=0.551, lr=0.000256]\nSteps: 53%|█████▎ | 529/1000 [22:25<14:42, 1.87s/it, loss=0.606, lr=0.000255]\nSteps: 53%|█████▎ | 530/1000 [22:26<14:40, 1.87s/it, loss=0.606, lr=0.000255]\nSteps: 53%|█████▎ | 530/1000 [22:26<14:40, 1.87s/it, loss=0.932, lr=0.000254]\nSteps: 53%|█████▎ | 531/1000 [22:28<14:38, 1.87s/it, loss=0.932, lr=0.000254]\nSteps: 53%|█████▎ | 531/1000 [22:28<14:38, 1.87s/it, loss=0.52, lr=0.000254] \nSteps: 53%|█████▎ | 532/1000 [22:30<14:36, 1.87s/it, loss=0.52, lr=0.000254]\nSteps: 53%|█████▎ | 532/1000 [22:30<14:36, 1.87s/it, loss=0.558, lr=0.000253]\nSteps: 53%|█████▎ | 533/1000 [22:32<14:34, 1.87s/it, loss=0.558, lr=0.000253]\nSteps: 53%|█████▎ | 533/1000 [22:32<14:34, 1.87s/it, loss=0.606, lr=0.000252]\nSteps: 53%|█████▎ | 534/1000 [22:34<14:32, 1.87s/it, loss=0.606, lr=0.000252]\nSteps: 53%|█████▎ | 534/1000 [22:34<14:32, 1.87s/it, loss=0.358, lr=0.000251]\nSteps: 54%|█████▎ | 535/1000 [22:36<14:31, 1.87s/it, loss=0.358, lr=0.000251]\nSteps: 54%|█████▎ | 535/1000 [22:36<14:31, 1.87s/it, loss=0.323, lr=0.00025] \nSteps: 54%|█████▎ | 536/1000 [22:38<14:29, 1.87s/it, loss=0.323, lr=0.00025]\nSteps: 54%|█████▎ | 536/1000 [22:38<14:29, 1.87s/it, loss=0.517, lr=0.00025]\nSteps: 54%|█████▎ | 537/1000 [22:39<14:27, 1.87s/it, loss=0.517, lr=0.00025]\nSteps: 54%|█████▎ | 537/1000 [22:40<14:27, 1.87s/it, loss=0.376, lr=0.000249]\nSteps: 54%|█████▍ | 538/1000 [22:41<14:25, 1.87s/it, loss=0.376, lr=0.000249]\nSteps: 54%|█████▍ | 538/1000 [22:41<14:25, 1.87s/it, loss=0.298, lr=0.000248]\nSteps: 54%|█████▍ | 539/1000 [22:43<14:22, 1.87s/it, loss=0.298, lr=0.000248]\nSteps: 54%|█████▍ | 539/1000 [22:43<14:22, 1.87s/it, loss=0.557, lr=0.000247]\nSteps: 54%|█████▍ | 540/1000 [22:45<14:21, 1.87s/it, loss=0.557, lr=0.000247]\nSteps: 54%|█████▍ | 540/1000 [22:45<14:21, 1.87s/it, loss=0.401, lr=0.000247]\nSteps: 54%|█████▍ | 541/1000 [22:53<27:46, 3.63s/it, loss=0.401, lr=0.000247]\nSteps: 54%|█████▍ | 541/1000 [22:53<27:46, 3.63s/it, loss=0.696, lr=0.000246]\nSteps: 54%|█████▍ | 542/1000 [22:55<23:40, 3.10s/it, loss=0.696, lr=0.000246]\nSteps: 54%|█████▍ | 542/1000 [22:55<23:40, 3.10s/it, loss=0.533, lr=0.000245]\nSteps: 54%|█████▍ | 543/1000 [22:57<20:49, 2.73s/it, loss=0.533, lr=0.000245]\nSteps: 54%|█████▍ | 543/1000 [22:57<20:49, 2.73s/it, loss=0.759, lr=0.000244]\nSteps: 54%|█████▍ | 544/1000 [22:58<18:47, 2.47s/it, loss=0.759, lr=0.000244]\nSteps: 54%|█████▍ | 544/1000 [22:58<18:47, 2.47s/it, loss=0.702, lr=0.000244]\nSteps: 55%|█████▍ | 545/1000 [23:00<17:23, 2.29s/it, loss=0.702, lr=0.000244]\nSteps: 55%|█████▍ | 545/1000 [23:00<17:23, 2.29s/it, loss=0.356, lr=0.000243]\nSteps: 55%|█████▍ | 546/1000 [23:02<16:23, 2.17s/it, loss=0.356, lr=0.000243]\nSteps: 55%|█████▍ | 546/1000 [23:02<16:23, 2.17s/it, loss=0.828, lr=0.000242]\nSteps: 55%|█████▍ | 547/1000 [23:04<15:40, 2.08s/it, loss=0.828, lr=0.000242]\nSteps: 55%|█████▍ | 547/1000 [23:04<15:40, 2.08s/it, loss=0.483, lr=0.000241]\nSteps: 55%|█████▍ | 548/1000 [23:06<15:10, 2.01s/it, loss=0.483, lr=0.000241]\nSteps: 55%|█████▍ | 548/1000 [23:06<15:10, 2.01s/it, loss=0.418, lr=0.000241]\nSteps: 55%|█████▍ | 549/1000 [23:08<14:49, 1.97s/it, loss=0.418, lr=0.000241]\nSteps: 55%|█████▍ | 549/1000 [23:08<14:49, 1.97s/it, loss=0.678, lr=0.00024] \nSteps: 55%|█████▌ | 550/1000 [23:10<14:34, 1.94s/it, loss=0.678, lr=0.00024]\nSteps: 55%|█████▌ | 550/1000 [23:10<14:34, 1.94s/it, loss=0.363, lr=0.000239]\nSteps: 55%|█████▌ | 551/1000 [23:12<14:21, 1.92s/it, loss=0.363, lr=0.000239]\nSteps: 55%|█████▌ | 551/1000 [23:12<14:21, 1.92s/it, loss=0.89, lr=0.000238] \nSteps: 55%|█████▌ | 552/1000 [23:13<14:13, 1.91s/it, loss=0.89, lr=0.000238]\nSteps: 55%|█████▌ | 552/1000 [23:13<14:13, 1.91s/it, loss=0.366, lr=0.000237]\nSteps: 55%|█████▌ | 553/1000 [23:15<14:07, 1.89s/it, loss=0.366, lr=0.000237]\nSteps: 55%|█████▌ | 553/1000 [23:15<14:07, 1.89s/it, loss=0.379, lr=0.000237]\nSteps: 55%|█████▌ | 554/1000 [23:17<14:01, 1.89s/it, loss=0.379, lr=0.000237]\nSteps: 55%|█████▌ | 554/1000 [23:17<14:01, 1.89s/it, loss=0.333, lr=0.000236]\nSteps: 56%|█████▌ | 555/1000 [23:19<13:57, 1.88s/it, loss=0.333, lr=0.000236]\nSteps: 56%|█████▌ | 555/1000 [23:19<13:57, 1.88s/it, loss=0.532, lr=0.000235]\nSteps: 56%|█████▌ | 556/1000 [23:21<13:54, 1.88s/it, loss=0.532, lr=0.000235]\nSteps: 56%|█████▌ | 556/1000 [23:21<13:54, 1.88s/it, loss=0.584, lr=0.000234]\nSteps: 56%|█████▌ | 557/1000 [23:23<13:51, 1.88s/it, loss=0.584, lr=0.000234]\nSteps: 56%|█████▌ | 557/1000 [23:23<13:51, 1.88s/it, loss=0.409, lr=0.000234]\nSteps: 56%|█████▌ | 558/1000 [23:25<13:49, 1.88s/it, loss=0.409, lr=0.000234]\nSteps: 56%|█████▌ | 558/1000 [23:25<13:49, 1.88s/it, loss=0.335, lr=0.000233]\nSteps: 56%|█████▌ | 559/1000 [23:27<13:46, 1.87s/it, loss=0.335, lr=0.000233]\nSteps: 56%|█████▌ | 559/1000 [23:27<13:46, 1.87s/it, loss=0.624, lr=0.000232]\nSteps: 56%|█████▌ | 560/1000 [23:28<13:44, 1.87s/it, loss=0.624, lr=0.000232]\nSteps: 56%|█████▌ | 560/1000 [23:28<13:44, 1.87s/it, loss=1.03, lr=0.000231] \nSteps: 56%|█████▌ | 561/1000 [23:30<13:42, 1.87s/it, loss=1.03, lr=0.000231]\nSteps: 56%|█████▌ | 561/1000 [23:30<13:42, 1.87s/it, loss=0.635, lr=0.000231]\nSteps: 56%|█████▌ | 562/1000 [23:32<13:40, 1.87s/it, loss=0.635, lr=0.000231]\nSteps: 56%|█████▌ | 562/1000 [23:32<13:40, 1.87s/it, loss=0.686, lr=0.00023] \nSteps: 56%|█████▋ | 563/1000 [23:34<13:38, 1.87s/it, loss=0.686, lr=0.00023]\nSteps: 56%|█████▋ | 563/1000 [23:34<13:38, 1.87s/it, loss=0.336, lr=0.000229]\nSteps: 56%|█████▋ | 564/1000 [23:36<13:36, 1.87s/it, loss=0.336, lr=0.000229]\nSteps: 56%|█████▋ | 564/1000 [23:36<13:36, 1.87s/it, loss=0.67, lr=0.000228] \nSteps: 56%|█████▋ | 565/1000 [23:41<21:23, 2.95s/it, loss=0.67, lr=0.000228]\nSteps: 56%|█████▋ | 565/1000 [23:41<21:23, 2.95s/it, loss=0.439, lr=0.000227]\nSteps: 57%|█████▋ | 566/1000 [23:43<18:57, 2.62s/it, loss=0.439, lr=0.000227]\nSteps: 57%|█████▋ | 566/1000 [23:43<18:57, 2.62s/it, loss=0.308, lr=0.000227]\nSteps: 57%|█████▋ | 567/1000 [23:45<17:17, 2.40s/it, loss=0.308, lr=0.000227]\nSteps: 57%|█████▋ | 567/1000 [23:45<17:17, 2.40s/it, loss=0.57, lr=0.000226] \nSteps: 57%|█████▋ | 568/1000 [23:47<16:07, 2.24s/it, loss=0.57, lr=0.000226]\nSteps: 57%|█████▋ | 568/1000 [23:47<16:07, 2.24s/it, loss=0.34, lr=0.000225]\nSteps: 57%|█████▋ | 569/1000 [23:49<15:17, 2.13s/it, loss=0.34, lr=0.000225]\nSteps: 57%|█████▋ | 569/1000 [23:49<15:17, 2.13s/it, loss=0.604, lr=0.000224]\nSteps: 57%|█████▋ | 570/1000 [23:51<14:42, 2.05s/it, loss=0.604, lr=0.000224]\nSteps: 57%|█████▋ | 570/1000 [23:51<14:42, 2.05s/it, loss=0.75, lr=0.000224] \nSteps: 57%|█████▋ | 571/1000 [23:58<26:56, 3.77s/it, loss=0.75, lr=0.000224]\nSteps: 57%|█████▋ | 571/1000 [23:58<26:56, 3.77s/it, loss=0.399, lr=0.000223]\nSteps: 57%|█████▋ | 572/1000 [24:00<22:48, 3.20s/it, loss=0.399, lr=0.000223]\nSteps: 57%|█████▋ | 572/1000 [24:00<22:48, 3.20s/it, loss=0.568, lr=0.000222]\nSteps: 57%|█████▋ | 573/1000 [24:02<19:54, 2.80s/it, loss=0.568, lr=0.000222]\nSteps: 57%|█████▋ | 573/1000 [24:02<19:54, 2.80s/it, loss=0.318, lr=0.000221]\nSteps: 57%|█████▋ | 574/1000 [24:04<17:53, 2.52s/it, loss=0.318, lr=0.000221]\nSteps: 57%|█████▋ | 574/1000 [24:04<17:53, 2.52s/it, loss=0.267, lr=0.00022] \nSteps: 57%|█████▊ | 575/1000 [24:06<16:27, 2.32s/it, loss=0.267, lr=0.00022]\nSteps: 57%|█████▊ | 575/1000 [24:06<16:27, 2.32s/it, loss=0.557, lr=0.00022]\nSteps: 58%|█████▊ | 576/1000 [24:08<15:27, 2.19s/it, loss=0.557, lr=0.00022]\nSteps: 58%|█████▊ | 576/1000 [24:08<15:27, 2.19s/it, loss=0.874, lr=0.000219]\nSteps: 58%|█████▊ | 577/1000 [24:10<14:45, 2.09s/it, loss=0.874, lr=0.000219]\nSteps: 58%|█████▊ | 577/1000 [24:10<14:45, 2.09s/it, loss=0.478, lr=0.000218]\nSteps: 58%|█████▊ | 578/1000 [24:12<14:15, 2.03s/it, loss=0.478, lr=0.000218]\nSteps: 58%|█████▊ | 578/1000 [24:12<14:15, 2.03s/it, loss=1.02, lr=0.000217] \nSteps: 58%|█████▊ | 579/1000 [24:13<13:54, 1.98s/it, loss=1.02, lr=0.000217]\nSteps: 58%|█████▊ | 579/1000 [24:13<13:54, 1.98s/it, loss=0.802, lr=0.000216]\nSteps: 58%|█████▊ | 580/1000 [24:15<13:38, 1.95s/it, loss=0.802, lr=0.000216]\nSteps: 58%|█████▊ | 580/1000 [24:15<13:38, 1.95s/it, loss=0.638, lr=0.000216]\nSteps: 58%|█████▊ | 581/1000 [24:17<13:26, 1.93s/it, loss=0.638, lr=0.000216]\nSteps: 58%|█████▊ | 581/1000 [24:17<13:26, 1.93s/it, loss=0.338, lr=0.000215]\nSteps: 58%|█████▊ | 582/1000 [24:19<13:17, 1.91s/it, loss=0.338, lr=0.000215]\nSteps: 58%|█████▊ | 582/1000 [24:19<13:17, 1.91s/it, loss=0.299, lr=0.000214]\nSteps: 58%|█████▊ | 583/1000 [24:21<13:11, 1.90s/it, loss=0.299, lr=0.000214]\nSteps: 58%|█████▊ | 583/1000 [24:21<13:11, 1.90s/it, loss=0.568, lr=0.000213]\nSteps: 58%|█████▊ | 584/1000 [24:23<13:05, 1.89s/it, loss=0.568, lr=0.000213]\nSteps: 58%|█████▊ | 584/1000 [24:23<13:05, 1.89s/it, loss=0.278, lr=0.000213]\nSteps: 58%|█████▊ | 585/1000 [24:25<13:01, 1.88s/it, loss=0.278, lr=0.000213]\nSteps: 58%|█████▊ | 585/1000 [24:25<13:01, 1.88s/it, loss=0.555, lr=0.000212]\nSteps: 59%|█████▊ | 586/1000 [24:27<12:59, 1.88s/it, loss=0.555, lr=0.000212]\nSteps: 59%|█████▊ | 586/1000 [24:27<12:59, 1.88s/it, loss=0.329, lr=0.000211]\nSteps: 59%|█████▊ | 587/1000 [24:28<12:56, 1.88s/it, loss=0.329, lr=0.000211]\nSteps: 59%|█████▊ | 587/1000 [24:28<12:56, 1.88s/it, loss=0.315, lr=0.00021] \nSteps: 59%|█████▉ | 588/1000 [24:30<12:52, 1.88s/it, loss=0.315, lr=0.00021]\nSteps: 59%|█████▉ | 588/1000 [24:30<12:52, 1.88s/it, loss=0.919, lr=0.000209]\nSteps: 59%|█████▉ | 589/1000 [24:32<12:49, 1.87s/it, loss=0.919, lr=0.000209]\nSteps: 59%|█████▉ | 589/1000 [24:32<12:49, 1.87s/it, loss=0.477, lr=0.000209]\nSteps: 59%|█████▉ | 590/1000 [24:34<12:48, 1.87s/it, loss=0.477, lr=0.000209]\nSteps: 59%|█████▉ | 590/1000 [24:34<12:48, 1.87s/it, loss=0.706, lr=0.000208]\nSteps: 59%|█████▉ | 591/1000 [24:36<12:45, 1.87s/it, loss=0.706, lr=0.000208]\nSteps: 59%|█████▉ | 591/1000 [24:36<12:45, 1.87s/it, loss=0.396, lr=0.000207]\nSteps: 59%|█████▉ | 592/1000 [24:38<12:44, 1.87s/it, loss=0.396, lr=0.000207]\nSteps: 59%|█████▉ | 592/1000 [24:38<12:44, 1.87s/it, loss=0.563, lr=0.000206]\nSteps: 59%|█████▉ | 593/1000 [24:40<12:42, 1.87s/it, loss=0.563, lr=0.000206]\nSteps: 59%|█████▉ | 593/1000 [24:40<12:42, 1.87s/it, loss=0.316, lr=0.000205]\nSteps: 59%|█████▉ | 594/1000 [24:41<12:39, 1.87s/it, loss=0.316, lr=0.000205]\nSteps: 59%|█████▉ | 594/1000 [24:42<12:39, 1.87s/it, loss=0.312, lr=0.000205]\nSteps: 60%|█████▉ | 595/1000 [24:43<12:37, 1.87s/it, loss=0.312, lr=0.000205]\nSteps: 60%|█████▉ | 595/1000 [24:43<12:37, 1.87s/it, loss=1.03, lr=0.000204] \nSteps: 60%|█████▉ | 596/1000 [24:45<12:35, 1.87s/it, loss=1.03, lr=0.000204]\nSteps: 60%|█████▉ | 596/1000 [24:45<12:35, 1.87s/it, loss=0.493, lr=0.000203]\nSteps: 60%|█████▉ | 597/1000 [24:47<12:33, 1.87s/it, loss=0.493, lr=0.000203]\nSteps: 60%|█████▉ | 597/1000 [24:47<12:33, 1.87s/it, loss=0.395, lr=0.000202]\nSteps: 60%|█████▉ | 598/1000 [24:49<12:31, 1.87s/it, loss=0.395, lr=0.000202]\nSteps: 60%|█████▉ | 598/1000 [24:49<12:31, 1.87s/it, loss=0.312, lr=0.000202]\nSteps: 60%|█████▉ | 599/1000 [24:51<12:31, 1.87s/it, loss=0.312, lr=0.000202]\nSteps: 60%|█████▉ | 599/1000 [24:51<12:31, 1.87s/it, loss=0.5, lr=0.000201] \nSteps: 60%|██████ | 600/1000 [24:53<12:28, 1.87s/it, loss=0.5, lr=0.000201]\nSteps: 60%|██████ | 600/1000 [24:53<12:28, 1.87s/it, loss=0.747, lr=0.0002]\nSteps: 60%|██████ | 601/1000 [25:01<24:21, 3.66s/it, loss=0.747, lr=0.0002]\nSteps: 60%|██████ | 601/1000 [25:01<24:21, 3.66s/it, loss=0.944, lr=0.000199]\nSteps: 60%|██████ | 602/1000 [25:02<20:43, 3.12s/it, loss=0.944, lr=0.000199]\nSteps: 60%|██████ | 602/1000 [25:02<20:43, 3.12s/it, loss=0.316, lr=0.000198]\nSteps: 60%|██████ | 603/1000 [25:04<18:11, 2.75s/it, loss=0.316, lr=0.000198]\nSteps: 60%|██████ | 603/1000 [25:04<18:11, 2.75s/it, loss=0.32, lr=0.000198] \nSteps: 60%|██████ | 604/1000 [25:06<16:24, 2.49s/it, loss=0.32, lr=0.000198]\nSteps: 60%|██████ | 604/1000 [25:06<16:24, 2.49s/it, loss=0.311, lr=0.000197]\nSteps: 60%|██████ | 605/1000 [25:08<15:08, 2.30s/it, loss=0.311, lr=0.000197]\nSteps: 60%|██████ | 605/1000 [25:08<15:08, 2.30s/it, loss=0.359, lr=0.000196]\nSteps: 61%|██████ | 606/1000 [25:10<14:15, 2.17s/it, loss=0.359, lr=0.000196]\nSteps: 61%|██████ | 606/1000 [25:10<14:15, 2.17s/it, loss=0.376, lr=0.000195]\nSteps: 61%|██████ | 607/1000 [25:12<13:37, 2.08s/it, loss=0.376, lr=0.000195]\nSteps: 61%|██████ | 607/1000 [25:12<13:37, 2.08s/it, loss=0.972, lr=0.000195]\nSteps: 61%|██████ | 608/1000 [25:14<13:10, 2.02s/it, loss=0.972, lr=0.000195]\nSteps: 61%|██████ | 608/1000 [25:14<13:10, 2.02s/it, loss=0.953, lr=0.000194]\nSteps: 61%|██████ | 609/1000 [25:16<12:51, 1.97s/it, loss=0.953, lr=0.000194]\nSteps: 61%|██████ | 609/1000 [25:16<12:51, 1.97s/it, loss=0.748, lr=0.000193]\nSteps: 61%|██████ | 610/1000 [25:17<12:37, 1.94s/it, loss=0.748, lr=0.000193]\nSteps: 61%|██████ | 610/1000 [25:17<12:37, 1.94s/it, loss=0.962, lr=0.000192]\nSteps: 61%|██████ | 611/1000 [25:19<12:28, 1.92s/it, loss=0.962, lr=0.000192]\nSteps: 61%|██████ | 611/1000 [25:19<12:28, 1.92s/it, loss=0.633, lr=0.000191]\nSteps: 61%|██████ | 612/1000 [25:21<12:20, 1.91s/it, loss=0.633, lr=0.000191]\nSteps: 61%|██████ | 612/1000 [25:21<12:20, 1.91s/it, loss=0.295, lr=0.000191]\nSteps: 61%|██████▏ | 613/1000 [25:23<12:14, 1.90s/it, loss=0.295, lr=0.000191]\nSteps: 61%|██████▏ | 613/1000 [25:23<12:14, 1.90s/it, loss=0.616, lr=0.00019] \nSteps: 61%|██████▏ | 614/1000 [25:25<12:09, 1.89s/it, loss=0.616, lr=0.00019]\nSteps: 61%|██████▏ | 614/1000 [25:25<12:09, 1.89s/it, loss=0.307, lr=0.000189]\nSteps: 62%|██████▏ | 615/1000 [25:27<12:05, 1.89s/it, loss=0.307, lr=0.000189]\nSteps: 62%|██████▏ | 615/1000 [25:27<12:05, 1.89s/it, loss=0.786, lr=0.000188]\nSteps: 62%|██████▏ | 616/1000 [25:29<12:02, 1.88s/it, loss=0.786, lr=0.000188]\nSteps: 62%|██████▏ | 616/1000 [25:29<12:02, 1.88s/it, loss=0.587, lr=0.000187]\nSteps: 62%|██████▏ | 617/1000 [25:31<11:59, 1.88s/it, loss=0.587, lr=0.000187]\nSteps: 62%|██████▏ | 617/1000 [25:31<11:59, 1.88s/it, loss=0.473, lr=0.000187]\nSteps: 62%|██████▏ | 618/1000 [25:32<11:56, 1.88s/it, loss=0.473, lr=0.000187]\nSteps: 62%|██████▏ | 618/1000 [25:32<11:56, 1.88s/it, loss=0.537, lr=0.000186]\nSteps: 62%|██████▏ | 619/1000 [25:34<11:54, 1.87s/it, loss=0.537, lr=0.000186]\nSteps: 62%|██████▏ | 619/1000 [25:34<11:54, 1.87s/it, loss=0.884, lr=0.000185]\nSteps: 62%|██████▏ | 620/1000 [25:36<11:51, 1.87s/it, loss=0.884, lr=0.000185]\nSteps: 62%|██████▏ | 620/1000 [25:36<11:51, 1.87s/it, loss=0.388, lr=0.000184]\nSteps: 62%|██████▏ | 621/1000 [25:38<11:49, 1.87s/it, loss=0.388, lr=0.000184]\nSteps: 62%|██████▏ | 621/1000 [25:38<11:49, 1.87s/it, loss=0.436, lr=0.000184]\nSteps: 62%|██████▏ | 622/1000 [25:40<11:48, 1.87s/it, loss=0.436, lr=0.000184]\nSteps: 62%|██████▏ | 622/1000 [25:40<11:48, 1.87s/it, loss=0.293, lr=0.000183]\nSteps: 62%|██████▏ | 623/1000 [25:42<11:46, 1.87s/it, loss=0.293, lr=0.000183]\nSteps: 62%|██████▏ | 623/1000 [25:42<11:46, 1.87s/it, loss=0.54, lr=0.000182] \nSteps: 62%|██████▏ | 624/1000 [25:44<11:44, 1.87s/it, loss=0.54, lr=0.000182]\nSteps: 62%|██████▏ | 624/1000 [25:44<11:44, 1.87s/it, loss=0.226, lr=0.000181]\nSteps: 62%|██████▎ | 625/1000 [25:45<11:42, 1.87s/it, loss=0.226, lr=0.000181]\nSteps: 62%|██████▎ | 625/1000 [25:45<11:42, 1.87s/it, loss=0.816, lr=0.00018] \nSteps: 63%|██████▎ | 626/1000 [25:47<11:40, 1.87s/it, loss=0.816, lr=0.00018]\nSteps: 63%|██████▎ | 626/1000 [25:47<11:40, 1.87s/it, loss=0.36, lr=0.00018] \nSteps: 63%|██████▎ | 627/1000 [25:49<11:38, 1.87s/it, loss=0.36, lr=0.00018]\nSteps: 63%|██████▎ | 627/1000 [25:49<11:38, 1.87s/it, loss=0.569, lr=0.000179]\nSteps: 63%|██████▎ | 628/1000 [25:51<11:36, 1.87s/it, loss=0.569, lr=0.000179]\nSteps: 63%|██████▎ | 628/1000 [25:51<11:36, 1.87s/it, loss=0.617, lr=0.000178]\nSteps: 63%|██████▎ | 629/1000 [25:53<11:34, 1.87s/it, loss=0.617, lr=0.000178]\nSteps: 63%|██████▎ | 629/1000 [25:53<11:34, 1.87s/it, loss=0.592, lr=0.000177]\nSteps: 63%|██████▎ | 630/1000 [25:55<11:32, 1.87s/it, loss=0.592, lr=0.000177]\nSteps: 63%|██████▎ | 630/1000 [25:55<11:32, 1.87s/it, loss=0.288, lr=0.000176]\nSteps: 63%|██████▎ | 631/1000 [26:02<22:01, 3.58s/it, loss=0.288, lr=0.000176]\nSteps: 63%|██████▎ | 631/1000 [26:02<22:01, 3.58s/it, loss=0.517, lr=0.000176]\nSteps: 63%|██████▎ | 632/1000 [26:04<18:49, 3.07s/it, loss=0.517, lr=0.000176]\nSteps: 63%|██████▎ | 632/1000 [26:04<18:49, 3.07s/it, loss=0.57, lr=0.000175] \nSteps: 63%|██████▎ | 633/1000 [26:06<16:33, 2.71s/it, loss=0.57, lr=0.000175]\nSteps: 63%|██████▎ | 633/1000 [26:06<16:33, 2.71s/it, loss=1.01, lr=0.000174]\nSteps: 63%|██████▎ | 634/1000 [26:08<14:58, 2.46s/it, loss=1.01, lr=0.000174]\nSteps: 63%|██████▎ | 634/1000 [26:08<14:58, 2.46s/it, loss=0.282, lr=0.000173]\nSteps: 64%|██████▎ | 635/1000 [26:10<13:51, 2.28s/it, loss=0.282, lr=0.000173]\nSteps: 64%|██████▎ | 635/1000 [26:10<13:51, 2.28s/it, loss=0.437, lr=0.000173]\nSteps: 64%|██████▎ | 636/1000 [26:12<13:05, 2.16s/it, loss=0.437, lr=0.000173]\nSteps: 64%|██████▎ | 636/1000 [26:12<13:05, 2.16s/it, loss=0.302, lr=0.000172]\nSteps: 64%|██████▎ | 637/1000 [26:14<12:32, 2.07s/it, loss=0.302, lr=0.000172]\nSteps: 64%|██████▎ | 637/1000 [26:14<12:32, 2.07s/it, loss=0.353, lr=0.000171]\nSteps: 64%|██████▍ | 638/1000 [26:16<12:07, 2.01s/it, loss=0.353, lr=0.000171]\nSteps: 64%|██████▍ | 638/1000 [26:16<12:07, 2.01s/it, loss=0.327, lr=0.00017] \nSteps: 64%|██████▍ | 639/1000 [26:17<11:50, 1.97s/it, loss=0.327, lr=0.00017]\nSteps: 64%|██████▍ | 639/1000 [26:17<11:50, 1.97s/it, loss=0.421, lr=0.000169]\nSteps: 64%|██████▍ | 640/1000 [26:19<11:37, 1.94s/it, loss=0.421, lr=0.000169]\nSteps: 64%|██████▍ | 640/1000 [26:19<11:37, 1.94s/it, loss=0.428, lr=0.000169]\nSteps: 64%|██████▍ | 641/1000 [26:21<11:28, 1.92s/it, loss=0.428, lr=0.000169]\nSteps: 64%|██████▍ | 641/1000 [26:21<11:28, 1.92s/it, loss=0.291, lr=0.000168]\nSteps: 64%|██████▍ | 642/1000 [26:23<11:21, 1.90s/it, loss=0.291, lr=0.000168]\nSteps: 64%|██████▍ | 642/1000 [26:23<11:21, 1.90s/it, loss=0.631, lr=0.000167]\nSteps: 64%|██████▍ | 643/1000 [26:25<11:16, 1.89s/it, loss=0.631, lr=0.000167]\nSteps: 64%|██████▍ | 643/1000 [26:25<11:16, 1.89s/it, loss=0.315, lr=0.000166]\nSteps: 64%|██████▍ | 644/1000 [26:27<11:12, 1.89s/it, loss=0.315, lr=0.000166]\nSteps: 64%|██████▍ | 644/1000 [26:27<11:12, 1.89s/it, loss=0.345, lr=0.000166]\nSteps: 64%|██████▍ | 645/1000 [26:29<11:08, 1.88s/it, loss=0.345, lr=0.000166]\nSteps: 64%|██████▍ | 645/1000 [26:29<11:08, 1.88s/it, loss=0.714, lr=0.000165]\nSteps: 65%|██████▍ | 646/1000 [26:30<11:05, 1.88s/it, loss=0.714, lr=0.000165]\nSteps: 65%|██████▍ | 646/1000 [26:30<11:05, 1.88s/it, loss=1.03, lr=0.000164] \nSteps: 65%|██████▍ | 647/1000 [26:32<11:02, 1.88s/it, loss=1.03, lr=0.000164]\nSteps: 65%|██████▍ | 647/1000 [26:32<11:02, 1.88s/it, loss=0.933, lr=0.000163]\nSteps: 65%|██████▍ | 648/1000 [26:34<10:59, 1.87s/it, loss=0.933, lr=0.000163]\nSteps: 65%|██████▍ | 648/1000 [26:34<10:59, 1.87s/it, loss=0.308, lr=0.000163]\nSteps: 65%|██████▍ | 649/1000 [26:36<10:57, 1.87s/it, loss=0.308, lr=0.000163]\nSteps: 65%|██████▍ | 649/1000 [26:36<10:57, 1.87s/it, loss=0.408, lr=0.000162]\nSteps: 65%|██████▌ | 650/1000 [26:38<10:55, 1.87s/it, loss=0.408, lr=0.000162]\nSteps: 65%|██████▌ | 650/1000 [26:38<10:55, 1.87s/it, loss=0.374, lr=0.000161]\nSteps: 65%|██████▌ | 651/1000 [26:40<10:53, 1.87s/it, loss=0.374, lr=0.000161]\nSteps: 65%|██████▌ | 651/1000 [26:40<10:53, 1.87s/it, loss=0.295, lr=0.00016] \nSteps: 65%|██████▌ | 652/1000 [26:42<10:51, 1.87s/it, loss=0.295, lr=0.00016]\nSteps: 65%|██████▌ | 652/1000 [26:42<10:51, 1.87s/it, loss=0.458, lr=0.000159]\nSteps: 65%|██████▌ | 653/1000 [26:44<10:49, 1.87s/it, loss=0.458, lr=0.000159]\nSteps: 65%|██████▌ | 653/1000 [26:44<10:49, 1.87s/it, loss=0.286, lr=0.000159]\nSteps: 65%|██████▌ | 654/1000 [26:45<10:47, 1.87s/it, loss=0.286, lr=0.000159]\nSteps: 65%|██████▌ | 654/1000 [26:45<10:47, 1.87s/it, loss=0.394, lr=0.000158]\nSteps: 66%|██████▌ | 655/1000 [26:47<10:46, 1.87s/it, loss=0.394, lr=0.000158]\nSteps: 66%|██████▌ | 655/1000 [26:47<10:46, 1.87s/it, loss=0.894, lr=0.000157]\nSteps: 66%|██████▌ | 656/1000 [26:49<10:44, 1.87s/it, loss=0.894, lr=0.000157]\nSteps: 66%|██████▌ | 656/1000 [26:49<10:44, 1.87s/it, loss=0.28, lr=0.000156] \nSteps: 66%|██████▌ | 657/1000 [26:51<10:41, 1.87s/it, loss=0.28, lr=0.000156]\nSteps: 66%|██████▌ | 657/1000 [26:51<10:41, 1.87s/it, loss=0.316, lr=0.000156]\nSteps: 66%|██████▌ | 658/1000 [26:53<10:40, 1.87s/it, loss=0.316, lr=0.000156]\nSteps: 66%|██████▌ | 658/1000 [26:53<10:40, 1.87s/it, loss=0.992, lr=0.000155]\nSteps: 66%|██████▌ | 659/1000 [26:55<10:38, 1.87s/it, loss=0.992, lr=0.000155]\nSteps: 66%|██████▌ | 659/1000 [26:55<10:38, 1.87s/it, loss=0.338, lr=0.000154]\nSteps: 66%|██████▌ | 660/1000 [26:57<10:36, 1.87s/it, loss=0.338, lr=0.000154]\nSteps: 66%|██████▌ | 660/1000 [26:57<10:36, 1.87s/it, loss=0.535, lr=0.000153]\nSteps: 66%|██████▌ | 661/1000 [27:04<20:18, 3.59s/it, loss=0.535, lr=0.000153]\nSteps: 66%|██████▌ | 661/1000 [27:04<20:18, 3.59s/it, loss=0.435, lr=0.000153]\nSteps: 66%|██████▌ | 662/1000 [27:06<17:19, 3.08s/it, loss=0.435, lr=0.000153]\nSteps: 66%|██████▌ | 662/1000 [27:06<17:19, 3.08s/it, loss=0.683, lr=0.000152]\nSteps: 66%|██████▋ | 663/1000 [27:08<15:14, 2.71s/it, loss=0.683, lr=0.000152]\nSteps: 66%|██████▋ | 663/1000 [27:08<15:14, 2.71s/it, loss=0.694, lr=0.000151]\nSteps: 66%|██████▋ | 664/1000 [27:10<13:47, 2.46s/it, loss=0.694, lr=0.000151]\nSteps: 66%|██████▋ | 664/1000 [27:10<13:47, 2.46s/it, loss=0.385, lr=0.00015] \nSteps: 66%|██████▋ | 665/1000 [27:12<12:45, 2.28s/it, loss=0.385, lr=0.00015]\nSteps: 66%|██████▋ | 665/1000 [27:12<12:45, 2.28s/it, loss=0.316, lr=0.00015]\nSteps: 67%|██████▋ | 666/1000 [27:14<12:01, 2.16s/it, loss=0.316, lr=0.00015]\nSteps: 67%|██████▋ | 666/1000 [27:14<12:01, 2.16s/it, loss=0.866, lr=0.000149]\nSteps: 67%|██████▋ | 667/1000 [27:16<11:30, 2.07s/it, loss=0.866, lr=0.000149]\nSteps: 67%|██████▋ | 667/1000 [27:16<11:30, 2.07s/it, loss=0.656, lr=0.000148]\nSteps: 67%|██████▋ | 668/1000 [27:17<11:07, 2.01s/it, loss=0.656, lr=0.000148]\nSteps: 67%|██████▋ | 668/1000 [27:17<11:07, 2.01s/it, loss=0.43, lr=0.000147] \nSteps: 67%|██████▋ | 669/1000 [27:19<10:52, 1.97s/it, loss=0.43, lr=0.000147]\nSteps: 67%|██████▋ | 669/1000 [27:19<10:52, 1.97s/it, loss=1.02, lr=0.000146]\nSteps: 67%|██████▋ | 670/1000 [27:21<10:40, 1.94s/it, loss=1.02, lr=0.000146]\nSteps: 67%|██████▋ | 670/1000 [27:21<10:40, 1.94s/it, loss=0.334, lr=0.000146]\nSteps: 67%|██████▋ | 671/1000 [27:23<10:31, 1.92s/it, loss=0.334, lr=0.000146]\nSteps: 67%|██████▋ | 671/1000 [27:23<10:31, 1.92s/it, loss=0.28, lr=0.000145] \nSteps: 67%|██████▋ | 672/1000 [27:25<10:24, 1.91s/it, loss=0.28, lr=0.000145]\nSteps: 67%|██████▋ | 672/1000 [27:25<10:24, 1.91s/it, loss=0.327, lr=0.000144]\nSteps: 67%|██████▋ | 673/1000 [27:27<10:19, 1.89s/it, loss=0.327, lr=0.000144]\nSteps: 67%|██████▋ | 673/1000 [27:27<10:19, 1.89s/it, loss=1, lr=0.000143] \nSteps: 67%|██████▋ | 674/1000 [27:29<10:15, 1.89s/it, loss=1, lr=0.000143]\nSteps: 67%|██████▋ | 674/1000 [27:29<10:15, 1.89s/it, loss=0.946, lr=0.000143]\nSteps: 68%|██████▊ | 675/1000 [27:30<10:11, 1.88s/it, loss=0.946, lr=0.000143]\nSteps: 68%|██████▊ | 675/1000 [27:30<10:11, 1.88s/it, loss=0.582, lr=0.000142]\nSteps: 68%|██████▊ | 676/1000 [27:32<10:09, 1.88s/it, loss=0.582, lr=0.000142]\nSteps: 68%|██████▊ | 676/1000 [27:32<10:09, 1.88s/it, loss=0.33, lr=0.000141] \nSteps: 68%|██████▊ | 677/1000 [27:34<10:06, 1.88s/it, loss=0.33, lr=0.000141]\nSteps: 68%|██████▊ | 677/1000 [27:34<10:06, 1.88s/it, loss=0.237, lr=0.00014]\nSteps: 68%|██████▊ | 678/1000 [27:36<10:04, 1.88s/it, loss=0.237, lr=0.00014]\nSteps: 68%|██████▊ | 678/1000 [27:36<10:04, 1.88s/it, loss=0.393, lr=0.00014]\nSteps: 68%|██████▊ | 679/1000 [27:38<10:02, 1.88s/it, loss=0.393, lr=0.00014]\nSteps: 68%|██████▊ | 679/1000 [27:38<10:02, 1.88s/it, loss=0.812, lr=0.000139]\nSteps: 68%|██████▊ | 680/1000 [27:40<10:00, 1.88s/it, loss=0.812, lr=0.000139]\nSteps: 68%|██████▊ | 680/1000 [27:40<10:00, 1.88s/it, loss=0.74, lr=0.000138] \nSteps: 68%|██████▊ | 681/1000 [27:42<09:57, 1.87s/it, loss=0.74, lr=0.000138]\nSteps: 68%|██████▊ | 681/1000 [27:42<09:57, 1.87s/it, loss=1.04, lr=0.000137]\nSteps: 68%|██████▊ | 682/1000 [27:44<09:55, 1.87s/it, loss=1.04, lr=0.000137]\nSteps: 68%|██████▊ | 682/1000 [27:44<09:55, 1.87s/it, loss=0.292, lr=0.000137]\nSteps: 68%|██████▊ | 683/1000 [27:45<09:53, 1.87s/it, loss=0.292, lr=0.000137]\nSteps: 68%|██████▊ | 683/1000 [27:45<09:53, 1.87s/it, loss=0.491, lr=0.000136]\nSteps: 68%|██████▊ | 684/1000 [27:47<09:50, 1.87s/it, loss=0.491, lr=0.000136]\nSteps: 68%|██████▊ | 684/1000 [27:47<09:50, 1.87s/it, loss=0.56, lr=0.000135] \nSteps: 68%|██████▊ | 685/1000 [27:49<09:49, 1.87s/it, loss=0.56, lr=0.000135]\nSteps: 68%|██████▊ | 685/1000 [27:49<09:49, 1.87s/it, loss=0.931, lr=0.000134]\nSteps: 69%|██████▊ | 686/1000 [27:51<09:47, 1.87s/it, loss=0.931, lr=0.000134]\nSteps: 69%|██████▊ | 686/1000 [27:51<09:47, 1.87s/it, loss=0.9, lr=0.000134] \nSteps: 69%|██████▊ | 687/1000 [27:53<09:45, 1.87s/it, loss=0.9, lr=0.000134]\nSteps: 69%|██████▊ | 687/1000 [27:53<09:45, 1.87s/it, loss=0.472, lr=0.000133]\nSteps: 69%|██████▉ | 688/1000 [27:55<09:44, 1.87s/it, loss=0.472, lr=0.000133]\nSteps: 69%|██████▉ | 688/1000 [27:55<09:44, 1.87s/it, loss=0.273, lr=0.000132]\nSteps: 69%|██████▉ | 689/1000 [27:57<09:42, 1.87s/it, loss=0.273, lr=0.000132]\nSteps: 69%|██████▉ | 689/1000 [27:57<09:42, 1.87s/it, loss=0.333, lr=0.000132]\nSteps: 69%|██████▉ | 690/1000 [27:59<09:40, 1.87s/it, loss=0.333, lr=0.000132]\nSteps: 69%|██████▉ | 690/1000 [27:59<09:40, 1.87s/it, loss=0.755, lr=0.000131]\nSteps: 69%|██████▉ | 691/1000 [28:06<18:42, 3.63s/it, loss=0.755, lr=0.000131]\nSteps: 69%|██████▉ | 691/1000 [28:06<18:42, 3.63s/it, loss=0.336, lr=0.00013] \nSteps: 69%|██████▉ | 692/1000 [28:08<15:55, 3.10s/it, loss=0.336, lr=0.00013]\nSteps: 69%|██████▉ | 692/1000 [28:08<15:55, 3.10s/it, loss=0.546, lr=0.000129]\nSteps: 69%|██████▉ | 693/1000 [28:10<13:59, 2.73s/it, loss=0.546, lr=0.000129]\nSteps: 69%|██████▉ | 693/1000 [28:10<13:59, 2.73s/it, loss=0.302, lr=0.000129]\nSteps: 69%|██████▉ | 694/1000 [28:12<12:37, 2.48s/it, loss=0.302, lr=0.000129]\nSteps: 69%|██████▉ | 694/1000 [28:12<12:37, 2.48s/it, loss=0.268, lr=0.000128]\nSteps: 70%|██████▉ | 695/1000 [28:14<11:39, 2.29s/it, loss=0.268, lr=0.000128]\nSteps: 70%|██████▉ | 695/1000 [28:14<11:39, 2.29s/it, loss=0.666, lr=0.000127]\nSteps: 70%|██████▉ | 696/1000 [28:16<10:58, 2.17s/it, loss=0.666, lr=0.000127]\nSteps: 70%|██████▉ | 696/1000 [28:16<10:58, 2.17s/it, loss=0.302, lr=0.000126]\nSteps: 70%|██████▉ | 697/1000 [28:18<10:29, 2.08s/it, loss=0.302, lr=0.000126]\nSteps: 70%|██████▉ | 697/1000 [28:18<10:29, 2.08s/it, loss=0.704, lr=0.000126]\nSteps: 70%|██████▉ | 698/1000 [28:19<10:08, 2.02s/it, loss=0.704, lr=0.000126]\nSteps: 70%|██████▉ | 698/1000 [28:19<10:08, 2.02s/it, loss=0.329, lr=0.000125]\nSteps: 70%|██████▉ | 699/1000 [28:21<09:53, 1.97s/it, loss=0.329, lr=0.000125]\nSteps: 70%|██████▉ | 699/1000 [28:21<09:53, 1.97s/it, loss=0.309, lr=0.000124]\nSteps: 70%|███████ | 700/1000 [28:23<09:42, 1.94s/it, loss=0.309, lr=0.000124]\nSteps: 70%|███████ | 700/1000 [28:23<09:42, 1.94s/it, loss=0.715, lr=0.000123]\nSteps: 70%|███████ | 701/1000 [28:25<09:34, 1.92s/it, loss=0.715, lr=0.000123]\nSteps: 70%|███████ | 701/1000 [28:25<09:34, 1.92s/it, loss=0.756, lr=0.000123]\nSteps: 70%|███████ | 702/1000 [28:27<09:27, 1.90s/it, loss=0.756, lr=0.000123]\nSteps: 70%|███████ | 702/1000 [28:27<09:27, 1.90s/it, loss=0.805, lr=0.000122]\nSteps: 70%|███████ | 703/1000 [28:29<09:22, 1.89s/it, loss=0.805, lr=0.000122]\nSteps: 70%|███████ | 703/1000 [28:29<09:22, 1.89s/it, loss=0.541, lr=0.000121]\nSteps: 70%|███████ | 704/1000 [28:31<09:18, 1.89s/it, loss=0.541, lr=0.000121]\nSteps: 70%|███████ | 704/1000 [28:31<09:18, 1.89s/it, loss=0.535, lr=0.000121]\nSteps: 70%|███████ | 705/1000 [28:32<09:15, 1.88s/it, loss=0.535, lr=0.000121]\nSteps: 70%|███████ | 705/1000 [28:32<09:15, 1.88s/it, loss=0.319, lr=0.00012] \nSteps: 71%|███████ | 706/1000 [28:34<09:12, 1.88s/it, loss=0.319, lr=0.00012]\nSteps: 71%|███████ | 706/1000 [28:34<09:12, 1.88s/it, loss=0.533, lr=0.000119]\nSteps: 71%|███████ | 707/1000 [28:36<09:09, 1.88s/it, loss=0.533, lr=0.000119]\nSteps: 71%|███████ | 707/1000 [28:36<09:09, 1.88s/it, loss=0.304, lr=0.000118]\nSteps: 71%|███████ | 708/1000 [28:38<09:07, 1.88s/it, loss=0.304, lr=0.000118]\nSteps: 71%|███████ | 708/1000 [28:38<09:07, 1.88s/it, loss=0.296, lr=0.000118]\nSteps: 71%|███████ | 709/1000 [28:40<09:04, 1.87s/it, loss=0.296, lr=0.000118]\nSteps: 71%|███████ | 709/1000 [28:40<09:04, 1.87s/it, loss=0.554, lr=0.000117]\nSteps: 71%|███████ | 710/1000 [28:42<09:02, 1.87s/it, loss=0.554, lr=0.000117]\nSteps: 71%|███████ | 710/1000 [28:42<09:02, 1.87s/it, loss=0.404, lr=0.000116]\nSteps: 71%|███████ | 711/1000 [28:44<09:01, 1.87s/it, loss=0.404, lr=0.000116]\nSteps: 71%|███████ | 711/1000 [28:44<09:01, 1.87s/it, loss=0.494, lr=0.000116]\nSteps: 71%|███████ | 712/1000 [28:46<08:59, 1.87s/it, loss=0.494, lr=0.000116]\nSteps: 71%|███████ | 712/1000 [28:46<08:59, 1.87s/it, loss=0.575, lr=0.000115]\nSteps: 71%|███████▏ | 713/1000 [28:47<08:57, 1.87s/it, loss=0.575, lr=0.000115]\nSteps: 71%|███████▏ | 713/1000 [28:47<08:57, 1.87s/it, loss=0.809, lr=0.000114]\nSteps: 71%|███████▏ | 714/1000 [28:49<08:55, 1.87s/it, loss=0.809, lr=0.000114]\nSteps: 71%|███████▏ | 714/1000 [28:49<08:55, 1.87s/it, loss=0.242, lr=0.000113]\nSteps: 72%|███████▏ | 715/1000 [28:51<08:53, 1.87s/it, loss=0.242, lr=0.000113]\nSteps: 72%|███████▏ | 715/1000 [28:51<08:53, 1.87s/it, loss=0.922, lr=0.000113]\nSteps: 72%|███████▏ | 716/1000 [28:53<08:51, 1.87s/it, loss=0.922, lr=0.000113]\nSteps: 72%|███████▏ | 716/1000 [28:53<08:51, 1.87s/it, loss=0.338, lr=0.000112]\nSteps: 72%|███████▏ | 717/1000 [28:55<08:49, 1.87s/it, loss=0.338, lr=0.000112]\nSteps: 72%|███████▏ | 717/1000 [28:55<08:49, 1.87s/it, loss=0.4, lr=0.000111] \nSteps: 72%|███████▏ | 718/1000 [28:57<08:47, 1.87s/it, loss=0.4, lr=0.000111]\nSteps: 72%|███████▏ | 718/1000 [28:57<08:47, 1.87s/it, loss=0.472, lr=0.000111]\nSteps: 72%|███████▏ | 719/1000 [28:59<08:45, 1.87s/it, loss=0.472, lr=0.000111]\nSteps: 72%|███████▏ | 719/1000 [28:59<08:45, 1.87s/it, loss=0.346, lr=0.00011] \nSteps: 72%|███████▏ | 720/1000 [29:01<08:44, 1.87s/it, loss=0.346, lr=0.00011]\nSteps: 72%|███████▏ | 720/1000 [29:01<08:44, 1.87s/it, loss=0.278, lr=0.000109]\nSteps: 72%|███████▏ | 721/1000 [29:08<16:45, 3.61s/it, loss=0.278, lr=0.000109]\nSteps: 72%|███████▏ | 721/1000 [29:08<16:45, 3.61s/it, loss=0.41, lr=0.000109] \nSteps: 72%|███████▏ | 722/1000 [29:10<14:17, 3.08s/it, loss=0.41, lr=0.000109]\nSteps: 72%|███████▏ | 722/1000 [29:10<14:17, 3.08s/it, loss=0.684, lr=0.000108]\nSteps: 72%|███████▏ | 723/1000 [29:12<12:33, 2.72s/it, loss=0.684, lr=0.000108]\nSteps: 72%|███████▏ | 723/1000 [29:12<12:33, 2.72s/it, loss=0.397, lr=0.000107]\nSteps: 72%|███████▏ | 724/1000 [29:14<11:20, 2.46s/it, loss=0.397, lr=0.000107]\nSteps: 72%|███████▏ | 724/1000 [29:14<11:20, 2.46s/it, loss=0.553, lr=0.000106]\nSteps: 72%|███████▎ | 725/1000 [29:16<10:28, 2.29s/it, loss=0.553, lr=0.000106]\nSteps: 72%|███████▎ | 725/1000 [29:16<10:28, 2.29s/it, loss=0.656, lr=0.000106]\nSteps: 73%|███████▎ | 726/1000 [29:18<09:52, 2.16s/it, loss=0.656, lr=0.000106]\nSteps: 73%|███████▎ | 726/1000 [29:18<09:52, 2.16s/it, loss=0.394, lr=0.000105]\nSteps: 73%|███████▎ | 727/1000 [29:19<09:25, 2.07s/it, loss=0.394, lr=0.000105]\nSteps: 73%|███████▎ | 727/1000 [29:19<09:25, 2.07s/it, loss=0.329, lr=0.000104]\nSteps: 73%|███████▎ | 728/1000 [29:21<09:07, 2.01s/it, loss=0.329, lr=0.000104]\nSteps: 73%|███████▎ | 728/1000 [29:21<09:07, 2.01s/it, loss=0.849, lr=0.000104]\nSteps: 73%|███████▎ | 729/1000 [29:23<08:53, 1.97s/it, loss=0.849, lr=0.000104]\nSteps: 73%|███████▎ | 729/1000 [29:23<08:53, 1.97s/it, loss=0.514, lr=0.000103]\nSteps: 73%|███████▎ | 730/1000 [29:25<08:43, 1.94s/it, loss=0.514, lr=0.000103]\nSteps: 73%|███████▎ | 730/1000 [29:25<08:43, 1.94s/it, loss=0.35, lr=0.000102] \nSteps: 73%|███████▎ | 731/1000 [29:27<08:36, 1.92s/it, loss=0.35, lr=0.000102]\nSteps: 73%|███████▎ | 731/1000 [29:27<08:36, 1.92s/it, loss=0.565, lr=0.000102]\nSteps: 73%|███████▎ | 732/1000 [29:29<08:30, 1.91s/it, loss=0.565, lr=0.000102]\nSteps: 73%|███████▎ | 732/1000 [29:29<08:30, 1.91s/it, loss=0.907, lr=0.000101]\nSteps: 73%|███████▎ | 733/1000 [29:31<08:25, 1.90s/it, loss=0.907, lr=0.000101]\nSteps: 73%|███████▎ | 733/1000 [29:31<08:25, 1.90s/it, loss=0.68, lr=0.0001] \nSteps: 73%|███████▎ | 734/1000 [29:32<08:21, 1.89s/it, loss=0.68, lr=0.0001]\nSteps: 73%|███████▎ | 734/1000 [29:33<08:21, 1.89s/it, loss=0.374, lr=9.95e-5]\nSteps: 74%|███████▎ | 735/1000 [29:34<08:18, 1.88s/it, loss=0.374, lr=9.95e-5]\nSteps: 74%|███████▎ | 735/1000 [29:34<08:18, 1.88s/it, loss=0.32, lr=9.89e-5] \nSteps: 74%|███████▎ | 736/1000 [29:36<08:15, 1.88s/it, loss=0.32, lr=9.89e-5]\nSteps: 74%|███████▎ | 736/1000 [29:36<08:15, 1.88s/it, loss=0.331, lr=9.82e-5]\nSteps: 74%|███████▎ | 737/1000 [29:38<08:12, 1.87s/it, loss=0.331, lr=9.82e-5]\nSteps: 74%|███████▎ | 737/1000 [29:38<08:12, 1.87s/it, loss=0.579, lr=9.75e-5]\nSteps: 74%|███████▍ | 738/1000 [29:40<08:10, 1.87s/it, loss=0.579, lr=9.75e-5]\nSteps: 74%|███████▍ | 738/1000 [29:40<08:10, 1.87s/it, loss=0.369, lr=9.68e-5]\nSteps: 74%|███████▍ | 739/1000 [29:42<08:07, 1.87s/it, loss=0.369, lr=9.68e-5]\nSteps: 74%|███████▍ | 739/1000 [29:42<08:07, 1.87s/it, loss=0.469, lr=9.62e-5]\nSteps: 74%|███████▍ | 740/1000 [29:44<08:05, 1.87s/it, loss=0.469, lr=9.62e-5]\nSteps: 74%|███████▍ | 740/1000 [29:44<08:05, 1.87s/it, loss=0.932, lr=9.55e-5]\nSteps: 74%|███████▍ | 741/1000 [29:46<08:03, 1.87s/it, loss=0.932, lr=9.55e-5]\nSteps: 74%|███████▍ | 741/1000 [29:46<08:03, 1.87s/it, loss=0.518, lr=9.48e-5]\nSteps: 74%|███████▍ | 742/1000 [29:47<08:01, 1.87s/it, loss=0.518, lr=9.48e-5]\nSteps: 74%|███████▍ | 742/1000 [29:47<08:01, 1.87s/it, loss=0.301, lr=9.42e-5]\nSteps: 74%|███████▍ | 743/1000 [29:49<07:59, 1.87s/it, loss=0.301, lr=9.42e-5]\nSteps: 74%|███████▍ | 743/1000 [29:49<07:59, 1.87s/it, loss=0.681, lr=9.35e-5]\nSteps: 74%|███████▍ | 744/1000 [29:51<07:57, 1.87s/it, loss=0.681, lr=9.35e-5]\nSteps: 74%|███████▍ | 744/1000 [29:51<07:57, 1.87s/it, loss=0.229, lr=9.28e-5]\nSteps: 74%|███████▍ | 745/1000 [29:53<07:56, 1.87s/it, loss=0.229, lr=9.28e-5]\nSteps: 74%|███████▍ | 745/1000 [29:53<07:56, 1.87s/it, loss=0.42, lr=9.22e-5] \nSteps: 75%|███████▍ | 746/1000 [29:55<07:54, 1.87s/it, loss=0.42, lr=9.22e-5]\nSteps: 75%|███████▍ | 746/1000 [29:55<07:54, 1.87s/it, loss=0.654, lr=9.15e-5]\nSteps: 75%|███████▍ | 747/1000 [29:57<07:52, 1.87s/it, loss=0.654, lr=9.15e-5]\nSteps: 75%|███████▍ | 747/1000 [29:57<07:52, 1.87s/it, loss=0.484, lr=9.09e-5]\nSteps: 75%|███████▍ | 748/1000 [29:59<07:50, 1.87s/it, loss=0.484, lr=9.09e-5]\nSteps: 75%|███████▍ | 748/1000 [29:59<07:50, 1.87s/it, loss=0.28, lr=9.02e-5] \nSteps: 75%|███████▍ | 749/1000 [30:00<07:48, 1.87s/it, loss=0.28, lr=9.02e-5]\nSteps: 75%|███████▍ | 749/1000 [30:00<07:48, 1.87s/it, loss=0.429, lr=8.95e-5]\nSteps: 75%|███████▌ | 750/1000 [30:02<07:46, 1.87s/it, loss=0.429, lr=8.95e-5]\nSteps: 75%|███████▌ | 750/1000 [30:02<07:46, 1.87s/it, loss=0.43, lr=8.89e-5] \nSteps: 75%|███████▌ | 751/1000 [30:10<14:55, 3.59s/it, loss=0.43, lr=8.89e-5]\nSteps: 75%|███████▌ | 751/1000 [30:10<14:55, 3.59s/it, loss=0.583, lr=8.82e-5]\nSteps: 75%|███████▌ | 752/1000 [30:12<12:43, 3.08s/it, loss=0.583, lr=8.82e-5]\nSteps: 75%|███████▌ | 752/1000 [30:12<12:43, 3.08s/it, loss=0.377, lr=8.76e-5]\nSteps: 75%|███████▌ | 753/1000 [30:14<11:10, 2.72s/it, loss=0.377, lr=8.76e-5]\nSteps: 75%|███████▌ | 753/1000 [30:14<11:10, 2.72s/it, loss=0.544, lr=8.69e-5]\nSteps: 75%|███████▌ | 754/1000 [30:16<10:05, 2.46s/it, loss=0.544, lr=8.69e-5]\nSteps: 75%|███████▌ | 754/1000 [30:16<10:05, 2.46s/it, loss=0.281, lr=8.63e-5]\nSteps: 76%|███████▌ | 755/1000 [30:17<09:19, 2.28s/it, loss=0.281, lr=8.63e-5]\nSteps: 76%|███████▌ | 755/1000 [30:17<09:19, 2.28s/it, loss=1.04, lr=8.56e-5] \nSteps: 76%|███████▌ | 756/1000 [30:19<08:46, 2.16s/it, loss=1.04, lr=8.56e-5]\nSteps: 76%|███████▌ | 756/1000 [30:19<08:46, 2.16s/it, loss=0.313, lr=8.5e-5]\nSteps: 76%|███████▌ | 757/1000 [30:21<08:23, 2.07s/it, loss=0.313, lr=8.5e-5]\nSteps: 76%|███████▌ | 757/1000 [30:21<08:23, 2.07s/it, loss=0.71, lr=8.44e-5]\nSteps: 76%|███████▌ | 758/1000 [30:23<08:07, 2.01s/it, loss=0.71, lr=8.44e-5]\nSteps: 76%|███████▌ | 758/1000 [30:23<08:07, 2.01s/it, loss=0.784, lr=8.37e-5]\nSteps: 76%|███████▌ | 759/1000 [30:25<07:55, 1.97s/it, loss=0.784, lr=8.37e-5]\nSteps: 76%|███████▌ | 759/1000 [30:25<07:55, 1.97s/it, loss=0.59, lr=8.31e-5] \nSteps: 76%|███████▌ | 760/1000 [30:27<07:45, 1.94s/it, loss=0.59, lr=8.31e-5]\nSteps: 76%|███████▌ | 760/1000 [30:27<07:45, 1.94s/it, loss=0.81, lr=8.24e-5]\nSteps: 76%|███████▌ | 761/1000 [30:29<07:38, 1.92s/it, loss=0.81, lr=8.24e-5]\nSteps: 76%|███████▌ | 761/1000 [30:29<07:38, 1.92s/it, loss=0.661, lr=8.18e-5]\nSteps: 76%|███████▌ | 762/1000 [30:31<07:33, 1.90s/it, loss=0.661, lr=8.18e-5]\nSteps: 76%|███████▌ | 762/1000 [30:31<07:33, 1.90s/it, loss=0.452, lr=8.12e-5]\nSteps: 76%|███████▋ | 763/1000 [30:32<07:28, 1.89s/it, loss=0.452, lr=8.12e-5]\nSteps: 76%|███████▋ | 763/1000 [30:32<07:28, 1.89s/it, loss=0.422, lr=8.05e-5]\nSteps: 76%|███████▋ | 764/1000 [30:34<07:25, 1.89s/it, loss=0.422, lr=8.05e-5]\nSteps: 76%|███████▋ | 764/1000 [30:34<07:25, 1.89s/it, loss=0.53, lr=7.99e-5] \nSteps: 76%|███████▋ | 765/1000 [30:36<07:22, 1.88s/it, loss=0.53, lr=7.99e-5]\nSteps: 76%|███████▋ | 765/1000 [30:36<07:22, 1.88s/it, loss=0.322, lr=7.93e-5]\nSteps: 77%|███████▋ | 766/1000 [30:38<07:19, 1.88s/it, loss=0.322, lr=7.93e-5]\nSteps: 77%|███████▋ | 766/1000 [30:38<07:19, 1.88s/it, loss=0.702, lr=7.87e-5]\nSteps: 77%|███████▋ | 767/1000 [30:40<07:17, 1.88s/it, loss=0.702, lr=7.87e-5]\nSteps: 77%|███████▋ | 767/1000 [30:40<07:17, 1.88s/it, loss=0.364, lr=7.8e-5] \nSteps: 77%|███████▋ | 768/1000 [30:42<07:15, 1.88s/it, loss=0.364, lr=7.8e-5]\nSteps: 77%|███████▋ | 768/1000 [30:42<07:15, 1.88s/it, loss=0.327, lr=7.74e-5]\nSteps: 77%|███████▋ | 769/1000 [30:44<07:13, 1.88s/it, loss=0.327, lr=7.74e-5]\nSteps: 77%|███████▋ | 769/1000 [30:44<07:13, 1.88s/it, loss=0.331, lr=7.68e-5]\nSteps: 77%|███████▋ | 770/1000 [30:46<07:11, 1.87s/it, loss=0.331, lr=7.68e-5]\nSteps: 77%|███████▋ | 770/1000 [30:46<07:11, 1.87s/it, loss=0.455, lr=7.62e-5]\nSteps: 77%|███████▋ | 771/1000 [30:47<07:09, 1.88s/it, loss=0.455, lr=7.62e-5]\nSteps: 77%|███████▋ | 771/1000 [30:47<07:09, 1.88s/it, loss=0.404, lr=7.56e-5]\nSteps: 77%|███████▋ | 772/1000 [30:49<07:07, 1.87s/it, loss=0.404, lr=7.56e-5]\nSteps: 77%|███████▋ | 772/1000 [30:49<07:07, 1.87s/it, loss=1.01, lr=7.5e-5] \nSteps: 77%|███████▋ | 773/1000 [30:51<07:05, 1.87s/it, loss=1.01, lr=7.5e-5]\nSteps: 77%|███████▋ | 773/1000 [30:51<07:05, 1.87s/it, loss=0.762, lr=7.43e-5]\nSteps: 77%|███████▋ | 774/1000 [30:53<07:03, 1.88s/it, loss=0.762, lr=7.43e-5]\nSteps: 77%|███████▋ | 774/1000 [30:53<07:03, 1.88s/it, loss=0.371, lr=7.37e-5]\nSteps: 78%|███████▊ | 775/1000 [30:55<07:01, 1.88s/it, loss=0.371, lr=7.37e-5]\nSteps: 78%|███████▊ | 775/1000 [30:55<07:01, 1.88s/it, loss=0.976, lr=7.31e-5]\nSteps: 78%|███████▊ | 776/1000 [30:57<06:59, 1.87s/it, loss=0.976, lr=7.31e-5]\nSteps: 78%|███████▊ | 776/1000 [30:57<06:59, 1.87s/it, loss=0.998, lr=7.25e-5]\nSteps: 78%|███████▊ | 777/1000 [30:59<06:57, 1.87s/it, loss=0.998, lr=7.25e-5]\nSteps: 78%|███████▊ | 777/1000 [30:59<06:57, 1.87s/it, loss=0.697, lr=7.19e-5]\nSteps: 78%|███████▊ | 778/1000 [31:01<06:55, 1.87s/it, loss=0.697, lr=7.19e-5]\nSteps: 78%|███████▊ | 778/1000 [31:01<06:55, 1.87s/it, loss=0.637, lr=7.13e-5]\nSteps: 78%|███████▊ | 779/1000 [31:02<06:53, 1.87s/it, loss=0.637, lr=7.13e-5]\nSteps: 78%|███████▊ | 779/1000 [31:02<06:53, 1.87s/it, loss=0.719, lr=7.07e-5]\nSteps: 78%|███████▊ | 780/1000 [31:04<06:52, 1.87s/it, loss=0.719, lr=7.07e-5]\nSteps: 78%|███████▊ | 780/1000 [31:04<06:52, 1.87s/it, loss=0.402, lr=7.01e-5]\nSteps: 78%|███████▊ | 781/1000 [31:12<13:11, 3.61s/it, loss=0.402, lr=7.01e-5]\nSteps: 78%|███████▊ | 781/1000 [31:12<13:11, 3.61s/it, loss=0.353, lr=6.95e-5]\nSteps: 78%|███████▊ | 782/1000 [31:14<11:13, 3.09s/it, loss=0.353, lr=6.95e-5]\nSteps: 78%|███████▊ | 782/1000 [31:14<11:13, 3.09s/it, loss=0.519, lr=6.89e-5]\nSteps: 78%|███████▊ | 783/1000 [31:16<09:51, 2.72s/it, loss=0.519, lr=6.89e-5]\nSteps: 78%|███████▊ | 783/1000 [31:16<09:51, 2.72s/it, loss=0.33, lr=6.83e-5] \nSteps: 78%|███████▊ | 784/1000 [31:18<08:53, 2.47s/it, loss=0.33, lr=6.83e-5]\nSteps: 78%|███████▊ | 784/1000 [31:18<08:53, 2.47s/it, loss=0.923, lr=6.77e-5]\nSteps: 78%|███████▊ | 785/1000 [31:19<08:12, 2.29s/it, loss=0.923, lr=6.77e-5]\nSteps: 78%|███████▊ | 785/1000 [31:19<08:12, 2.29s/it, loss=0.756, lr=6.71e-5]\nSteps: 79%|███████▊ | 786/1000 [31:21<07:42, 2.16s/it, loss=0.756, lr=6.71e-5]\nSteps: 79%|███████▊ | 786/1000 [31:21<07:42, 2.16s/it, loss=0.279, lr=6.66e-5]\nSteps: 79%|███████▊ | 787/1000 [31:23<07:22, 2.08s/it, loss=0.279, lr=6.66e-5]\nSteps: 79%|███████▊ | 787/1000 [31:23<07:22, 2.08s/it, loss=0.66, lr=6.6e-5] \nSteps: 79%|███████▉ | 788/1000 [31:25<07:07, 2.01s/it, loss=0.66, lr=6.6e-5]\nSteps: 79%|███████▉ | 788/1000 [31:25<07:07, 2.01s/it, loss=0.417, lr=6.54e-5]\nSteps: 79%|███████▉ | 789/1000 [31:27<06:56, 1.97s/it, loss=0.417, lr=6.54e-5]\nSteps: 79%|███████▉ | 789/1000 [31:27<06:56, 1.97s/it, loss=0.785, lr=6.48e-5]\nSteps: 79%|███████▉ | 790/1000 [31:29<06:48, 1.94s/it, loss=0.785, lr=6.48e-5]\nSteps: 79%|███████▉ | 790/1000 [31:29<06:48, 1.94s/it, loss=0.428, lr=6.42e-5]\nSteps: 79%|███████▉ | 791/1000 [31:31<06:42, 1.92s/it, loss=0.428, lr=6.42e-5]\nSteps: 79%|███████▉ | 791/1000 [31:31<06:42, 1.92s/it, loss=0.274, lr=6.37e-5]\nSteps: 79%|███████▉ | 792/1000 [31:33<06:36, 1.91s/it, loss=0.274, lr=6.37e-5]\nSteps: 79%|███████▉ | 792/1000 [31:33<06:36, 1.91s/it, loss=0.798, lr=6.31e-5]\nSteps: 79%|███████▉ | 793/1000 [31:34<06:32, 1.90s/it, loss=0.798, lr=6.31e-5]\nSteps: 79%|███████▉ | 793/1000 [31:34<06:32, 1.90s/it, loss=0.288, lr=6.25e-5]\nSteps: 79%|███████▉ | 794/1000 [31:36<06:28, 1.89s/it, loss=0.288, lr=6.25e-5]\nSteps: 79%|███████▉ | 794/1000 [31:36<06:28, 1.89s/it, loss=0.728, lr=6.19e-5]\nSteps: 80%|███████▉ | 795/1000 [31:38<06:26, 1.88s/it, loss=0.728, lr=6.19e-5]\nSteps: 80%|███████▉ | 795/1000 [31:38<06:26, 1.88s/it, loss=0.617, lr=6.14e-5]\nSteps: 80%|███████▉ | 796/1000 [31:40<06:23, 1.88s/it, loss=0.617, lr=6.14e-5]\nSteps: 80%|███████▉ | 796/1000 [31:40<06:23, 1.88s/it, loss=0.68, lr=6.08e-5] \nSteps: 80%|███████▉ | 797/1000 [31:42<06:21, 1.88s/it, loss=0.68, lr=6.08e-5]\nSteps: 80%|███████▉ | 797/1000 [31:42<06:21, 1.88s/it, loss=0.577, lr=6.03e-5]\nSteps: 80%|███████▉ | 798/1000 [31:44<06:19, 1.88s/it, loss=0.577, lr=6.03e-5]\nSteps: 80%|███████▉ | 798/1000 [31:44<06:19, 1.88s/it, loss=0.427, lr=5.97e-5]\nSteps: 80%|███████▉ | 799/1000 [31:46<06:17, 1.88s/it, loss=0.427, lr=5.97e-5]\nSteps: 80%|███████▉ | 799/1000 [31:46<06:17, 1.88s/it, loss=0.317, lr=5.91e-5]\nSteps: 80%|████████ | 800/1000 [31:48<06:15, 1.88s/it, loss=0.317, lr=5.91e-5]\nSteps: 80%|████████ | 800/1000 [31:48<06:15, 1.88s/it, loss=0.501, lr=5.86e-5]\nSteps: 80%|████████ | 801/1000 [31:49<06:13, 1.87s/it, loss=0.501, lr=5.86e-5]\nSteps: 80%|████████ | 801/1000 [31:49<06:13, 1.87s/it, loss=0.288, lr=5.8e-5] \nSteps: 80%|████████ | 802/1000 [31:51<06:11, 1.87s/it, loss=0.288, lr=5.8e-5]\nSteps: 80%|████████ | 802/1000 [31:51<06:11, 1.87s/it, loss=0.983, lr=5.75e-5]\nSteps: 80%|████████ | 803/1000 [31:53<06:09, 1.87s/it, loss=0.983, lr=5.75e-5]\nSteps: 80%|████████ | 803/1000 [31:53<06:09, 1.87s/it, loss=0.357, lr=5.69e-5]\nSteps: 80%|████████ | 804/1000 [31:55<06:07, 1.87s/it, loss=0.357, lr=5.69e-5]\nSteps: 80%|████████ | 804/1000 [31:55<06:07, 1.87s/it, loss=0.227, lr=5.64e-5]\nSteps: 80%|████████ | 805/1000 [31:57<06:05, 1.87s/it, loss=0.227, lr=5.64e-5]\nSteps: 80%|████████ | 805/1000 [31:57<06:05, 1.87s/it, loss=0.467, lr=5.58e-5]\nSteps: 81%|████████ | 806/1000 [31:59<06:03, 1.87s/it, loss=0.467, lr=5.58e-5]\nSteps: 81%|████████ | 806/1000 [31:59<06:03, 1.87s/it, loss=0.31, lr=5.53e-5] \nSteps: 81%|████████ | 807/1000 [32:01<06:01, 1.87s/it, loss=0.31, lr=5.53e-5]\nSteps: 81%|████████ | 807/1000 [32:01<06:01, 1.87s/it, loss=0.681, lr=5.47e-5]\nSteps: 81%|████████ | 808/1000 [32:02<05:59, 1.87s/it, loss=0.681, lr=5.47e-5]\nSteps: 81%|████████ | 808/1000 [32:03<05:59, 1.87s/it, loss=0.285, lr=5.42e-5]\nSteps: 81%|████████ | 809/1000 [32:04<05:58, 1.87s/it, loss=0.285, lr=5.42e-5]\nSteps: 81%|████████ | 809/1000 [32:04<05:58, 1.87s/it, loss=0.362, lr=5.37e-5]\nSteps: 81%|████████ | 810/1000 [32:06<05:56, 1.88s/it, loss=0.362, lr=5.37e-5]\nSteps: 81%|████████ | 810/1000 [32:06<05:56, 1.88s/it, loss=0.746, lr=5.31e-5]\nSteps: 81%|████████ | 811/1000 [32:14<11:29, 3.65s/it, loss=0.746, lr=5.31e-5]\nSteps: 81%|████████ | 811/1000 [32:14<11:29, 3.65s/it, loss=1.01, lr=5.26e-5] \nSteps: 81%|████████ | 812/1000 [32:16<09:45, 3.11s/it, loss=1.01, lr=5.26e-5]\nSteps: 81%|████████ | 812/1000 [32:16<09:45, 3.11s/it, loss=0.306, lr=5.21e-5]\nSteps: 81%|████████▏ | 813/1000 [32:18<08:32, 2.74s/it, loss=0.306, lr=5.21e-5]\nSteps: 81%|████████▏ | 813/1000 [32:18<08:32, 2.74s/it, loss=0.647, lr=5.15e-5]\nSteps: 81%|████████▏ | 814/1000 [32:20<07:41, 2.48s/it, loss=0.647, lr=5.15e-5]\nSteps: 81%|████████▏ | 814/1000 [32:20<07:41, 2.48s/it, loss=0.805, lr=5.1e-5] \nSteps: 82%|████████▏ | 815/1000 [32:22<07:04, 2.30s/it, loss=0.805, lr=5.1e-5]\nSteps: 82%|████████▏ | 815/1000 [32:22<07:04, 2.30s/it, loss=0.457, lr=5.05e-5]\nSteps: 82%|████████▏ | 816/1000 [32:23<06:39, 2.17s/it, loss=0.457, lr=5.05e-5]\nSteps: 82%|████████▏ | 816/1000 [32:23<06:39, 2.17s/it, loss=0.582, lr=5e-5] \nSteps: 82%|████████▏ | 817/1000 [32:25<06:20, 2.08s/it, loss=0.582, lr=5e-5]\nSteps: 82%|████████▏ | 817/1000 [32:25<06:20, 2.08s/it, loss=0.313, lr=4.95e-5]\nSteps: 82%|████████▏ | 818/1000 [32:27<06:07, 2.02s/it, loss=0.313, lr=4.95e-5]\nSteps: 82%|████████▏ | 818/1000 [32:27<06:07, 2.02s/it, loss=0.524, lr=4.89e-5]\nSteps: 82%|████████▏ | 819/1000 [32:29<05:57, 1.97s/it, loss=0.524, lr=4.89e-5]\nSteps: 82%|████████▏ | 819/1000 [32:29<05:57, 1.97s/it, loss=0.812, lr=4.84e-5]\nSteps: 82%|████████▏ | 820/1000 [32:31<05:49, 1.94s/it, loss=0.812, lr=4.84e-5]\nSteps: 82%|████████▏ | 820/1000 [32:31<05:49, 1.94s/it, loss=0.401, lr=4.79e-5]\nSteps: 82%|████████▏ | 821/1000 [32:33<05:43, 1.92s/it, loss=0.401, lr=4.79e-5]\nSteps: 82%|████████▏ | 821/1000 [32:33<05:43, 1.92s/it, loss=0.325, lr=4.74e-5]\nSteps: 82%|████████▏ | 822/1000 [32:35<05:39, 1.91s/it, loss=0.325, lr=4.74e-5]\nSteps: 82%|████████▏ | 822/1000 [32:35<05:39, 1.91s/it, loss=0.639, lr=4.69e-5]\nSteps: 82%|████████▏ | 823/1000 [32:36<05:35, 1.89s/it, loss=0.639, lr=4.69e-5]\nSteps: 82%|████████▏ | 823/1000 [32:36<05:35, 1.89s/it, loss=0.799, lr=4.64e-5]\nSteps: 82%|████████▏ | 824/1000 [32:38<05:32, 1.89s/it, loss=0.799, lr=4.64e-5]\nSteps: 82%|████████▏ | 824/1000 [32:38<05:32, 1.89s/it, loss=0.505, lr=4.59e-5]\nSteps: 82%|████████▎ | 825/1000 [32:40<05:29, 1.88s/it, loss=0.505, lr=4.59e-5]\nSteps: 82%|████████▎ | 825/1000 [32:40<05:29, 1.88s/it, loss=0.998, lr=4.54e-5]\nSteps: 83%|████████▎ | 826/1000 [32:42<05:27, 1.88s/it, loss=0.998, lr=4.54e-5]\nSteps: 83%|████████▎ | 826/1000 [32:42<05:27, 1.88s/it, loss=0.37, lr=4.49e-5] \nSteps: 83%|████████▎ | 827/1000 [32:44<05:24, 1.88s/it, loss=0.37, lr=4.49e-5]\nSteps: 83%|████████▎ | 827/1000 [32:44<05:24, 1.88s/it, loss=0.67, lr=4.44e-5]\nSteps: 83%|████████▎ | 828/1000 [32:46<05:22, 1.88s/it, loss=0.67, lr=4.44e-5]\nSteps: 83%|████████▎ | 828/1000 [32:46<05:22, 1.88s/it, loss=0.298, lr=4.39e-5]\nSteps: 83%|████████▎ | 829/1000 [32:48<05:20, 1.87s/it, loss=0.298, lr=4.39e-5]\nSteps: 83%|████████▎ | 829/1000 [32:48<05:20, 1.87s/it, loss=0.783, lr=4.34e-5]\nSteps: 83%|████████▎ | 830/1000 [32:50<05:18, 1.87s/it, loss=0.783, lr=4.34e-5]\nSteps: 83%|████████▎ | 830/1000 [32:50<05:18, 1.87s/it, loss=0.355, lr=4.29e-5]\nSteps: 83%|████████▎ | 831/1000 [32:51<05:16, 1.87s/it, loss=0.355, lr=4.29e-5]\nSteps: 83%|████████▎ | 831/1000 [32:51<05:16, 1.87s/it, loss=0.796, lr=4.25e-5]\nSteps: 83%|████████▎ | 832/1000 [32:53<05:14, 1.87s/it, loss=0.796, lr=4.25e-5]\nSteps: 83%|████████▎ | 832/1000 [32:53<05:14, 1.87s/it, loss=0.467, lr=4.2e-5] \nSteps: 83%|████████▎ | 833/1000 [32:55<05:12, 1.87s/it, loss=0.467, lr=4.2e-5]\nSteps: 83%|████████▎ | 833/1000 [32:55<05:12, 1.87s/it, loss=0.29, lr=4.15e-5]\nSteps: 83%|████████▎ | 834/1000 [32:57<05:10, 1.87s/it, loss=0.29, lr=4.15e-5]\nSteps: 83%|████████▎ | 834/1000 [32:57<05:10, 1.87s/it, loss=0.222, lr=4.1e-5]\nSteps: 84%|████████▎ | 835/1000 [32:59<05:08, 1.87s/it, loss=0.222, lr=4.1e-5]\nSteps: 84%|████████▎ | 835/1000 [32:59<05:08, 1.87s/it, loss=0.359, lr=4.05e-5]\nSteps: 84%|████████▎ | 836/1000 [33:01<05:06, 1.87s/it, loss=0.359, lr=4.05e-5]\nSteps: 84%|████████▎ | 836/1000 [33:01<05:06, 1.87s/it, loss=0.51, lr=4.01e-5] \nSteps: 84%|████████▎ | 837/1000 [33:03<05:04, 1.87s/it, loss=0.51, lr=4.01e-5]\nSteps: 84%|████████▎ | 837/1000 [33:03<05:04, 1.87s/it, loss=0.674, lr=3.96e-5]\nSteps: 84%|████████▍ | 838/1000 [33:05<05:02, 1.87s/it, loss=0.674, lr=3.96e-5]\nSteps: 84%|████████▍ | 838/1000 [33:05<05:02, 1.87s/it, loss=0.796, lr=3.91e-5]\nSteps: 84%|████████▍ | 839/1000 [33:06<05:00, 1.87s/it, loss=0.796, lr=3.91e-5]\nSteps: 84%|████████▍ | 839/1000 [33:06<05:00, 1.87s/it, loss=0.477, lr=3.87e-5]\nSteps: 84%|████████▍ | 840/1000 [33:08<04:58, 1.87s/it, loss=0.477, lr=3.87e-5]\nSteps: 84%|████████▍ | 840/1000 [33:08<04:58, 1.87s/it, loss=0.394, lr=3.82e-5]\nSteps: 84%|████████▍ | 841/1000 [33:16<09:41, 3.66s/it, loss=0.394, lr=3.82e-5]\nSteps: 84%|████████▍ | 841/1000 [33:16<09:41, 3.66s/it, loss=0.318, lr=3.77e-5]\nSteps: 84%|████████▍ | 842/1000 [33:18<08:12, 3.12s/it, loss=0.318, lr=3.77e-5]\nSteps: 84%|████████▍ | 842/1000 [33:18<08:12, 3.12s/it, loss=0.402, lr=3.73e-5]\nSteps: 84%|████████▍ | 843/1000 [33:20<07:11, 2.75s/it, loss=0.402, lr=3.73e-5]\nSteps: 84%|████████▍ | 843/1000 [33:20<07:11, 2.75s/it, loss=0.834, lr=3.68e-5]\nSteps: 84%|████████▍ | 844/1000 [33:22<06:27, 2.49s/it, loss=0.834, lr=3.68e-5]\nSteps: 84%|████████▍ | 844/1000 [33:22<06:27, 2.49s/it, loss=0.346, lr=3.64e-5]\nSteps: 84%|████████▍ | 845/1000 [33:24<05:56, 2.30s/it, loss=0.346, lr=3.64e-5]\nSteps: 84%|████████▍ | 845/1000 [33:24<05:56, 2.30s/it, loss=0.486, lr=3.59e-5]\nSteps: 85%|████████▍ | 846/1000 [33:25<05:34, 2.17s/it, loss=0.486, lr=3.59e-5]\nSteps: 85%|████████▍ | 846/1000 [33:25<05:34, 2.17s/it, loss=0.326, lr=3.55e-5]\nSteps: 85%|████████▍ | 847/1000 [33:27<05:18, 2.08s/it, loss=0.326, lr=3.55e-5]\nSteps: 85%|████████▍ | 847/1000 [33:27<05:18, 2.08s/it, loss=0.328, lr=3.5e-5] \nSteps: 85%|████████▍ | 848/1000 [33:29<05:06, 2.02s/it, loss=0.328, lr=3.5e-5]\nSteps: 85%|████████▍ | 848/1000 [33:29<05:06, 2.02s/it, loss=0.697, lr=3.46e-5]\nSteps: 85%|████████▍ | 849/1000 [33:31<04:58, 1.97s/it, loss=0.697, lr=3.46e-5]\nSteps: 85%|████████▍ | 849/1000 [33:31<04:58, 1.97s/it, loss=0.375, lr=3.41e-5]\nSteps: 85%|████████▌ | 850/1000 [33:33<04:51, 1.94s/it, loss=0.375, lr=3.41e-5]\nSteps: 85%|████████▌ | 850/1000 [33:33<04:51, 1.94s/it, loss=0.996, lr=3.37e-5]\nSteps: 85%|████████▌ | 851/1000 [33:35<04:46, 1.92s/it, loss=0.996, lr=3.37e-5]\nSteps: 85%|████████▌ | 851/1000 [33:35<04:46, 1.92s/it, loss=0.817, lr=3.33e-5]\nSteps: 85%|████████▌ | 852/1000 [33:37<04:42, 1.91s/it, loss=0.817, lr=3.33e-5]\nSteps: 85%|████████▌ | 852/1000 [33:37<04:42, 1.91s/it, loss=0.285, lr=3.28e-5]\nSteps: 85%|████████▌ | 853/1000 [33:39<04:38, 1.90s/it, loss=0.285, lr=3.28e-5]\nSteps: 85%|████████▌ | 853/1000 [33:39<04:38, 1.90s/it, loss=0.641, lr=3.24e-5]\nSteps: 85%|████████▌ | 854/1000 [33:40<04:35, 1.89s/it, loss=0.641, lr=3.24e-5]\nSteps: 85%|████████▌ | 854/1000 [33:40<04:35, 1.89s/it, loss=0.678, lr=3.2e-5] \nSteps: 86%|████████▌ | 855/1000 [33:42<04:33, 1.89s/it, loss=0.678, lr=3.2e-5]\nSteps: 86%|████████▌ | 855/1000 [33:42<04:33, 1.89s/it, loss=0.953, lr=3.16e-5]\nSteps: 86%|████████▌ | 856/1000 [33:44<04:31, 1.88s/it, loss=0.953, lr=3.16e-5]\nSteps: 86%|████████▌ | 856/1000 [33:44<04:31, 1.88s/it, loss=0.33, lr=3.11e-5] \nSteps: 86%|████████▌ | 857/1000 [33:46<04:28, 1.88s/it, loss=0.33, lr=3.11e-5]\nSteps: 86%|████████▌ | 857/1000 [33:46<04:28, 1.88s/it, loss=0.782, lr=3.07e-5]\nSteps: 86%|████████▌ | 858/1000 [33:48<04:26, 1.88s/it, loss=0.782, lr=3.07e-5]\nSteps: 86%|████████▌ | 858/1000 [33:48<04:26, 1.88s/it, loss=0.652, lr=3.03e-5]\nSteps: 86%|████████▌ | 859/1000 [33:50<04:24, 1.88s/it, loss=0.652, lr=3.03e-5]\nSteps: 86%|████████▌ | 859/1000 [33:50<04:24, 1.88s/it, loss=0.55, lr=2.99e-5] \nSteps: 86%|████████▌ | 860/1000 [33:52<04:22, 1.88s/it, loss=0.55, lr=2.99e-5]\nSteps: 86%|████████▌ | 860/1000 [33:52<04:22, 1.88s/it, loss=0.467, lr=2.95e-5]\nSteps: 86%|████████▌ | 861/1000 [33:54<04:20, 1.88s/it, loss=0.467, lr=2.95e-5]\nSteps: 86%|████████▌ | 861/1000 [33:54<04:20, 1.88s/it, loss=0.636, lr=2.91e-5]\nSteps: 86%|████████▌ | 862/1000 [33:55<04:18, 1.87s/it, loss=0.636, lr=2.91e-5]\nSteps: 86%|████████▌ | 862/1000 [33:55<04:18, 1.87s/it, loss=0.502, lr=2.87e-5]\nSteps: 86%|████████▋ | 863/1000 [33:57<04:16, 1.87s/it, loss=0.502, lr=2.87e-5]\nSteps: 86%|████████▋ | 863/1000 [33:57<04:16, 1.87s/it, loss=0.29, lr=2.83e-5] \nSteps: 86%|████████▋ | 864/1000 [33:59<04:14, 1.87s/it, loss=0.29, lr=2.83e-5]\nSteps: 86%|████████▋ | 864/1000 [33:59<04:14, 1.87s/it, loss=0.379, lr=2.79e-5]\nSteps: 86%|████████▋ | 865/1000 [34:01<04:12, 1.87s/it, loss=0.379, lr=2.79e-5]\nSteps: 86%|████████▋ | 865/1000 [34:01<04:12, 1.87s/it, loss=0.47, lr=2.75e-5] \nSteps: 87%|████████▋ | 866/1000 [34:03<04:10, 1.87s/it, loss=0.47, lr=2.75e-5]\nSteps: 87%|████████▋ | 866/1000 [34:03<04:10, 1.87s/it, loss=0.333, lr=2.71e-5]\nSteps: 87%|████████▋ | 867/1000 [34:05<04:09, 1.87s/it, loss=0.333, lr=2.71e-5]\nSteps: 87%|████████▋ | 867/1000 [34:05<04:09, 1.87s/it, loss=0.916, lr=2.67e-5]\nSteps: 87%|████████▋ | 868/1000 [34:07<04:07, 1.87s/it, loss=0.916, lr=2.67e-5]\nSteps: 87%|████████▋ | 868/1000 [34:07<04:07, 1.87s/it, loss=0.406, lr=2.63e-5]\nSteps: 87%|████████▋ | 869/1000 [34:09<04:05, 1.87s/it, loss=0.406, lr=2.63e-5]\nSteps: 87%|████████▋ | 869/1000 [34:09<04:05, 1.87s/it, loss=0.387, lr=2.59e-5]\nSteps: 87%|████████▋ | 870/1000 [34:10<04:03, 1.87s/it, loss=0.387, lr=2.59e-5]\nSteps: 87%|████████▋ | 870/1000 [34:10<04:03, 1.87s/it, loss=0.272, lr=2.55e-5]\nSteps: 87%|████████▋ | 871/1000 [34:18<07:45, 3.61s/it, loss=0.272, lr=2.55e-5]\nSteps: 87%|████████▋ | 871/1000 [34:18<07:45, 3.61s/it, loss=0.311, lr=2.51e-5]\nSteps: 87%|████████▋ | 872/1000 [34:20<06:35, 3.09s/it, loss=0.311, lr=2.51e-5]\nSteps: 87%|████████▋ | 872/1000 [34:20<06:35, 3.09s/it, loss=0.616, lr=2.47e-5]\nSteps: 87%|████████▋ | 873/1000 [34:22<05:45, 2.72s/it, loss=0.616, lr=2.47e-5]\nSteps: 87%|████████▋ | 873/1000 [34:22<05:45, 2.72s/it, loss=0.909, lr=2.44e-5]\nSteps: 87%|████████▋ | 874/1000 [34:24<05:10, 2.47s/it, loss=0.909, lr=2.44e-5]\nSteps: 87%|████████▋ | 874/1000 [34:24<05:10, 2.47s/it, loss=0.92, lr=2.4e-5] \nSteps: 88%|████████▊ | 875/1000 [34:26<04:45, 2.29s/it, loss=0.92, lr=2.4e-5]\nSteps: 88%|████████▊ | 875/1000 [34:26<04:45, 2.29s/it, loss=0.308, lr=2.36e-5]\nSteps: 88%|████████▊ | 876/1000 [34:27<04:28, 2.16s/it, loss=0.308, lr=2.36e-5]\nSteps: 88%|████████▊ | 876/1000 [34:27<04:28, 2.16s/it, loss=0.602, lr=2.32e-5]\nSteps: 88%|████████▊ | 877/1000 [34:29<04:15, 2.07s/it, loss=0.602, lr=2.32e-5]\nSteps: 88%|████████▊ | 877/1000 [34:29<04:15, 2.07s/it, loss=0.335, lr=2.29e-5]\nSteps: 88%|████████▊ | 878/1000 [34:31<04:05, 2.01s/it, loss=0.335, lr=2.29e-5]\nSteps: 88%|████████▊ | 878/1000 [34:31<04:05, 2.01s/it, loss=0.42, lr=2.25e-5] \nSteps: 88%|████████▊ | 879/1000 [34:33<03:58, 1.97s/it, loss=0.42, lr=2.25e-5]\nSteps: 88%|████████▊ | 879/1000 [34:33<03:58, 1.97s/it, loss=0.296, lr=2.22e-5]\nSteps: 88%|████████▊ | 880/1000 [34:35<03:52, 1.94s/it, loss=0.296, lr=2.22e-5]\nSteps: 88%|████████▊ | 880/1000 [34:35<03:52, 1.94s/it, loss=0.369, lr=2.18e-5]\nSteps: 88%|████████▊ | 881/1000 [34:37<03:48, 1.92s/it, loss=0.369, lr=2.18e-5]\nSteps: 88%|████████▊ | 881/1000 [34:37<03:48, 1.92s/it, loss=0.855, lr=2.14e-5]\nSteps: 88%|████████▊ | 882/1000 [34:39<03:44, 1.90s/it, loss=0.855, lr=2.14e-5]\nSteps: 88%|████████▊ | 882/1000 [34:39<03:44, 1.90s/it, loss=0.897, lr=2.11e-5]\nSteps: 88%|████████▊ | 883/1000 [34:41<03:41, 1.90s/it, loss=0.897, lr=2.11e-5]\nSteps: 88%|████████▊ | 883/1000 [34:41<03:41, 1.90s/it, loss=0.313, lr=2.07e-5]\nSteps: 88%|████████▊ | 884/1000 [34:42<03:39, 1.89s/it, loss=0.313, lr=2.07e-5]\nSteps: 88%|████████▊ | 884/1000 [34:42<03:39, 1.89s/it, loss=0.438, lr=2.04e-5]\nSteps: 88%|████████▊ | 885/1000 [34:44<03:36, 1.88s/it, loss=0.438, lr=2.04e-5]\nSteps: 88%|████████▊ | 885/1000 [34:44<03:36, 1.88s/it, loss=1.02, lr=2.01e-5] \nSteps: 89%|████████▊ | 886/1000 [34:46<03:34, 1.88s/it, loss=1.02, lr=2.01e-5]\nSteps: 89%|████████▊ | 886/1000 [34:46<03:34, 1.88s/it, loss=1.03, lr=1.97e-5]\nSteps: 89%|████████▊ | 887/1000 [34:48<03:32, 1.88s/it, loss=1.03, lr=1.97e-5]\nSteps: 89%|████████▊ | 887/1000 [34:48<03:32, 1.88s/it, loss=0.438, lr=1.94e-5]\nSteps: 89%|████████▉ | 888/1000 [34:50<03:30, 1.88s/it, loss=0.438, lr=1.94e-5]\nSteps: 89%|████████▉ | 888/1000 [34:50<03:30, 1.88s/it, loss=0.478, lr=1.9e-5] \nSteps: 89%|████████▉ | 889/1000 [34:52<03:28, 1.88s/it, loss=0.478, lr=1.9e-5]\nSteps: 89%|████████▉ | 889/1000 [34:52<03:28, 1.88s/it, loss=0.345, lr=1.87e-5]\nSteps: 89%|████████▉ | 890/1000 [34:54<03:26, 1.87s/it, loss=0.345, lr=1.87e-5]\nSteps: 89%|████████▉ | 890/1000 [34:54<03:26, 1.87s/it, loss=0.646, lr=1.84e-5]\nSteps: 89%|████████▉ | 891/1000 [34:55<03:24, 1.87s/it, loss=0.646, lr=1.84e-5]\nSteps: 89%|████████▉ | 891/1000 [34:56<03:24, 1.87s/it, loss=0.328, lr=1.8e-5] \nSteps: 89%|████████▉ | 892/1000 [34:57<03:22, 1.87s/it, loss=0.328, lr=1.8e-5]\nSteps: 89%|████████▉ | 892/1000 [34:57<03:22, 1.87s/it, loss=0.561, lr=1.77e-5]\nSteps: 89%|████████▉ | 893/1000 [34:59<03:20, 1.87s/it, loss=0.561, lr=1.77e-5]\nSteps: 89%|████████▉ | 893/1000 [34:59<03:20, 1.87s/it, loss=0.326, lr=1.74e-5]\nSteps: 89%|████████▉ | 894/1000 [35:01<03:18, 1.87s/it, loss=0.326, lr=1.74e-5]\nSteps: 89%|████████▉ | 894/1000 [35:01<03:18, 1.87s/it, loss=0.299, lr=1.71e-5]\nSteps: 90%|████████▉ | 895/1000 [35:03<03:16, 1.87s/it, loss=0.299, lr=1.71e-5]\nSteps: 90%|████████▉ | 895/1000 [35:03<03:16, 1.87s/it, loss=0.333, lr=1.68e-5]\nSteps: 90%|████████▉ | 896/1000 [35:05<03:14, 1.87s/it, loss=0.333, lr=1.68e-5]\nSteps: 90%|████████▉ | 896/1000 [35:05<03:14, 1.87s/it, loss=0.362, lr=1.64e-5]\nSteps: 90%|████████▉ | 897/1000 [35:07<03:12, 1.87s/it, loss=0.362, lr=1.64e-5]\nSteps: 90%|████████▉ | 897/1000 [35:07<03:12, 1.87s/it, loss=0.921, lr=1.61e-5]\nSteps: 90%|████████▉ | 898/1000 [35:09<03:11, 1.87s/it, loss=0.921, lr=1.61e-5]\nSteps: 90%|████████▉ | 898/1000 [35:09<03:11, 1.87s/it, loss=0.953, lr=1.58e-5]\nSteps: 90%|████████▉ | 899/1000 [35:10<03:09, 1.87s/it, loss=0.953, lr=1.58e-5]\nSteps: 90%|████████▉ | 899/1000 [35:10<03:09, 1.87s/it, loss=0.381, lr=1.55e-5]\nSteps: 90%|█████████ | 900/1000 [35:12<03:07, 1.87s/it, loss=0.381, lr=1.55e-5]\nSteps: 90%|█████████ | 900/1000 [35:12<03:07, 1.87s/it, loss=0.381, lr=1.52e-5]\nSteps: 90%|█████████ | 901/1000 [35:20<06:00, 3.64s/it, loss=0.381, lr=1.52e-5]\nSteps: 90%|█████████ | 901/1000 [35:20<06:00, 3.64s/it, loss=0.543, lr=1.49e-5]\nSteps: 90%|█████████ | 902/1000 [35:22<05:04, 3.11s/it, loss=0.543, lr=1.49e-5]\nSteps: 90%|█████████ | 902/1000 [35:22<05:04, 3.11s/it, loss=0.503, lr=1.46e-5]\nSteps: 90%|█████████ | 903/1000 [35:24<04:25, 2.74s/it, loss=0.503, lr=1.46e-5]\nSteps: 90%|█████████ | 903/1000 [35:24<04:25, 2.74s/it, loss=0.302, lr=1.43e-5]\nSteps: 90%|█████████ | 904/1000 [35:26<03:57, 2.48s/it, loss=0.302, lr=1.43e-5]\nSteps: 90%|█████████ | 904/1000 [35:26<03:57, 2.48s/it, loss=0.296, lr=1.4e-5] \nSteps: 90%|█████████ | 905/1000 [35:28<03:37, 2.29s/it, loss=0.296, lr=1.4e-5]\nSteps: 90%|█████████ | 905/1000 [35:28<03:37, 2.29s/it, loss=0.609, lr=1.38e-5]\nSteps: 91%|█████████ | 906/1000 [35:29<03:23, 2.17s/it, loss=0.609, lr=1.38e-5]\nSteps: 91%|█████████ | 906/1000 [35:29<03:23, 2.17s/it, loss=0.326, lr=1.35e-5]\nSteps: 91%|█████████ | 907/1000 [35:31<03:13, 2.08s/it, loss=0.326, lr=1.35e-5]\nSteps: 91%|█████████ | 907/1000 [35:31<03:13, 2.08s/it, loss=0.318, lr=1.32e-5]\nSteps: 91%|█████████ | 908/1000 [35:33<03:05, 2.02s/it, loss=0.318, lr=1.32e-5]\nSteps: 91%|█████████ | 908/1000 [35:33<03:05, 2.02s/it, loss=0.327, lr=1.29e-5]\nSteps: 91%|█████████ | 909/1000 [35:35<02:59, 1.97s/it, loss=0.327, lr=1.29e-5]\nSteps: 91%|█████████ | 909/1000 [35:35<02:59, 1.97s/it, loss=0.337, lr=1.26e-5]\nSteps: 91%|█████████ | 910/1000 [35:37<02:54, 1.94s/it, loss=0.337, lr=1.26e-5]\nSteps: 91%|█████████ | 910/1000 [35:37<02:54, 1.94s/it, loss=0.396, lr=1.24e-5]\nSteps: 91%|█████████ | 911/1000 [35:39<02:50, 1.92s/it, loss=0.396, lr=1.24e-5]\nSteps: 91%|█████████ | 911/1000 [35:39<02:50, 1.92s/it, loss=0.422, lr=1.21e-5]\nSteps: 91%|█████████ | 912/1000 [35:41<02:47, 1.91s/it, loss=0.422, lr=1.21e-5]\nSteps: 91%|█████████ | 912/1000 [35:41<02:47, 1.91s/it, loss=0.291, lr=1.18e-5]\nSteps: 91%|█████████▏| 913/1000 [35:43<02:44, 1.89s/it, loss=0.291, lr=1.18e-5]\nSteps: 91%|█████████▏| 913/1000 [35:43<02:44, 1.89s/it, loss=0.289, lr=1.16e-5]\nSteps: 91%|█████████▏| 914/1000 [35:44<02:42, 1.89s/it, loss=0.289, lr=1.16e-5]\nSteps: 91%|█████████▏| 914/1000 [35:44<02:42, 1.89s/it, loss=0.414, lr=1.13e-5]\nSteps: 92%|█████████▏| 915/1000 [35:46<02:40, 1.88s/it, loss=0.414, lr=1.13e-5]\nSteps: 92%|█████████▏| 915/1000 [35:46<02:40, 1.88s/it, loss=0.315, lr=1.1e-5] \nSteps: 92%|█████████▏| 916/1000 [35:48<02:37, 1.88s/it, loss=0.315, lr=1.1e-5]\nSteps: 92%|█████████▏| 916/1000 [35:48<02:37, 1.88s/it, loss=0.331, lr=1.08e-5]\nSteps: 92%|█████████▏| 917/1000 [35:50<02:35, 1.88s/it, loss=0.331, lr=1.08e-5]\nSteps: 92%|█████████▏| 917/1000 [35:50<02:35, 1.88s/it, loss=0.368, lr=1.05e-5]\nSteps: 92%|█████████▏| 918/1000 [35:52<02:33, 1.88s/it, loss=0.368, lr=1.05e-5]\nSteps: 92%|█████████▏| 918/1000 [35:52<02:33, 1.88s/it, loss=0.57, lr=1.03e-5] \nSteps: 92%|█████████▏| 919/1000 [35:54<02:31, 1.88s/it, loss=0.57, lr=1.03e-5]\nSteps: 92%|█████████▏| 919/1000 [35:54<02:31, 1.88s/it, loss=0.748, lr=1e-5] \nSteps: 92%|█████████▏| 920/1000 [35:56<02:30, 1.88s/it, loss=0.748, lr=1e-5]\nSteps: 92%|█████████▏| 920/1000 [35:56<02:30, 1.88s/it, loss=0.379, lr=9.79e-6]\nSteps: 92%|█████████▏| 921/1000 [35:58<02:27, 1.87s/it, loss=0.379, lr=9.79e-6]\nSteps: 92%|█████████▏| 921/1000 [35:58<02:27, 1.87s/it, loss=0.331, lr=9.55e-6]\nSteps: 92%|█████████▏| 922/1000 [35:59<02:26, 1.87s/it, loss=0.331, lr=9.55e-6]\nSteps: 92%|█████████▏| 922/1000 [35:59<02:26, 1.87s/it, loss=0.358, lr=9.31e-6]\nSteps: 92%|█████████▏| 923/1000 [36:01<02:24, 1.88s/it, loss=0.358, lr=9.31e-6]\nSteps: 92%|█████████▏| 923/1000 [36:01<02:24, 1.88s/it, loss=0.816, lr=9.07e-6]\nSteps: 92%|█████████▏| 924/1000 [36:03<02:23, 1.88s/it, loss=0.816, lr=9.07e-6]\nSteps: 92%|█████████▏| 924/1000 [36:03<02:23, 1.88s/it, loss=1.03, lr=8.84e-6] \nSteps: 92%|█████████▎| 925/1000 [36:05<02:21, 1.88s/it, loss=1.03, lr=8.84e-6]\nSteps: 92%|█████████▎| 925/1000 [36:05<02:21, 1.88s/it, loss=0.466, lr=8.61e-6]\nSteps: 93%|█████████▎| 926/1000 [36:07<02:19, 1.88s/it, loss=0.466, lr=8.61e-6]\nSteps: 93%|█████████▎| 926/1000 [36:07<02:19, 1.88s/it, loss=0.505, lr=8.39e-6]\nSteps: 93%|█████████▎| 927/1000 [36:09<02:16, 1.88s/it, loss=0.505, lr=8.39e-6]\nSteps: 93%|█████████▎| 927/1000 [36:09<02:16, 1.88s/it, loss=0.736, lr=8.16e-6]\nSteps: 93%|█████████▎| 928/1000 [36:11<02:14, 1.87s/it, loss=0.736, lr=8.16e-6]\nSteps: 93%|█████████▎| 928/1000 [36:11<02:14, 1.87s/it, loss=0.274, lr=7.94e-6]\nSteps: 93%|█████████▎| 929/1000 [36:13<02:12, 1.87s/it, loss=0.274, lr=7.94e-6]\nSteps: 93%|█████████▎| 929/1000 [36:13<02:12, 1.87s/it, loss=0.434, lr=7.72e-6]\nSteps: 93%|█████████▎| 930/1000 [36:14<02:10, 1.87s/it, loss=0.434, lr=7.72e-6]\nSteps: 93%|█████████▎| 930/1000 [36:14<02:10, 1.87s/it, loss=0.471, lr=7.51e-6]\nSteps: 93%|█████████▎| 931/1000 [36:22<04:14, 3.69s/it, loss=0.471, lr=7.51e-6]\nSteps: 93%|█████████▎| 931/1000 [36:22<04:14, 3.69s/it, loss=0.424, lr=7.3e-6] \nSteps: 93%|█████████▎| 932/1000 [36:24<03:33, 3.14s/it, loss=0.424, lr=7.3e-6]\nSteps: 93%|█████████▎| 932/1000 [36:24<03:33, 3.14s/it, loss=0.324, lr=7.09e-6]\nSteps: 93%|█████████▎| 933/1000 [36:26<03:04, 2.76s/it, loss=0.324, lr=7.09e-6]\nSteps: 93%|█████████▎| 933/1000 [36:26<03:04, 2.76s/it, loss=0.827, lr=6.88e-6]\nSteps: 93%|█████████▎| 934/1000 [36:28<02:44, 2.49s/it, loss=0.827, lr=6.88e-6]\nSteps: 93%|█████████▎| 934/1000 [36:28<02:44, 2.49s/it, loss=0.567, lr=6.68e-6]\nSteps: 94%|█████████▎| 935/1000 [36:30<02:29, 2.31s/it, loss=0.567, lr=6.68e-6]\nSteps: 94%|█████████▎| 935/1000 [36:30<02:29, 2.31s/it, loss=0.363, lr=6.48e-6]\nSteps: 94%|█████████▎| 936/1000 [36:32<02:19, 2.18s/it, loss=0.363, lr=6.48e-6]\nSteps: 94%|█████████▎| 936/1000 [36:32<02:19, 2.18s/it, loss=0.556, lr=6.28e-6]\nSteps: 94%|█████████▎| 937/1000 [36:34<02:11, 2.08s/it, loss=0.556, lr=6.28e-6]\nSteps: 94%|█████████▎| 937/1000 [36:34<02:11, 2.08s/it, loss=0.445, lr=6.09e-6]\nSteps: 94%|█████████▍| 938/1000 [36:35<02:05, 2.02s/it, loss=0.445, lr=6.09e-6]\nSteps: 94%|█████████▍| 938/1000 [36:35<02:05, 2.02s/it, loss=0.685, lr=5.9e-6] \nSteps: 94%|█████████▍| 939/1000 [36:37<02:00, 1.97s/it, loss=0.685, lr=5.9e-6]\nSteps: 94%|█████████▍| 939/1000 [36:37<02:00, 1.97s/it, loss=0.334, lr=5.71e-6]\nSteps: 94%|█████████▍| 940/1000 [36:39<01:56, 1.94s/it, loss=0.334, lr=5.71e-6]\nSteps: 94%|█████████▍| 940/1000 [36:39<01:56, 1.94s/it, loss=0.332, lr=5.53e-6]\nSteps: 94%|█████████▍| 941/1000 [36:41<01:53, 1.92s/it, loss=0.332, lr=5.53e-6]\nSteps: 94%|█████████▍| 941/1000 [36:41<01:53, 1.92s/it, loss=1.02, lr=5.34e-6] \nSteps: 94%|█████████▍| 942/1000 [36:43<01:50, 1.91s/it, loss=1.02, lr=5.34e-6]\nSteps: 94%|█████████▍| 942/1000 [36:43<01:50, 1.91s/it, loss=0.346, lr=5.17e-6]\nSteps: 94%|█████████▍| 943/1000 [36:45<01:48, 1.90s/it, loss=0.346, lr=5.17e-6]\nSteps: 94%|█████████▍| 943/1000 [36:45<01:48, 1.90s/it, loss=0.341, lr=4.99e-6]\nSteps: 94%|█████████▍| 944/1000 [36:47<01:45, 1.89s/it, loss=0.341, lr=4.99e-6]\nSteps: 94%|█████████▍| 944/1000 [36:47<01:45, 1.89s/it, loss=0.681, lr=4.82e-6]\nSteps: 94%|█████████▍| 945/1000 [36:49<01:43, 1.88s/it, loss=0.681, lr=4.82e-6]\nSteps: 94%|█████████▍| 945/1000 [36:49<01:43, 1.88s/it, loss=0.317, lr=4.65e-6]\nSteps: 95%|█████████▍| 946/1000 [36:50<01:41, 1.88s/it, loss=0.317, lr=4.65e-6]\nSteps: 95%|█████████▍| 946/1000 [36:50<01:41, 1.88s/it, loss=1.03, lr=4.48e-6] \nSteps: 95%|█████████▍| 947/1000 [36:52<01:39, 1.88s/it, loss=1.03, lr=4.48e-6]\nSteps: 95%|█████████▍| 947/1000 [36:52<01:39, 1.88s/it, loss=0.624, lr=4.32e-6]\nSteps: 95%|█████████▍| 948/1000 [36:54<01:37, 1.87s/it, loss=0.624, lr=4.32e-6]\nSteps: 95%|█████████▍| 948/1000 [36:54<01:37, 1.87s/it, loss=0.504, lr=4.16e-6]\nSteps: 95%|█████████▍| 949/1000 [36:56<01:35, 1.87s/it, loss=0.504, lr=4.16e-6]\nSteps: 95%|█████████▍| 949/1000 [36:56<01:35, 1.87s/it, loss=0.628, lr=4e-6] \nSteps: 95%|█████████▌| 950/1000 [36:58<01:33, 1.87s/it, loss=0.628, lr=4e-6]\nSteps: 95%|█████████▌| 950/1000 [36:58<01:33, 1.87s/it, loss=0.607, lr=3.84e-6]\nSteps: 95%|█████████▌| 951/1000 [37:00<01:31, 1.87s/it, loss=0.607, lr=3.84e-6]\nSteps: 95%|█████████▌| 951/1000 [37:00<01:31, 1.87s/it, loss=0.364, lr=3.69e-6]\nSteps: 95%|█████████▌| 952/1000 [37:02<01:29, 1.87s/it, loss=0.364, lr=3.69e-6]\nSteps: 95%|█████████▌| 952/1000 [37:02<01:29, 1.87s/it, loss=0.557, lr=3.54e-6]\nSteps: 95%|█████████▌| 953/1000 [37:03<01:27, 1.87s/it, loss=0.557, lr=3.54e-6]\nSteps: 95%|█████████▌| 953/1000 [37:03<01:27, 1.87s/it, loss=0.282, lr=3.4e-6] \nSteps: 95%|█████████▌| 954/1000 [37:05<01:25, 1.87s/it, loss=0.282, lr=3.4e-6]\nSteps: 95%|█████████▌| 954/1000 [37:05<01:25, 1.87s/it, loss=0.285, lr=3.25e-6]\nSteps: 96%|█████████▌| 955/1000 [37:07<01:23, 1.87s/it, loss=0.285, lr=3.25e-6]\nSteps: 96%|█████████▌| 955/1000 [37:07<01:23, 1.87s/it, loss=0.333, lr=3.11e-6]\nSteps: 96%|█████████▌| 956/1000 [37:09<01:22, 1.87s/it, loss=0.333, lr=3.11e-6]\nSteps: 96%|█████████▌| 956/1000 [37:09<01:22, 1.87s/it, loss=0.295, lr=2.98e-6]\nSteps: 96%|█████████▌| 957/1000 [37:11<01:20, 1.87s/it, loss=0.295, lr=2.98e-6]\nSteps: 96%|█████████▌| 957/1000 [37:11<01:20, 1.87s/it, loss=0.399, lr=2.84e-6]\nSteps: 96%|█████████▌| 958/1000 [37:13<01:18, 1.86s/it, loss=0.399, lr=2.84e-6]\nSteps: 96%|█████████▌| 958/1000 [37:13<01:18, 1.86s/it, loss=0.416, lr=2.71e-6]\nSteps: 96%|█████████▌| 959/1000 [37:15<01:16, 1.86s/it, loss=0.416, lr=2.71e-6]\nSteps: 96%|█████████▌| 959/1000 [37:15<01:16, 1.86s/it, loss=0.496, lr=2.59e-6]\nSteps: 96%|█████████▌| 960/1000 [37:17<01:14, 1.87s/it, loss=0.496, lr=2.59e-6]\nSteps: 96%|█████████▌| 960/1000 [37:17<01:14, 1.87s/it, loss=0.52, lr=2.46e-6] \nSteps: 96%|█████████▌| 961/1000 [37:24<02:24, 3.70s/it, loss=0.52, lr=2.46e-6]\nSteps: 96%|█████████▌| 961/1000 [37:24<02:24, 3.70s/it, loss=0.607, lr=2.34e-6]\nSteps: 96%|█████████▌| 962/1000 [37:26<01:59, 3.15s/it, loss=0.607, lr=2.34e-6]\nSteps: 96%|█████████▌| 962/1000 [37:26<01:59, 3.15s/it, loss=0.305, lr=2.22e-6]\nSteps: 96%|█████████▋| 963/1000 [37:28<01:42, 2.76s/it, loss=0.305, lr=2.22e-6]\nSteps: 96%|█████████▋| 963/1000 [37:28<01:42, 2.76s/it, loss=0.302, lr=2.11e-6]\nSteps: 96%|█████████▋| 964/1000 [37:30<01:29, 2.50s/it, loss=0.302, lr=2.11e-6]\nSteps: 96%|█████████▋| 964/1000 [37:30<01:29, 2.50s/it, loss=0.363, lr=2e-6] \nSteps: 96%|█████████▋| 965/1000 [37:32<01:20, 2.31s/it, loss=0.363, lr=2e-6]\nSteps: 96%|█████████▋| 965/1000 [37:32<01:20, 2.31s/it, loss=0.786, lr=1.89e-6]\nSteps: 97%|█████████▋| 966/1000 [37:34<01:14, 2.18s/it, loss=0.786, lr=1.89e-6]\nSteps: 97%|█████████▋| 966/1000 [37:34<01:14, 2.18s/it, loss=0.582, lr=1.78e-6]\nSteps: 97%|█████████▋| 967/1000 [37:36<01:08, 2.09s/it, loss=0.582, lr=1.78e-6]\nSteps: 97%|█████████▋| 967/1000 [37:36<01:08, 2.09s/it, loss=0.393, lr=1.68e-6]\nSteps: 97%|█████████▋| 968/1000 [37:38<01:04, 2.02s/it, loss=0.393, lr=1.68e-6]\nSteps: 97%|█████████▋| 968/1000 [37:38<01:04, 2.02s/it, loss=0.404, lr=1.58e-6]\nSteps: 97%|█████████▋| 969/1000 [37:39<01:01, 1.98s/it, loss=0.404, lr=1.58e-6]\nSteps: 97%|█████████▋| 969/1000 [37:39<01:01, 1.98s/it, loss=0.289, lr=1.48e-6]\nSteps: 97%|█████████▋| 970/1000 [37:41<00:58, 1.95s/it, loss=0.289, lr=1.48e-6]\nSteps: 97%|█████████▋| 970/1000 [37:41<00:58, 1.95s/it, loss=0.325, lr=1.39e-6]\nSteps: 97%|█████████▋| 971/1000 [37:43<00:55, 1.92s/it, loss=0.325, lr=1.39e-6]\nSteps: 97%|█████████▋| 971/1000 [37:43<00:55, 1.92s/it, loss=1.08, lr=1.3e-6] \nSteps: 97%|█████████▋| 972/1000 [37:45<00:53, 1.91s/it, loss=1.08, lr=1.3e-6]\nSteps: 97%|█████████▋| 972/1000 [37:45<00:53, 1.91s/it, loss=0.492, lr=1.21e-6]\nSteps: 97%|█████████▋| 973/1000 [37:47<00:51, 1.90s/it, loss=0.492, lr=1.21e-6]\nSteps: 97%|█████████▋| 973/1000 [37:47<00:51, 1.90s/it, loss=0.571, lr=1.12e-6]\nSteps: 97%|█████████▋| 974/1000 [37:49<00:49, 1.89s/it, loss=0.571, lr=1.12e-6]\nSteps: 97%|█████████▋| 974/1000 [37:49<00:49, 1.89s/it, loss=0.343, lr=1.04e-6]\nSteps: 98%|█████████▊| 975/1000 [37:51<00:47, 1.89s/it, loss=0.343, lr=1.04e-6]\nSteps: 98%|█████████▊| 975/1000 [37:51<00:47, 1.89s/it, loss=0.655, lr=9.63e-7]\nSteps: 98%|█████████▊| 976/1000 [37:53<00:45, 1.88s/it, loss=0.655, lr=9.63e-7]\nSteps: 98%|█████████▊| 976/1000 [37:53<00:45, 1.88s/it, loss=0.474, lr=8.88e-7]\nSteps: 98%|█████████▊| 977/1000 [37:54<00:43, 1.88s/it, loss=0.474, lr=8.88e-7]\nSteps: 98%|█████████▊| 977/1000 [37:54<00:43, 1.88s/it, loss=0.344, lr=8.15e-7]\nSteps: 98%|█████████▊| 978/1000 [37:56<00:41, 1.88s/it, loss=0.344, lr=8.15e-7]\nSteps: 98%|█████████▊| 978/1000 [37:56<00:41, 1.88s/it, loss=0.565, lr=7.46e-7]\nSteps: 98%|█████████▊| 979/1000 [37:58<00:39, 1.88s/it, loss=0.565, lr=7.46e-7]\nSteps: 98%|█████████▊| 979/1000 [37:58<00:39, 1.88s/it, loss=0.311, lr=6.8e-7] \nSteps: 98%|█████████▊| 980/1000 [38:00<00:37, 1.87s/it, loss=0.311, lr=6.8e-7]\nSteps: 98%|█████████▊| 980/1000 [38:00<00:37, 1.87s/it, loss=0.762, lr=6.17e-7]\nSteps: 98%|█████████▊| 981/1000 [38:02<00:35, 1.87s/it, loss=0.762, lr=6.17e-7]\nSteps: 98%|█████████▊| 981/1000 [38:02<00:35, 1.87s/it, loss=0.832, lr=5.56e-7]\nSteps: 98%|█████████▊| 982/1000 [38:04<00:33, 1.87s/it, loss=0.832, lr=5.56e-7]\nSteps: 98%|█████████▊| 982/1000 [38:04<00:33, 1.87s/it, loss=0.289, lr=4.99e-7]\nSteps: 98%|█████████▊| 983/1000 [38:06<00:31, 1.87s/it, loss=0.289, lr=4.99e-7]\nSteps: 98%|█████████▊| 983/1000 [38:06<00:31, 1.87s/it, loss=0.513, lr=4.46e-7]\nSteps: 98%|█████████▊| 984/1000 [38:08<00:29, 1.87s/it, loss=0.513, lr=4.46e-7]\nSteps: 98%|█████████▊| 984/1000 [38:08<00:29, 1.87s/it, loss=0.227, lr=3.95e-7]\nSteps: 98%|█████████▊| 985/1000 [38:09<00:28, 1.87s/it, loss=0.227, lr=3.95e-7]\nSteps: 98%|█████████▊| 985/1000 [38:09<00:28, 1.87s/it, loss=0.385, lr=3.47e-7]\nSteps: 99%|█████████▊| 986/1000 [38:11<00:26, 1.87s/it, loss=0.385, lr=3.47e-7]\nSteps: 99%|█████████▊| 986/1000 [38:11<00:26, 1.87s/it, loss=0.451, lr=3.02e-7]\nSteps: 99%|█████████▊| 987/1000 [38:13<00:24, 1.87s/it, loss=0.451, lr=3.02e-7]\nSteps: 99%|█████████▊| 987/1000 [38:13<00:24, 1.87s/it, loss=0.391, lr=2.61e-7]\nSteps: 99%|█████████▉| 988/1000 [38:15<00:22, 1.88s/it, loss=0.391, lr=2.61e-7]\nSteps: 99%|█████████▉| 988/1000 [38:15<00:22, 1.88s/it, loss=0.337, lr=2.22e-7]\nSteps: 99%|█████████▉| 989/1000 [38:17<00:20, 1.88s/it, loss=0.337, lr=2.22e-7]\nSteps: 99%|█████████▉| 989/1000 [38:17<00:20, 1.88s/it, loss=0.342, lr=1.87e-7]\nSteps: 99%|█████████▉| 990/1000 [38:19<00:18, 1.87s/it, loss=0.342, lr=1.87e-7]\nSteps: 99%|█████████▉| 990/1000 [38:19<00:18, 1.87s/it, loss=0.278, lr=1.54e-7]\nSteps: 99%|█████████▉| 991/1000 [38:26<00:32, 3.62s/it, loss=0.278, lr=1.54e-7]\nSteps: 99%|█████████▉| 991/1000 [38:26<00:32, 3.62s/it, loss=0.339, lr=1.25e-7]\nSteps: 99%|█████████▉| 992/1000 [38:28<00:24, 3.09s/it, loss=0.339, lr=1.25e-7]\nSteps: 99%|█████████▉| 992/1000 [38:28<00:24, 3.09s/it, loss=0.54, lr=9.87e-8] \nSteps: 99%|█████████▉| 993/1000 [38:30<00:19, 2.73s/it, loss=0.54, lr=9.87e-8]\nSteps: 99%|█████████▉| 993/1000 [38:30<00:19, 2.73s/it, loss=0.88, lr=7.56e-8]\nSteps: 99%|█████████▉| 994/1000 [38:32<00:14, 2.47s/it, loss=0.88, lr=7.56e-8]\nSteps: 99%|█████████▉| 994/1000 [38:32<00:14, 2.47s/it, loss=0.269, lr=5.55e-8]\nSteps: 100%|█████████▉| 995/1000 [38:34<00:11, 2.29s/it, loss=0.269, lr=5.55e-8]\nSteps: 100%|█████████▉| 995/1000 [38:34<00:11, 2.29s/it, loss=0.283, lr=3.86e-8]\nSteps: 100%|█████████▉| 996/1000 [38:36<00:08, 2.16s/it, loss=0.283, lr=3.86e-8]\nSteps: 100%|█████████▉| 996/1000 [38:36<00:08, 2.16s/it, loss=0.801, lr=2.47e-8]\nSteps: 100%|█████████▉| 997/1000 [38:38<00:06, 2.08s/it, loss=0.801, lr=2.47e-8]\nSteps: 100%|█████████▉| 997/1000 [38:38<00:06, 2.08s/it, loss=1, lr=1.39e-8] \nSteps: 100%|█████████▉| 998/1000 [38:40<00:04, 2.02s/it, loss=1, lr=1.39e-8]\nSteps: 100%|█████████▉| 998/1000 [38:40<00:04, 2.02s/it, loss=0.874, lr=6.17e-9]\nSteps: 100%|█████████▉| 999/1000 [38:41<00:01, 1.97s/it, loss=0.874, lr=6.17e-9]\nSteps: 100%|█████████▉| 999/1000 [38:41<00:01, 1.97s/it, loss=0.505, lr=1.54e-9]\nSteps: 100%|██████████| 1000/1000 [38:43<00:00, 1.94s/it, loss=0.505, lr=1.54e-9]\nSteps: 100%|██████████| 1000/1000 [38:43<00:00, 1.94s/it, loss=0.424, lr=0] \nSteps: 100%|██████████| 1000/1000 [38:47<00:00, 2.33s/it, loss=0.424, lr=0]\n---Tar up output directory---\nmochi-lora/\nmochi-lora/pytorch_lora_weights.safetensors\nUploading to Hugging Face: lucataco/mochi-lora-disney\nHF Repo URL: https://huggingface.co/lucataco/mochi-lora-disney\npytorch_lora_weights.safetensors: 0%| | 0.00/76.1M [00:00<?, ?B/s]\npytorch_lora_weights.safetensors: 2%|▏ | 1.69M/76.1M [00:00<00:04, 16.8MB/s]\npytorch_lora_weights.safetensors: 21%|██ | 16.0M/76.1M [00:00<00:01, 43.1MB/s]\npytorch_lora_weights.safetensors: 42%|████▏ | 32.0M/76.1M [00:00<00:00, 54.4MB/s]\npytorch_lora_weights.safetensors: 63%|██████▎ | 48.0M/76.1M [00:00<00:00, 61.1MB/s]\npytorch_lora_weights.safetensors: 84%|████████▍ | 64.0M/76.1M [00:01<00:00, 61.1MB/s]\npytorch_lora_weights.safetensors: 100%|██████████| 76.1M/76.1M [00:01<00:00, 56.8MB/s]\nSuccessfully uploaded model to https://huggingface.co/lucataco/mochi-lora-disney", "metrics": { "predict_time": 2382.770900905, "total_time": 2452.026534 }, "output": { "weights": "https://replicate.delivery/xezq/8M2egxAio8VqOaRduflWdkA5J7DP5QCIBG70eP7PCrfHdZnPB/trained_model.tar" }, "started_at": "2024-12-11T15:06:43.255633Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/w6kpq5g261rme0ckps0rad27hc", "cancel": "https://api.replicate.com/v1/predictions/w6kpq5g261rme0ckps0rad27hc/cancel" }, "version": "170ea99fb48a30fef98cb1c9fb403a2882ab9d60c2ba15ad9383ace33c3fa385" }
Generated inCleaning up previous runs Extracted 60 files from zip to videos_input ---Starting to Trim input videos--- Processing: videos_input/0bb5f6dbf8ed2e0060f0ac4164b24847.mp4 videos_input/0bb5f6dbf8ed2e0060f0ac4164b24847.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/0bb5f6dbf8ed2e0060f0ac4164b24847.txt to videos_prepared/0bb5f6dbf8ed2e0060f0ac4164b24847.txt Moviepy - Building video videos_prepared/0bb5f6dbf8ed2e0060f0ac4164b24847.mp4. 0%| | 0/30 [00:00<?, ?it/s] 0%| | 0/30 [00:00<?, ?it/s] Moviepy - Writing video videos_prepared/0bb5f6dbf8ed2e0060f0ac4164b24847.mp4 0%| | 0/30 [00:00<?, ?it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] 0%| | 0/30 [00:00<?, ?it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/0bb5f6dbf8ed2e0060f0ac4164b24847.mp4 0%| | 0/30 [00:00<?, ?it/s] Processing: videos_input/1d50a3d9703f152758d5422c8b48010f.mp4 videos_input/1d50a3d9703f152758d5422c8b48010f.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/1d50a3d9703f152758d5422c8b48010f.txt to videos_prepared/1d50a3d9703f152758d5422c8b48010f.txt Moviepy - Building video videos_prepared/1d50a3d9703f152758d5422c8b48010f.mp4. Moviepy - Writing video videos_prepared/1d50a3d9703f152758d5422c8b48010f.mp4 3%|▎ | 1/30 [00:00<00:07, 3.78it/s] 3%|▎ | 1/30 [00:00<00:07, 3.78it/s] 3%|▎ | 1/30 [00:00<00:07, 3.78it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 98%|█████████▊| 39/40 [00:00<00:00, 385.32it/s, now=None] 3%|▎ | 1/30 [00:00<00:07, 3.78it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/1d50a3d9703f152758d5422c8b48010f.mp4 3%|▎ | 1/30 [00:00<00:07, 3.78it/s] Processing: videos_input/2c1ed5408882479b06681f7cf372916a.mp4 videos_input/2c1ed5408882479b06681f7cf372916a.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/2c1ed5408882479b06681f7cf372916a.txt to videos_prepared/2c1ed5408882479b06681f7cf372916a.txt 7%|▋ | 2/30 [00:00<00:07, 3.53it/s] 7%|▋ | 2/30 [00:00<00:07, 3.53it/s] Moviepy - Building video videos_prepared/2c1ed5408882479b06681f7cf372916a.mp4. Moviepy - Writing video videos_prepared/2c1ed5408882479b06681f7cf372916a.mp4 7%|▋ | 2/30 [00:00<00:07, 3.53it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 100%|██████████| 40/40 [00:00<00:00, 391.86it/s, now=None] 7%|▋ | 2/30 [00:00<00:07, 3.53it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/2c1ed5408882479b06681f7cf372916a.mp4 7%|▋ | 2/30 [00:00<00:07, 3.53it/s] Processing: videos_input/3f0979e6cae25447f416372c49ad5e07.mp4 videos_input/3f0979e6cae25447f416372c49ad5e07.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/3f0979e6cae25447f416372c49ad5e07.txt to videos_prepared/3f0979e6cae25447f416372c49ad5e07.txt Moviepy - Building video videos_prepared/3f0979e6cae25447f416372c49ad5e07.mp4. Moviepy - Writing video videos_prepared/3f0979e6cae25447f416372c49ad5e07.mp4 10%|█ | 3/30 [00:00<00:07, 3.53it/s] 10%|█ | 3/30 [00:00<00:07, 3.53it/s] 10%|█ | 3/30 [00:00<00:07, 3.53it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 98%|█████████▊| 39/40 [00:00<00:00, 384.65it/s, now=None] 10%|█ | 3/30 [00:01<00:07, 3.53it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/3f0979e6cae25447f416372c49ad5e07.mp4 10%|█ | 3/30 [00:01<00:07, 3.53it/s] Processing: videos_input/4adbb3a2945c9edd78785daccfd23e80.mp4 videos_input/4adbb3a2945c9edd78785daccfd23e80.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/4adbb3a2945c9edd78785daccfd23e80.txt to videos_prepared/4adbb3a2945c9edd78785daccfd23e80.txt Moviepy - Building video videos_prepared/4adbb3a2945c9edd78785daccfd23e80.mp4. Moviepy - Writing video videos_prepared/4adbb3a2945c9edd78785daccfd23e80.mp4 13%|█▎ | 4/30 [00:01<00:07, 3.54it/s] 13%|█▎ | 4/30 [00:01<00:07, 3.54it/s] 13%|█▎ | 4/30 [00:01<00:07, 3.54it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] 13%|█▎ | 4/30 [00:01<00:07, 3.54it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/4adbb3a2945c9edd78785daccfd23e80.mp4 13%|█▎ | 4/30 [00:01<00:07, 3.54it/s] Processing: videos_input/4c918b917308ff03120e9e86650a2d3c.mp4 videos_input/4c918b917308ff03120e9e86650a2d3c.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/4c918b917308ff03120e9e86650a2d3c.txt to videos_prepared/4c918b917308ff03120e9e86650a2d3c.txt Moviepy - Building video videos_prepared/4c918b917308ff03120e9e86650a2d3c.mp4. Moviepy - Writing video videos_prepared/4c918b917308ff03120e9e86650a2d3c.mp4 17%|█▋ | 5/30 [00:01<00:06, 3.68it/s] 17%|█▋ | 5/30 [00:01<00:06, 3.68it/s] 17%|█▋ | 5/30 [00:01<00:06, 3.68it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 98%|█████████▊| 39/40 [00:00<00:00, 388.06it/s, now=None] 17%|█▋ | 5/30 [00:01<00:06, 3.68it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/4c918b917308ff03120e9e86650a2d3c.mp4 17%|█▋ | 5/30 [00:01<00:06, 3.68it/s] Processing: videos_input/5a0229ffdb3bd9d8e81dca7988d7cdbb.mp4 videos_input/5a0229ffdb3bd9d8e81dca7988d7cdbb.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/5a0229ffdb3bd9d8e81dca7988d7cdbb.txt to videos_prepared/5a0229ffdb3bd9d8e81dca7988d7cdbb.txt Moviepy - Building video videos_prepared/5a0229ffdb3bd9d8e81dca7988d7cdbb.mp4. Moviepy - Writing video videos_prepared/5a0229ffdb3bd9d8e81dca7988d7cdbb.mp4 20%|██ | 6/30 [00:01<00:06, 3.60it/s] 20%|██ | 6/30 [00:01<00:06, 3.60it/s] 20%|██ | 6/30 [00:01<00:06, 3.60it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 92%|█████████▎| 37/40 [00:00<00:00, 366.42it/s, now=None] 20%|██ | 6/30 [00:01<00:06, 3.60it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/5a0229ffdb3bd9d8e81dca7988d7cdbb.mp4 20%|██ | 6/30 [00:01<00:06, 3.60it/s] Processing: videos_input/05a234b0164d015d468f2f53e771b4cf.mp4 videos_input/05a234b0164d015d468f2f53e771b4cf.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/05a234b0164d015d468f2f53e771b4cf.txt to videos_prepared/05a234b0164d015d468f2f53e771b4cf.txt Moviepy - Building video videos_prepared/05a234b0164d015d468f2f53e771b4cf.mp4. Moviepy - Writing video videos_prepared/05a234b0164d015d468f2f53e771b4cf.mp4 23%|██▎ | 7/30 [00:01<00:06, 3.61it/s] 23%|██▎ | 7/30 [00:01<00:06, 3.61it/s] 23%|██▎ | 7/30 [00:01<00:06, 3.61it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] 23%|██▎ | 7/30 [00:02<00:06, 3.61it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/05a234b0164d015d468f2f53e771b4cf.mp4 23%|██▎ | 7/30 [00:02<00:06, 3.61it/s] Processing: videos_input/05ccfa61ece031e881d173289761cf91.mp4 videos_input/05ccfa61ece031e881d173289761cf91.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/05ccfa61ece031e881d173289761cf91.txt to videos_prepared/05ccfa61ece031e881d173289761cf91.txt 27%|██▋ | 8/30 [00:02<00:06, 3.63it/s] Moviepy - Building video videos_prepared/05ccfa61ece031e881d173289761cf91.mp4. 27%|██▋ | 8/30 [00:02<00:06, 3.63it/s] Moviepy - Writing video videos_prepared/05ccfa61ece031e881d173289761cf91.mp4 27%|██▋ | 8/30 [00:02<00:06, 3.63it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 98%|█████████▊| 39/40 [00:00<00:00, 382.15it/s, now=None] 27%|██▋ | 8/30 [00:02<00:06, 3.63it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/05ccfa61ece031e881d173289761cf91.mp4 27%|██▋ | 8/30 [00:02<00:06, 3.63it/s] Processing: videos_input/7d6dcf13f5c3d45b85c5ea0544c429e4.mp4 videos_input/7d6dcf13f5c3d45b85c5ea0544c429e4.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/7d6dcf13f5c3d45b85c5ea0544c429e4.txt to videos_prepared/7d6dcf13f5c3d45b85c5ea0544c429e4.txt Moviepy - Building video videos_prepared/7d6dcf13f5c3d45b85c5ea0544c429e4.mp4. Moviepy - Writing video videos_prepared/7d6dcf13f5c3d45b85c5ea0544c429e4.mp4 30%|███ | 9/30 [00:02<00:05, 3.59it/s] 30%|███ | 9/30 [00:02<00:05, 3.59it/s] 30%|███ | 9/30 [00:02<00:05, 3.59it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 98%|█████████▊| 39/40 [00:00<00:00, 387.57it/s, now=None] 30%|███ | 9/30 [00:02<00:05, 3.59it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/7d6dcf13f5c3d45b85c5ea0544c429e4.mp4 30%|███ | 9/30 [00:02<00:05, 3.59it/s] Processing: videos_input/7fe0c83572de828da1cab0c118dece14.mp4 videos_input/7fe0c83572de828da1cab0c118dece14.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/7fe0c83572de828da1cab0c118dece14.txt to videos_prepared/7fe0c83572de828da1cab0c118dece14.txt Moviepy - Building video videos_prepared/7fe0c83572de828da1cab0c118dece14.mp4. Moviepy - Writing video videos_prepared/7fe0c83572de828da1cab0c118dece14.mp4 33%|███▎ | 10/30 [00:02<00:05, 3.49it/s] 33%|███▎ | 10/30 [00:02<00:05, 3.49it/s] 33%|███▎ | 10/30 [00:02<00:05, 3.49it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 98%|█████████▊| 39/40 [00:00<00:00, 385.04it/s, now=None] 33%|███▎ | 10/30 [00:03<00:05, 3.49it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/7fe0c83572de828da1cab0c118dece14.mp4 33%|███▎ | 10/30 [00:03<00:05, 3.49it/s] Processing: videos_input/8adfde998361b1d7c6f38a35481667fd.mp4 videos_input/8adfde998361b1d7c6f38a35481667fd.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/8adfde998361b1d7c6f38a35481667fd.txt to videos_prepared/8adfde998361b1d7c6f38a35481667fd.txt Moviepy - Building video videos_prepared/8adfde998361b1d7c6f38a35481667fd.mp4. Moviepy - Writing video videos_prepared/8adfde998361b1d7c6f38a35481667fd.mp4 37%|███▋ | 11/30 [00:03<00:05, 3.54it/s] 37%|███▋ | 11/30 [00:03<00:05, 3.54it/s] 37%|███▋ | 11/30 [00:03<00:05, 3.54it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 98%|█████████▊| 39/40 [00:00<00:00, 387.19it/s, now=None] 37%|███▋ | 11/30 [00:03<00:05, 3.54it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/8adfde998361b1d7c6f38a35481667fd.mp4 37%|███▋ | 11/30 [00:03<00:05, 3.54it/s] Processing: videos_input/8ae679ab483ab344c881d4a813e0cb51.mp4 videos_input/8ae679ab483ab344c881d4a813e0cb51.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/8ae679ab483ab344c881d4a813e0cb51.txt to videos_prepared/8ae679ab483ab344c881d4a813e0cb51.txt Moviepy - Building video videos_prepared/8ae679ab483ab344c881d4a813e0cb51.mp4. Moviepy - Writing video videos_prepared/8ae679ab483ab344c881d4a813e0cb51.mp4 40%|████ | 12/30 [00:03<00:05, 3.50it/s] 40%|████ | 12/30 [00:03<00:05, 3.50it/s] 40%|████ | 12/30 [00:03<00:05, 3.50it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] 40%|████ | 12/30 [00:03<00:05, 3.50it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/8ae679ab483ab344c881d4a813e0cb51.mp4 40%|████ | 12/30 [00:03<00:05, 3.50it/s] Processing: videos_input/8d616fee8e0a280d2d87e478b948a729.mp4 videos_input/8d616fee8e0a280d2d87e478b948a729.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/8d616fee8e0a280d2d87e478b948a729.txt to videos_prepared/8d616fee8e0a280d2d87e478b948a729.txt Moviepy - Building video videos_prepared/8d616fee8e0a280d2d87e478b948a729.mp4. Moviepy - Writing video videos_prepared/8d616fee8e0a280d2d87e478b948a729.mp4 43%|████▎ | 13/30 [00:03<00:04, 3.49it/s] 43%|████▎ | 13/30 [00:03<00:04, 3.49it/s] 43%|████▎ | 13/30 [00:03<00:04, 3.49it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 98%|█████████▊| 39/40 [00:00<00:00, 389.56it/s, now=None] 43%|████▎ | 13/30 [00:03<00:04, 3.49it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/8d616fee8e0a280d2d87e478b948a729.mp4 43%|████▎ | 13/30 [00:03<00:04, 3.49it/s] Processing: videos_input/8e7722634784cf969c15f4a597f3af4d.mp4 videos_input/8e7722634784cf969c15f4a597f3af4d.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/8e7722634784cf969c15f4a597f3af4d.txt to videos_prepared/8e7722634784cf969c15f4a597f3af4d.txt Moviepy - Building video videos_prepared/8e7722634784cf969c15f4a597f3af4d.mp4. 47%|████▋ | 14/30 [00:03<00:04, 3.46it/s] 47%|████▋ | 14/30 [00:03<00:04, 3.46it/s] Moviepy - Writing video videos_prepared/8e7722634784cf969c15f4a597f3af4d.mp4 47%|████▋ | 14/30 [00:03<00:04, 3.46it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 100%|██████████| 40/40 [00:00<00:00, 395.87it/s, now=None] 47%|████▋ | 14/30 [00:04<00:04, 3.46it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/8e7722634784cf969c15f4a597f3af4d.mp4 47%|████▋ | 14/30 [00:04<00:04, 3.46it/s] Processing: videos_input/12e51adf1acbf7acbb703a96a464a39b.mp4 videos_input/12e51adf1acbf7acbb703a96a464a39b.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/12e51adf1acbf7acbb703a96a464a39b.txt to videos_prepared/12e51adf1acbf7acbb703a96a464a39b.txt Moviepy - Building video videos_prepared/12e51adf1acbf7acbb703a96a464a39b.mp4. 50%|█████ | 15/30 [00:04<00:04, 3.46it/s] 50%|█████ | 15/30 [00:04<00:04, 3.46it/s] Moviepy - Writing video videos_prepared/12e51adf1acbf7acbb703a96a464a39b.mp4 50%|█████ | 15/30 [00:04<00:04, 3.46it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] 50%|█████ | 15/30 [00:04<00:04, 3.46it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/12e51adf1acbf7acbb703a96a464a39b.mp4 50%|█████ | 15/30 [00:04<00:04, 3.46it/s] Processing: videos_input/46e9d133d051655c956c7089b672f519.mp4 videos_input/46e9d133d051655c956c7089b672f519.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/46e9d133d051655c956c7089b672f519.txt to videos_prepared/46e9d133d051655c956c7089b672f519.txt Moviepy - Building video videos_prepared/46e9d133d051655c956c7089b672f519.mp4. Moviepy - Writing video videos_prepared/46e9d133d051655c956c7089b672f519.mp4 53%|█████▎ | 16/30 [00:04<00:04, 3.50it/s] 53%|█████▎ | 16/30 [00:04<00:04, 3.50it/s] 53%|█████▎ | 16/30 [00:04<00:04, 3.50it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 92%|█████████▎| 37/40 [00:00<00:00, 363.67it/s, now=None] Moviepy - Done ! 53%|█████▎ | 16/30 [00:04<00:04, 3.50it/s] Moviepy - video ready videos_prepared/46e9d133d051655c956c7089b672f519.mp4 53%|█████▎ | 16/30 [00:04<00:04, 3.50it/s] Processing: videos_input/46f4eee0864dd89c9225367d826a657f.mp4 videos_input/46f4eee0864dd89c9225367d826a657f.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/46f4eee0864dd89c9225367d826a657f.txt to videos_prepared/46f4eee0864dd89c9225367d826a657f.txt Moviepy - Building video videos_prepared/46f4eee0864dd89c9225367d826a657f.mp4. Moviepy - Writing video videos_prepared/46f4eee0864dd89c9225367d826a657f.mp4 57%|█████▋ | 17/30 [00:04<00:03, 3.47it/s] 57%|█████▋ | 17/30 [00:04<00:03, 3.47it/s] 57%|█████▋ | 17/30 [00:04<00:03, 3.47it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] 57%|█████▋ | 17/30 [00:05<00:03, 3.47it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/46f4eee0864dd89c9225367d826a657f.mp4 57%|█████▋ | 17/30 [00:05<00:03, 3.47it/s] Processing: videos_input/58b88d44575e945cd7dcd11b3aac6ff0.mp4 videos_input/58b88d44575e945cd7dcd11b3aac6ff0.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/58b88d44575e945cd7dcd11b3aac6ff0.txt to videos_prepared/58b88d44575e945cd7dcd11b3aac6ff0.txt Moviepy - Building video videos_prepared/58b88d44575e945cd7dcd11b3aac6ff0.mp4. 60%|██████ | 18/30 [00:05<00:03, 3.42it/s] 60%|██████ | 18/30 [00:05<00:03, 3.42it/s] Moviepy - Writing video videos_prepared/58b88d44575e945cd7dcd11b3aac6ff0.mp4 60%|██████ | 18/30 [00:05<00:03, 3.42it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 100%|██████████| 40/40 [00:00<00:00, 399.20it/s, now=None] 60%|██████ | 18/30 [00:05<00:03, 3.42it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/58b88d44575e945cd7dcd11b3aac6ff0.mp4 60%|██████ | 18/30 [00:05<00:03, 3.42it/s] Processing: videos_input/81c5dab878d73e6c21181d18d83f2808.mp4 videos_input/81c5dab878d73e6c21181d18d83f2808.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/81c5dab878d73e6c21181d18d83f2808.txt to videos_prepared/81c5dab878d73e6c21181d18d83f2808.txt Moviepy - Building video videos_prepared/81c5dab878d73e6c21181d18d83f2808.mp4. Moviepy - Writing video videos_prepared/81c5dab878d73e6c21181d18d83f2808.mp4 63%|██████▎ | 19/30 [00:05<00:03, 3.49it/s] 63%|██████▎ | 19/30 [00:05<00:03, 3.49it/s] 63%|██████▎ | 19/30 [00:05<00:03, 3.49it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 100%|██████████| 40/40 [00:00<00:00, 398.49it/s, now=None] 63%|██████▎ | 19/30 [00:05<00:03, 3.49it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/81c5dab878d73e6c21181d18d83f2808.mp4 63%|██████▎ | 19/30 [00:05<00:03, 3.49it/s] Processing: videos_input/96d342ea7c7cfddbe1106072bc34be5a.mp4 videos_input/96d342ea7c7cfddbe1106072bc34be5a.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/96d342ea7c7cfddbe1106072bc34be5a.txt to videos_prepared/96d342ea7c7cfddbe1106072bc34be5a.txt Moviepy - Building video videos_prepared/96d342ea7c7cfddbe1106072bc34be5a.mp4. Moviepy - Writing video videos_prepared/96d342ea7c7cfddbe1106072bc34be5a.mp4 67%|██████▋ | 20/30 [00:05<00:02, 3.53it/s] 67%|██████▋ | 20/30 [00:05<00:02, 3.53it/s] 67%|██████▋ | 20/30 [00:05<00:02, 3.53it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 98%|█████████▊| 39/40 [00:00<00:00, 389.71it/s, now=None] 67%|██████▋ | 20/30 [00:05<00:02, 3.53it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/96d342ea7c7cfddbe1106072bc34be5a.mp4 67%|██████▋ | 20/30 [00:05<00:02, 3.53it/s] Processing: videos_input/0288f3d69c08e816d81b014da620db49.mp4 videos_input/0288f3d69c08e816d81b014da620db49.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/0288f3d69c08e816d81b014da620db49.txt to videos_prepared/0288f3d69c08e816d81b014da620db49.txt Moviepy - Building video videos_prepared/0288f3d69c08e816d81b014da620db49.mp4. Moviepy - Writing video videos_prepared/0288f3d69c08e816d81b014da620db49.mp4 70%|███████ | 21/30 [00:05<00:02, 3.53it/s] 70%|███████ | 21/30 [00:05<00:02, 3.53it/s] 70%|███████ | 21/30 [00:05<00:02, 3.53it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 92%|█████████▎| 37/40 [00:00<00:00, 365.51it/s, now=None] 70%|███████ | 21/30 [00:06<00:02, 3.53it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/0288f3d69c08e816d81b014da620db49.mp4 70%|███████ | 21/30 [00:06<00:02, 3.53it/s] Processing: videos_input/328fc12cf9cf3d540e67efadeb893f61.mp4 videos_input/328fc12cf9cf3d540e67efadeb893f61.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/328fc12cf9cf3d540e67efadeb893f61.txt to videos_prepared/328fc12cf9cf3d540e67efadeb893f61.txt Moviepy - Building video videos_prepared/328fc12cf9cf3d540e67efadeb893f61.mp4. Moviepy - Writing video videos_prepared/328fc12cf9cf3d540e67efadeb893f61.mp4 73%|███████▎ | 22/30 [00:06<00:02, 3.48it/s] 73%|███████▎ | 22/30 [00:06<00:02, 3.48it/s] 73%|███████▎ | 22/30 [00:06<00:02, 3.48it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] 73%|███████▎ | 22/30 [00:06<00:02, 3.48it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/328fc12cf9cf3d540e67efadeb893f61.mp4 73%|███████▎ | 22/30 [00:06<00:02, 3.48it/s] Processing: videos_input/383cb4b496d17695554655f3ec79c587.mp4 videos_input/383cb4b496d17695554655f3ec79c587.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/383cb4b496d17695554655f3ec79c587.txt to videos_prepared/383cb4b496d17695554655f3ec79c587.txt Moviepy - Building video videos_prepared/383cb4b496d17695554655f3ec79c587.mp4. Moviepy - Writing video videos_prepared/383cb4b496d17695554655f3ec79c587.mp4 77%|███████▋ | 23/30 [00:06<00:02, 3.46it/s] 77%|███████▋ | 23/30 [00:06<00:02, 3.46it/s] 77%|███████▋ | 23/30 [00:06<00:02, 3.46it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 98%|█████████▊| 39/40 [00:00<00:00, 388.65it/s, now=None] 77%|███████▋ | 23/30 [00:06<00:02, 3.46it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/383cb4b496d17695554655f3ec79c587.mp4 77%|███████▋ | 23/30 [00:06<00:02, 3.46it/s] Processing: videos_input/485b43aa4524327f3c7a40d28e1cf7bc.mp4 videos_input/485b43aa4524327f3c7a40d28e1cf7bc.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/485b43aa4524327f3c7a40d28e1cf7bc.txt to videos_prepared/485b43aa4524327f3c7a40d28e1cf7bc.txt Moviepy - Building video videos_prepared/485b43aa4524327f3c7a40d28e1cf7bc.mp4. 80%|████████ | 24/30 [00:06<00:01, 3.46it/s] 80%|████████ | 24/30 [00:06<00:01, 3.46it/s] Moviepy - Writing video videos_prepared/485b43aa4524327f3c7a40d28e1cf7bc.mp4 80%|████████ | 24/30 [00:06<00:01, 3.46it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] 80%|████████ | 24/30 [00:07<00:01, 3.46it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/485b43aa4524327f3c7a40d28e1cf7bc.mp4 80%|████████ | 24/30 [00:07<00:01, 3.46it/s] Processing: videos_input/560c6472660330638c2809d823d59be3.mp4 videos_input/560c6472660330638c2809d823d59be3.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/560c6472660330638c2809d823d59be3.txt to videos_prepared/560c6472660330638c2809d823d59be3.txt Moviepy - Building video videos_prepared/560c6472660330638c2809d823d59be3.mp4. Moviepy - Writing video videos_prepared/560c6472660330638c2809d823d59be3.mp4 83%|████████▎ | 25/30 [00:07<00:01, 3.47it/s] 83%|████████▎ | 25/30 [00:07<00:01, 3.47it/s] 83%|████████▎ | 25/30 [00:07<00:01, 3.47it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 95%|█████████▌| 38/40 [00:00<00:00, 378.13it/s, now=None] 83%|████████▎ | 25/30 [00:07<00:01, 3.47it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/560c6472660330638c2809d823d59be3.mp4 83%|████████▎ | 25/30 [00:07<00:01, 3.47it/s] Processing: videos_input/614cf13ae1974436cf4072a5cc7d7c57.mp4 videos_input/614cf13ae1974436cf4072a5cc7d7c57.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/614cf13ae1974436cf4072a5cc7d7c57.txt to videos_prepared/614cf13ae1974436cf4072a5cc7d7c57.txt Moviepy - Building video videos_prepared/614cf13ae1974436cf4072a5cc7d7c57.mp4. Moviepy - Writing video videos_prepared/614cf13ae1974436cf4072a5cc7d7c57.mp4 87%|████████▋ | 26/30 [00:07<00:01, 3.47it/s] 87%|████████▋ | 26/30 [00:07<00:01, 3.47it/s] 87%|████████▋ | 26/30 [00:07<00:01, 3.47it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 100%|██████████| 40/40 [00:00<00:00, 396.80it/s, now=None] 87%|████████▋ | 26/30 [00:07<00:01, 3.47it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/614cf13ae1974436cf4072a5cc7d7c57.mp4 87%|████████▋ | 26/30 [00:07<00:01, 3.47it/s] Processing: videos_input/1151c01bd77450dfc603a2eb7352822e.mp4 videos_input/1151c01bd77450dfc603a2eb7352822e.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/1151c01bd77450dfc603a2eb7352822e.txt to videos_prepared/1151c01bd77450dfc603a2eb7352822e.txt 90%|█████████ | 27/30 [00:07<00:00, 3.50it/s] 90%|█████████ | 27/30 [00:07<00:00, 3.50it/s] Moviepy - Building video videos_prepared/1151c01bd77450dfc603a2eb7352822e.mp4. Moviepy - Writing video videos_prepared/1151c01bd77450dfc603a2eb7352822e.mp4 90%|█████████ | 27/30 [00:07<00:00, 3.50it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] 90%|█████████ | 27/30 [00:07<00:00, 3.50it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/1151c01bd77450dfc603a2eb7352822e.mp4 90%|█████████ | 27/30 [00:07<00:00, 3.50it/s] Processing: videos_input/2325e5f8e287753e50e47ab2fc2e8241.mp4 videos_input/2325e5f8e287753e50e47ab2fc2e8241.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/2325e5f8e287753e50e47ab2fc2e8241.txt to videos_prepared/2325e5f8e287753e50e47ab2fc2e8241.txt Moviepy - Building video videos_prepared/2325e5f8e287753e50e47ab2fc2e8241.mp4. 93%|█████████▎| 28/30 [00:07<00:00, 3.48it/s] 93%|█████████▎| 28/30 [00:07<00:00, 3.48it/s] Moviepy - Writing video videos_prepared/2325e5f8e287753e50e47ab2fc2e8241.mp4 93%|█████████▎| 28/30 [00:07<00:00, 3.48it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 98%|█████████▊| 39/40 [00:00<00:00, 388.53it/s, now=None] 93%|█████████▎| 28/30 [00:08<00:00, 3.48it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/2325e5f8e287753e50e47ab2fc2e8241.mp4 93%|█████████▎| 28/30 [00:08<00:00, 3.48it/s] Processing: videos_input/3108dd567bd8669967bc83e0bc50dab2.mp4 videos_input/3108dd567bd8669967bc83e0bc50dab2.mp4 as target resolution 480x848 is larger than input 640x360. So, upsampling the video. Copied videos_input/3108dd567bd8669967bc83e0bc50dab2.txt to videos_prepared/3108dd567bd8669967bc83e0bc50dab2.txt Moviepy - Building video videos_prepared/3108dd567bd8669967bc83e0bc50dab2.mp4. Moviepy - Writing video videos_prepared/3108dd567bd8669967bc83e0bc50dab2.mp4 97%|█████████▋| 29/30 [00:08<00:00, 3.53it/s] 97%|█████████▋| 29/30 [00:08<00:00, 3.53it/s] 97%|█████████▋| 29/30 [00:08<00:00, 3.53it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] 97%|█████████▋| 29/30 [00:08<00:00, 3.53it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/3108dd567bd8669967bc83e0bc50dab2.mp4 97%|█████████▋| 29/30 [00:08<00:00, 3.53it/s] 100%|██████████| 30/30 [00:08<00:00, 3.65it/s] 100%|██████████| 30/30 [00:08<00:00, 3.53it/s] ---Starting to Embed videos--- Loading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s] Loading checkpoint shards: 50%|█████ | 1/2 [00:00<00:00, 1.78it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:01<00:00, 1.90it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:01<00:00, 1.88it/s] Loading pipeline components...: 0%| | 0/3 [00:00<?, ?it/s] Loading pipeline components...: 100%|██████████| 3/3 [00:00<00:00, 651.69it/s] Processing videos_prepared/0288f3d69c08e816d81b014da620db49.mp4 Trimmed video from 40 to first 37 frames 0it [00:00, ?it/s] Processing videos_prepared/05a234b0164d015d468f2f53e771b4cf.mp4 Trimmed video from 40 to first 37 frames 1it [00:01, 1.40s/it] Processing videos_prepared/05ccfa61ece031e881d173289761cf91.mp4 Trimmed video from 40 to first 37 frames 2it [00:02, 1.14s/it] Processing videos_prepared/0bb5f6dbf8ed2e0060f0ac4164b24847.mp4 Trimmed video from 40 to first 37 frames 3it [00:03, 1.05s/it] Processing videos_prepared/1151c01bd77450dfc603a2eb7352822e.mp4 Trimmed video from 40 to first 37 frames 4it [00:04, 1.01s/it] Processing videos_prepared/12e51adf1acbf7acbb703a96a464a39b.mp4 Trimmed video from 40 to first 37 frames 5it [00:05, 1.01it/s] Processing videos_prepared/1d50a3d9703f152758d5422c8b48010f.mp4 Trimmed video from 40 to first 37 frames 6it [00:06, 1.02it/s] Processing videos_prepared/2325e5f8e287753e50e47ab2fc2e8241.mp4 Trimmed video from 40 to first 37 frames 7it [00:07, 1.03it/s] Processing videos_prepared/2c1ed5408882479b06681f7cf372916a.mp4 Trimmed video from 40 to first 37 frames 8it [00:08, 1.04it/s] Processing videos_prepared/3108dd567bd8669967bc83e0bc50dab2.mp4 Trimmed video from 40 to first 37 frames 9it [00:09, 1.01it/s] Processing videos_prepared/328fc12cf9cf3d540e67efadeb893f61.mp4 Trimmed video from 40 to first 37 frames 10it [00:10, 1.02it/s] Processing videos_prepared/383cb4b496d17695554655f3ec79c587.mp4 Trimmed video from 40 to first 37 frames 11it [00:11, 1.00s/it] Processing videos_prepared/3f0979e6cae25447f416372c49ad5e07.mp4 Trimmed video from 40 to first 37 frames 12it [00:12, 1.02it/s] Processing videos_prepared/46e9d133d051655c956c7089b672f519.mp4 Trimmed video from 40 to first 37 frames 13it [00:12, 1.03it/s] Processing videos_prepared/46f4eee0864dd89c9225367d826a657f.mp4 Trimmed video from 40 to first 37 frames 14it [00:13, 1.04it/s] Processing videos_prepared/485b43aa4524327f3c7a40d28e1cf7bc.mp4 Trimmed video from 40 to first 37 frames 15it [00:14, 1.04it/s] Processing videos_prepared/4adbb3a2945c9edd78785daccfd23e80.mp4 Trimmed video from 40 to first 37 frames 16it [00:15, 1.04it/s] Processing videos_prepared/4c918b917308ff03120e9e86650a2d3c.mp4 Trimmed video from 40 to first 37 frames 17it [00:16, 1.05it/s] Processing videos_prepared/560c6472660330638c2809d823d59be3.mp4 Trimmed video from 40 to first 37 frames 18it [00:17, 1.05it/s] Processing videos_prepared/58b88d44575e945cd7dcd11b3aac6ff0.mp4 Trimmed video from 40 to first 37 frames 19it [00:18, 1.02it/s] Processing videos_prepared/5a0229ffdb3bd9d8e81dca7988d7cdbb.mp4 Trimmed video from 40 to first 37 frames 20it [00:19, 1.03it/s] Processing videos_prepared/614cf13ae1974436cf4072a5cc7d7c57.mp4 Trimmed video from 40 to first 37 frames 21it [00:20, 1.04it/s] Processing videos_prepared/7d6dcf13f5c3d45b85c5ea0544c429e4.mp4 Trimmed video from 40 to first 37 frames 22it [00:21, 1.05it/s] Processing videos_prepared/7fe0c83572de828da1cab0c118dece14.mp4 Trimmed video from 40 to first 37 frames 23it [00:22, 1.05it/s] Processing videos_prepared/81c5dab878d73e6c21181d18d83f2808.mp4 Trimmed video from 40 to first 37 frames 24it [00:23, 1.05it/s] Processing videos_prepared/8adfde998361b1d7c6f38a35481667fd.mp4 Trimmed video from 40 to first 37 frames 25it [00:24, 1.05it/s] Processing videos_prepared/8ae679ab483ab344c881d4a813e0cb51.mp4 Trimmed video from 40 to first 37 frames 26it [00:25, 1.05it/s] Processing videos_prepared/8d616fee8e0a280d2d87e478b948a729.mp4 Trimmed video from 40 to first 37 frames 27it [00:26, 1.06it/s] Processing videos_prepared/8e7722634784cf969c15f4a597f3af4d.mp4 Trimmed video from 40 to first 37 frames 28it [00:27, 1.05it/s] Processing videos_prepared/96d342ea7c7cfddbe1106072bc34be5a.mp4 Trimmed video from 40 to first 37 frames 29it [00:28, 1.02it/s] 30it [00:29, 1.03it/s] 30it [00:29, 1.02it/s] ---Starting training--- Found 30 training videos in videos_prepared Loaded 30/30 valid file pairs. ===== Memory before training ===== memory_allocated=18.903 GB max_memory_allocated=18.903 GB max_memory_reserved=28.078 GB ***** Running training ***** Num trainable parameters = 19005440 Num examples = 30 Num batches each epoch = 30 Num epochs = 34 Instantaneous batch size per device = 1 Total train batch size (w. parallel, distributed & accumulation) = 1 Total optimization steps = 1000 Steps: 0%| | 0/1000 [00:00<?, ?it/s]W1211 15:09:46.660000 135675630435840 torch/fx/experimental/symbolic_shapes.py:4449] [0/0] xindex is not in var_ranges, defaulting to unknown range. W1211 15:09:46.674000 135675630435840 torch/fx/experimental/symbolic_shapes.py:4449] [0/0] xindex is not in var_ranges, defaulting to unknown range. W1211 15:09:46.812000 135675630435840 torch/fx/experimental/symbolic_shapes.py:4449] [0/0] xindex is not in var_ranges, defaulting to unknown range. Steps: 0%| | 1/1000 [04:18<71:48:02, 258.74s/it] Steps: 0%| | 1/1000 [04:18<71:48:02, 258.74s/it, loss=1.07, lr=2e-6] Steps: 0%| | 2/1000 [04:20<29:50:05, 107.62s/it, loss=1.07, lr=2e-6] Steps: 0%| | 2/1000 [04:20<29:50:05, 107.62s/it, loss=0.666, lr=4e-6] Steps: 0%| | 3/1000 [04:22<16:25:48, 59.33s/it, loss=0.666, lr=4e-6] Steps: 0%| | 3/1000 [04:22<16:25:48, 59.33s/it, loss=0.335, lr=6e-6] Steps: 0%| | 4/1000 [04:24<10:08:13, 36.64s/it, loss=0.335, lr=6e-6] Steps: 0%| | 4/1000 [04:24<10:08:13, 36.64s/it, loss=0.362, lr=8e-6] Steps: 0%| | 5/1000 [04:26<6:39:34, 24.09s/it, loss=0.362, lr=8e-6] Steps: 0%| | 5/1000 [04:26<6:39:34, 24.09s/it, loss=0.905, lr=1e-5] Steps: 1%| | 6/1000 [04:28<4:33:55, 16.53s/it, loss=0.905, lr=1e-5] Steps: 1%| | 6/1000 [04:28<4:33:55, 16.53s/it, loss=0.767, lr=1.2e-5] Steps: 1%| | 7/1000 [04:29<3:14:14, 11.74s/it, loss=0.767, lr=1.2e-5] Steps: 1%| | 7/1000 [04:29<3:14:14, 11.74s/it, loss=0.973, lr=1.4e-5] Steps: 1%| | 8/1000 [04:31<2:22:04, 8.59s/it, loss=0.973, lr=1.4e-5] Steps: 1%| | 8/1000 [04:31<2:22:04, 8.59s/it, loss=0.821, lr=1.6e-5] Steps: 1%| | 9/1000 [04:33<1:47:09, 6.49s/it, loss=0.821, lr=1.6e-5] Steps: 1%| | 9/1000 [04:33<1:47:09, 6.49s/it, loss=0.472, lr=1.8e-5] Steps: 1%| | 10/1000 [04:35<1:23:29, 5.06s/it, loss=0.472, lr=1.8e-5] Steps: 1%| | 10/1000 [04:35<1:23:29, 5.06s/it, loss=0.358, lr=2e-5] Steps: 1%| | 11/1000 [04:37<1:07:16, 4.08s/it, loss=0.358, lr=2e-5] Steps: 1%| | 11/1000 [04:37<1:07:16, 4.08s/it, loss=0.332, lr=2.2e-5] Steps: 1%| | 12/1000 [04:39<56:04, 3.41s/it, loss=0.332, lr=2.2e-5] Steps: 1%| | 12/1000 [04:39<56:04, 3.41s/it, loss=0.353, lr=2.4e-5] Steps: 1%|▏ | 13/1000 [04:41<48:22, 2.94s/it, loss=0.353, lr=2.4e-5] Steps: 1%|▏ | 13/1000 [04:41<48:22, 2.94s/it, loss=0.346, lr=2.6e-5] Steps: 1%|▏ | 14/1000 [04:42<42:57, 2.61s/it, loss=0.346, lr=2.6e-5] Steps: 1%|▏ | 14/1000 [04:42<42:57, 2.61s/it, loss=0.499, lr=2.8e-5] Steps: 2%|▏ | 15/1000 [04:44<39:12, 2.39s/it, loss=0.499, lr=2.8e-5] Steps: 2%|▏ | 15/1000 [04:44<39:12, 2.39s/it, loss=1.07, lr=3e-5] Steps: 2%|▏ | 16/1000 [04:46<36:35, 2.23s/it, loss=1.07, lr=3e-5] Steps: 2%|▏ | 16/1000 [04:46<36:35, 2.23s/it, loss=0.448, lr=3.2e-5] Steps: 2%|▏ | 17/1000 [04:48<34:44, 2.12s/it, loss=0.448, lr=3.2e-5] Steps: 2%|▏ | 17/1000 [04:48<34:44, 2.12s/it, loss=0.752, lr=3.4e-5] Steps: 2%|▏ | 18/1000 [04:50<33:26, 2.04s/it, loss=0.752, lr=3.4e-5] Steps: 2%|▏ | 18/1000 [04:50<33:26, 2.04s/it, loss=0.33, lr=3.6e-5] Steps: 2%|▏ | 19/1000 [04:52<32:30, 1.99s/it, loss=0.33, lr=3.6e-5] Steps: 2%|▏ | 19/1000 [04:52<32:30, 1.99s/it, loss=0.873, lr=3.8e-5] Steps: 2%|▏ | 20/1000 [04:54<31:52, 1.95s/it, loss=0.873, lr=3.8e-5] Steps: 2%|▏ | 20/1000 [04:54<31:52, 1.95s/it, loss=0.499, lr=4e-5] Steps: 2%|▏ | 21/1000 [04:55<31:25, 1.93s/it, loss=0.499, lr=4e-5] Steps: 2%|▏ | 21/1000 [04:55<31:25, 1.93s/it, loss=0.55, lr=4.2e-5] Steps: 2%|▏ | 22/1000 [04:57<31:07, 1.91s/it, loss=0.55, lr=4.2e-5] Steps: 2%|▏ | 22/1000 [04:57<31:07, 1.91s/it, loss=0.304, lr=4.4e-5] Steps: 2%|▏ | 23/1000 [04:59<30:52, 1.90s/it, loss=0.304, lr=4.4e-5] Steps: 2%|▏ | 23/1000 [04:59<30:52, 1.90s/it, loss=0.42, lr=4.6e-5] Steps: 2%|▏ | 24/1000 [05:01<30:41, 1.89s/it, loss=0.42, lr=4.6e-5] Steps: 2%|▏ | 24/1000 [05:01<30:41, 1.89s/it, loss=0.442, lr=4.8e-5] Steps: 2%|▎ | 25/1000 [05:03<30:34, 1.88s/it, loss=0.442, lr=4.8e-5] Steps: 2%|▎ | 25/1000 [05:03<30:34, 1.88s/it, loss=0.386, lr=5e-5] Steps: 3%|▎ | 26/1000 [05:05<30:28, 1.88s/it, loss=0.386, lr=5e-5] Steps: 3%|▎ | 26/1000 [05:05<30:28, 1.88s/it, loss=0.453, lr=5.2e-5] Steps: 3%|▎ | 27/1000 [05:07<30:22, 1.87s/it, loss=0.453, lr=5.2e-5] Steps: 3%|▎ | 27/1000 [05:07<30:22, 1.87s/it, loss=0.524, lr=5.4e-5] Steps: 3%|▎ | 28/1000 [05:09<30:18, 1.87s/it, loss=0.524, lr=5.4e-5] Steps: 3%|▎ | 28/1000 [05:09<30:18, 1.87s/it, loss=0.853, lr=5.6e-5] Steps: 3%|▎ | 29/1000 [05:10<30:14, 1.87s/it, loss=0.853, lr=5.6e-5] Steps: 3%|▎ | 29/1000 [05:10<30:14, 1.87s/it, loss=0.383, lr=5.8e-5] Steps: 3%|▎ | 30/1000 [05:12<30:12, 1.87s/it, loss=0.383, lr=5.8e-5] Steps: 3%|▎ | 30/1000 [05:12<30:12, 1.87s/it, loss=0.674, lr=6e-5] Steps: 3%|▎ | 31/1000 [05:20<58:22, 3.61s/it, loss=0.674, lr=6e-5] Steps: 3%|▎ | 31/1000 [05:20<58:22, 3.61s/it, loss=0.638, lr=6.2e-5] Steps: 3%|▎ | 32/1000 [05:22<49:53, 3.09s/it, loss=0.638, lr=6.2e-5] Steps: 3%|▎ | 32/1000 [05:22<49:53, 3.09s/it, loss=1.04, lr=6.4e-5] Steps: 3%|▎ | 33/1000 [05:24<43:53, 2.72s/it, loss=1.04, lr=6.4e-5] Steps: 3%|▎ | 33/1000 [05:24<43:53, 2.72s/it, loss=0.504, lr=6.6e-5] Steps: 3%|▎ | 34/1000 [05:26<39:41, 2.47s/it, loss=0.504, lr=6.6e-5] Steps: 3%|▎ | 34/1000 [05:26<39:41, 2.47s/it, loss=0.638, lr=6.8e-5] Steps: 4%|▎ | 35/1000 [05:27<36:45, 2.29s/it, loss=0.638, lr=6.8e-5] Steps: 4%|▎ | 35/1000 [05:27<36:45, 2.29s/it, loss=1.01, lr=7e-5] Steps: 4%|▎ | 36/1000 [05:29<34:41, 2.16s/it, loss=1.01, lr=7e-5] Steps: 4%|▎ | 36/1000 [05:29<34:41, 2.16s/it, loss=1.03, lr=7.2e-5] Steps: 4%|▎ | 37/1000 [05:31<33:16, 2.07s/it, loss=1.03, lr=7.2e-5] Steps: 4%|▎ | 37/1000 [05:31<33:16, 2.07s/it, loss=0.447, lr=7.4e-5] Steps: 4%|▍ | 38/1000 [05:33<32:15, 2.01s/it, loss=0.447, lr=7.4e-5] Steps: 4%|▍ | 38/1000 [05:33<32:15, 2.01s/it, loss=0.56, lr=7.6e-5] Steps: 4%|▍ | 39/1000 [05:35<31:30, 1.97s/it, loss=0.56, lr=7.6e-5] Steps: 4%|▍ | 39/1000 [05:35<31:30, 1.97s/it, loss=0.317, lr=7.8e-5] Steps: 4%|▍ | 40/1000 [05:37<30:58, 1.94s/it, loss=0.317, lr=7.8e-5] Steps: 4%|▍ | 40/1000 [05:37<30:58, 1.94s/it, loss=0.787, lr=8e-5] Steps: 4%|▍ | 41/1000 [05:39<30:38, 1.92s/it, loss=0.787, lr=8e-5] Steps: 4%|▍ | 41/1000 [05:39<30:38, 1.92s/it, loss=0.309, lr=8.2e-5] Steps: 4%|▍ | 42/1000 [05:40<30:21, 1.90s/it, loss=0.309, lr=8.2e-5] Steps: 4%|▍ | 42/1000 [05:40<30:21, 1.90s/it, loss=0.805, lr=8.4e-5] Steps: 4%|▍ | 43/1000 [05:42<30:09, 1.89s/it, loss=0.805, lr=8.4e-5] Steps: 4%|▍ | 43/1000 [05:42<30:09, 1.89s/it, loss=1.1, lr=8.6e-5] Steps: 4%|▍ | 44/1000 [05:44<30:00, 1.88s/it, loss=1.1, lr=8.6e-5] Steps: 4%|▍ | 44/1000 [05:44<30:00, 1.88s/it, loss=0.307, lr=8.8e-5] Steps: 4%|▍ | 45/1000 [05:46<29:54, 1.88s/it, loss=0.307, lr=8.8e-5] Steps: 4%|▍ | 45/1000 [05:46<29:54, 1.88s/it, loss=0.991, lr=9e-5] Steps: 5%|▍ | 46/1000 [05:48<29:50, 1.88s/it, loss=0.991, lr=9e-5] Steps: 5%|▍ | 46/1000 [05:48<29:50, 1.88s/it, loss=0.431, lr=9.2e-5] Steps: 5%|▍ | 47/1000 [05:50<29:45, 1.87s/it, loss=0.431, lr=9.2e-5] Steps: 5%|▍ | 47/1000 [05:50<29:45, 1.87s/it, loss=0.301, lr=9.4e-5] Steps: 5%|▍ | 48/1000 [05:52<29:42, 1.87s/it, loss=0.301, lr=9.4e-5] Steps: 5%|▍ | 48/1000 [05:52<29:42, 1.87s/it, loss=0.78, lr=9.6e-5] Steps: 5%|▍ | 49/1000 [05:54<29:37, 1.87s/it, loss=0.78, lr=9.6e-5] Steps: 5%|▍ | 49/1000 [05:54<29:37, 1.87s/it, loss=0.699, lr=9.8e-5] Steps: 5%|▌ | 50/1000 [05:55<29:36, 1.87s/it, loss=0.699, lr=9.8e-5] Steps: 5%|▌ | 50/1000 [05:55<29:36, 1.87s/it, loss=0.784, lr=0.0001] Steps: 5%|▌ | 51/1000 [05:57<29:33, 1.87s/it, loss=0.784, lr=0.0001] Steps: 5%|▌ | 51/1000 [05:57<29:33, 1.87s/it, loss=0.487, lr=0.000102] Steps: 5%|▌ | 52/1000 [05:59<29:31, 1.87s/it, loss=0.487, lr=0.000102] Steps: 5%|▌ | 52/1000 [05:59<29:31, 1.87s/it, loss=0.608, lr=0.000104] Steps: 5%|▌ | 53/1000 [06:01<29:30, 1.87s/it, loss=0.608, lr=0.000104] Steps: 5%|▌ | 53/1000 [06:01<29:30, 1.87s/it, loss=0.371, lr=0.000106] Steps: 5%|▌ | 54/1000 [06:03<29:29, 1.87s/it, loss=0.371, lr=0.000106] Steps: 5%|▌ | 54/1000 [06:03<29:29, 1.87s/it, loss=0.302, lr=0.000108] Steps: 6%|▌ | 55/1000 [06:05<29:26, 1.87s/it, loss=0.302, lr=0.000108] Steps: 6%|▌ | 55/1000 [06:05<29:26, 1.87s/it, loss=0.568, lr=0.00011] Steps: 6%|▌ | 56/1000 [06:07<29:25, 1.87s/it, loss=0.568, lr=0.00011] Steps: 6%|▌ | 56/1000 [06:07<29:25, 1.87s/it, loss=0.316, lr=0.000112] Steps: 6%|▌ | 57/1000 [06:08<29:22, 1.87s/it, loss=0.316, lr=0.000112] Steps: 6%|▌ | 57/1000 [06:09<29:22, 1.87s/it, loss=0.611, lr=0.000114] Steps: 6%|▌ | 58/1000 [06:10<29:21, 1.87s/it, loss=0.611, lr=0.000114] Steps: 6%|▌ | 58/1000 [06:10<29:21, 1.87s/it, loss=0.531, lr=0.000116] Steps: 6%|▌ | 59/1000 [06:12<29:20, 1.87s/it, loss=0.531, lr=0.000116] Steps: 6%|▌ | 59/1000 [06:12<29:20, 1.87s/it, loss=0.451, lr=0.000118] Steps: 6%|▌ | 60/1000 [06:14<29:18, 1.87s/it, loss=0.451, lr=0.000118] Steps: 6%|▌ | 60/1000 [06:14<29:18, 1.87s/it, loss=0.353, lr=0.00012] Steps: 6%|▌ | 61/1000 [06:22<56:51, 3.63s/it, loss=0.353, lr=0.00012] Steps: 6%|▌ | 61/1000 [06:22<56:51, 3.63s/it, loss=0.44, lr=0.000122] Steps: 6%|▌ | 62/1000 [06:24<48:31, 3.10s/it, loss=0.44, lr=0.000122] Steps: 6%|▌ | 62/1000 [06:24<48:31, 3.10s/it, loss=0.314, lr=0.000124] Steps: 6%|▋ | 63/1000 [06:26<42:42, 2.73s/it, loss=0.314, lr=0.000124] Steps: 6%|▋ | 63/1000 [06:26<42:42, 2.73s/it, loss=0.364, lr=0.000126] Steps: 6%|▋ | 64/1000 [06:27<38:36, 2.47s/it, loss=0.364, lr=0.000126] Steps: 6%|▋ | 64/1000 [06:27<38:36, 2.47s/it, loss=0.35, lr=0.000128] Steps: 6%|▋ | 65/1000 [06:29<35:42, 2.29s/it, loss=0.35, lr=0.000128] Steps: 6%|▋ | 65/1000 [06:29<35:42, 2.29s/it, loss=0.293, lr=0.00013] Steps: 7%|▋ | 66/1000 [06:31<33:40, 2.16s/it, loss=0.293, lr=0.00013] Steps: 7%|▋ | 66/1000 [06:31<33:40, 2.16s/it, loss=0.978, lr=0.000132] Steps: 7%|▋ | 67/1000 [06:33<32:14, 2.07s/it, loss=0.978, lr=0.000132] Steps: 7%|▋ | 67/1000 [06:33<32:14, 2.07s/it, loss=0.847, lr=0.000134] Steps: 7%|▋ | 68/1000 [06:35<31:15, 2.01s/it, loss=0.847, lr=0.000134] Steps: 7%|▋ | 68/1000 [06:35<31:15, 2.01s/it, loss=0.442, lr=0.000136] Steps: 7%|▋ | 69/1000 [06:37<30:34, 1.97s/it, loss=0.442, lr=0.000136] Steps: 7%|▋ | 69/1000 [06:37<30:34, 1.97s/it, loss=0.295, lr=0.000138] Steps: 7%|▋ | 70/1000 [06:39<30:03, 1.94s/it, loss=0.295, lr=0.000138] Steps: 7%|▋ | 70/1000 [06:39<30:03, 1.94s/it, loss=0.314, lr=0.00014] Steps: 7%|▋ | 71/1000 [06:41<29:42, 1.92s/it, loss=0.314, lr=0.00014] Steps: 7%|▋ | 71/1000 [06:41<29:42, 1.92s/it, loss=1.03, lr=0.000142] Steps: 7%|▋ | 72/1000 [06:42<29:27, 1.90s/it, loss=1.03, lr=0.000142] Steps: 7%|▋ | 72/1000 [06:42<29:27, 1.90s/it, loss=0.524, lr=0.000144] Steps: 7%|▋ | 73/1000 [06:44<29:15, 1.89s/it, loss=0.524, lr=0.000144] Steps: 7%|▋ | 73/1000 [06:44<29:15, 1.89s/it, loss=0.3, lr=0.000146] Steps: 7%|▋ | 74/1000 [06:46<29:06, 1.89s/it, loss=0.3, lr=0.000146] Steps: 7%|▋ | 74/1000 [06:46<29:06, 1.89s/it, loss=0.374, lr=0.000148] Steps: 8%|▊ | 75/1000 [06:48<28:59, 1.88s/it, loss=0.374, lr=0.000148] Steps: 8%|▊ | 75/1000 [06:48<28:59, 1.88s/it, loss=0.328, lr=0.00015] Steps: 8%|▊ | 76/1000 [06:50<28:53, 1.88s/it, loss=0.328, lr=0.00015] Steps: 8%|▊ | 76/1000 [06:50<28:53, 1.88s/it, loss=0.547, lr=0.000152] Steps: 8%|▊ | 77/1000 [06:52<28:50, 1.88s/it, loss=0.547, lr=0.000152] Steps: 8%|▊ | 77/1000 [06:52<28:50, 1.88s/it, loss=0.301, lr=0.000154] Steps: 8%|▊ | 78/1000 [06:54<28:46, 1.87s/it, loss=0.301, lr=0.000154] Steps: 8%|▊ | 78/1000 [06:54<28:46, 1.87s/it, loss=1.02, lr=0.000156] Steps: 8%|▊ | 79/1000 [06:55<28:44, 1.87s/it, loss=1.02, lr=0.000156] Steps: 8%|▊ | 79/1000 [06:56<28:44, 1.87s/it, loss=0.303, lr=0.000158] Steps: 8%|▊ | 80/1000 [06:57<28:41, 1.87s/it, loss=0.303, lr=0.000158] Steps: 8%|▊ | 80/1000 [06:57<28:41, 1.87s/it, loss=0.386, lr=0.00016] Steps: 8%|▊ | 81/1000 [06:59<28:40, 1.87s/it, loss=0.386, lr=0.00016] Steps: 8%|▊ | 81/1000 [06:59<28:40, 1.87s/it, loss=0.399, lr=0.000162] Steps: 8%|▊ | 82/1000 [07:01<28:36, 1.87s/it, loss=0.399, lr=0.000162] Steps: 8%|▊ | 82/1000 [07:01<28:36, 1.87s/it, loss=0.47, lr=0.000164] Steps: 8%|▊ | 83/1000 [07:03<28:34, 1.87s/it, loss=0.47, lr=0.000164] Steps: 8%|▊ | 83/1000 [07:03<28:34, 1.87s/it, loss=0.909, lr=0.000166] Steps: 8%|▊ | 84/1000 [07:05<28:33, 1.87s/it, loss=0.909, lr=0.000166] Steps: 8%|▊ | 84/1000 [07:05<28:33, 1.87s/it, loss=0.284, lr=0.000168] Steps: 8%|▊ | 85/1000 [07:07<28:32, 1.87s/it, loss=0.284, lr=0.000168] Steps: 8%|▊ | 85/1000 [07:07<28:32, 1.87s/it, loss=0.52, lr=0.00017] Steps: 9%|▊ | 86/1000 [07:09<28:30, 1.87s/it, loss=0.52, lr=0.00017] Steps: 9%|▊ | 86/1000 [07:09<28:30, 1.87s/it, loss=0.286, lr=0.000172] Steps: 9%|▊ | 87/1000 [07:10<28:27, 1.87s/it, loss=0.286, lr=0.000172] Steps: 9%|▊ | 87/1000 [07:10<28:27, 1.87s/it, loss=0.642, lr=0.000174] Steps: 9%|▉ | 88/1000 [07:12<28:24, 1.87s/it, loss=0.642, lr=0.000174] Steps: 9%|▉ | 88/1000 [07:12<28:24, 1.87s/it, loss=0.305, lr=0.000176] Steps: 9%|▉ | 89/1000 [07:14<28:23, 1.87s/it, loss=0.305, lr=0.000176] Steps: 9%|▉ | 89/1000 [07:14<28:23, 1.87s/it, loss=1.01, lr=0.000178] Steps: 9%|▉ | 90/1000 [07:16<28:21, 1.87s/it, loss=1.01, lr=0.000178] Steps: 9%|▉ | 90/1000 [07:16<28:21, 1.87s/it, loss=0.287, lr=0.00018] Steps: 9%|▉ | 91/1000 [07:24<54:57, 3.63s/it, loss=0.287, lr=0.00018] Steps: 9%|▉ | 91/1000 [07:24<54:57, 3.63s/it, loss=0.731, lr=0.000182] Steps: 9%|▉ | 92/1000 [07:26<46:54, 3.10s/it, loss=0.731, lr=0.000182] Steps: 9%|▉ | 92/1000 [07:26<46:54, 3.10s/it, loss=0.585, lr=0.000184] Steps: 9%|▉ | 93/1000 [07:28<41:15, 2.73s/it, loss=0.585, lr=0.000184] Steps: 9%|▉ | 93/1000 [07:28<41:15, 2.73s/it, loss=0.737, lr=0.000186] Steps: 9%|▉ | 94/1000 [07:29<37:18, 2.47s/it, loss=0.737, lr=0.000186] Steps: 9%|▉ | 94/1000 [07:29<37:18, 2.47s/it, loss=0.679, lr=0.000188] Steps: 10%|▉ | 95/1000 [07:31<34:32, 2.29s/it, loss=0.679, lr=0.000188] Steps: 10%|▉ | 95/1000 [07:31<34:32, 2.29s/it, loss=0.305, lr=0.00019] Steps: 10%|▉ | 96/1000 [07:33<32:34, 2.16s/it, loss=0.305, lr=0.00019] Steps: 10%|▉ | 96/1000 [07:33<32:34, 2.16s/it, loss=0.355, lr=0.000192] Steps: 10%|▉ | 97/1000 [07:35<31:12, 2.07s/it, loss=0.355, lr=0.000192] Steps: 10%|▉ | 97/1000 [07:35<31:12, 2.07s/it, loss=0.331, lr=0.000194] Steps: 10%|▉ | 98/1000 [07:37<30:16, 2.01s/it, loss=0.331, lr=0.000194] Steps: 10%|▉ | 98/1000 [07:37<30:16, 2.01s/it, loss=0.954, lr=0.000196] Steps: 10%|▉ | 99/1000 [07:39<29:34, 1.97s/it, loss=0.954, lr=0.000196] Steps: 10%|▉ | 99/1000 [07:39<29:34, 1.97s/it, loss=0.692, lr=0.000198] Steps: 10%|█ | 100/1000 [07:41<29:05, 1.94s/it, loss=0.692, lr=0.000198] Steps: 10%|█ | 100/1000 [07:41<29:05, 1.94s/it, loss=0.329, lr=0.0002] Steps: 10%|█ | 101/1000 [07:42<28:44, 1.92s/it, loss=0.329, lr=0.0002] Steps: 10%|█ | 101/1000 [07:42<28:44, 1.92s/it, loss=0.283, lr=0.000202] Steps: 10%|█ | 102/1000 [07:44<28:30, 1.90s/it, loss=0.283, lr=0.000202] Steps: 10%|█ | 102/1000 [07:44<28:30, 1.90s/it, loss=0.633, lr=0.000204] Steps: 10%|█ | 103/1000 [07:46<28:19, 1.89s/it, loss=0.633, lr=0.000204] Steps: 10%|█ | 103/1000 [07:46<28:19, 1.89s/it, loss=0.355, lr=0.000206] Steps: 10%|█ | 104/1000 [07:48<28:10, 1.89s/it, loss=0.355, lr=0.000206] Steps: 10%|█ | 104/1000 [07:48<28:10, 1.89s/it, loss=1.03, lr=0.000208] Steps: 10%|█ | 105/1000 [07:50<28:04, 1.88s/it, loss=1.03, lr=0.000208] Steps: 10%|█ | 105/1000 [07:50<28:04, 1.88s/it, loss=0.62, lr=0.00021] Steps: 11%|█ | 106/1000 [07:52<27:59, 1.88s/it, loss=0.62, lr=0.00021] Steps: 11%|█ | 106/1000 [07:52<27:59, 1.88s/it, loss=0.404, lr=0.000212] Steps: 11%|█ | 107/1000 [07:54<27:54, 1.87s/it, loss=0.404, lr=0.000212] Steps: 11%|█ | 107/1000 [07:54<27:54, 1.87s/it, loss=0.22, lr=0.000214] Steps: 11%|█ | 108/1000 [07:56<27:52, 1.87s/it, loss=0.22, lr=0.000214] Steps: 11%|█ | 108/1000 [07:56<27:52, 1.87s/it, loss=0.314, lr=0.000216] Steps: 11%|█ | 109/1000 [07:57<27:49, 1.87s/it, loss=0.314, lr=0.000216] Steps: 11%|█ | 109/1000 [07:57<27:49, 1.87s/it, loss=0.704, lr=0.000218] Steps: 11%|█ | 110/1000 [07:59<27:45, 1.87s/it, loss=0.704, lr=0.000218] Steps: 11%|█ | 110/1000 [07:59<27:45, 1.87s/it, loss=0.539, lr=0.00022] Steps: 11%|█ | 111/1000 [08:01<27:43, 1.87s/it, loss=0.539, lr=0.00022] Steps: 11%|█ | 111/1000 [08:01<27:43, 1.87s/it, loss=0.569, lr=0.000222] Steps: 11%|█ | 112/1000 [08:03<27:40, 1.87s/it, loss=0.569, lr=0.000222] Steps: 11%|█ | 112/1000 [08:03<27:40, 1.87s/it, loss=0.591, lr=0.000224] Steps: 11%|█▏ | 113/1000 [08:05<27:39, 1.87s/it, loss=0.591, lr=0.000224] Steps: 11%|█▏ | 113/1000 [08:05<27:39, 1.87s/it, loss=0.32, lr=0.000226] Steps: 11%|█▏ | 114/1000 [08:07<27:36, 1.87s/it, loss=0.32, lr=0.000226] Steps: 11%|█▏ | 114/1000 [08:07<27:36, 1.87s/it, loss=0.462, lr=0.000228] Steps: 12%|█▏ | 115/1000 [08:09<27:34, 1.87s/it, loss=0.462, lr=0.000228] Steps: 12%|█▏ | 115/1000 [08:09<27:34, 1.87s/it, loss=0.409, lr=0.00023] Steps: 12%|█▏ | 116/1000 [08:11<27:32, 1.87s/it, loss=0.409, lr=0.00023] Steps: 12%|█▏ | 116/1000 [08:11<27:32, 1.87s/it, loss=0.943, lr=0.000232] Steps: 12%|█▏ | 117/1000 [08:12<27:30, 1.87s/it, loss=0.943, lr=0.000232] Steps: 12%|█▏ | 117/1000 [08:12<27:30, 1.87s/it, loss=0.33, lr=0.000234] Steps: 12%|█▏ | 118/1000 [08:14<27:28, 1.87s/it, loss=0.33, lr=0.000234] Steps: 12%|█▏ | 118/1000 [08:14<27:28, 1.87s/it, loss=0.447, lr=0.000236] Steps: 12%|█▏ | 119/1000 [08:16<27:27, 1.87s/it, loss=0.447, lr=0.000236] Steps: 12%|█▏ | 119/1000 [08:16<27:27, 1.87s/it, loss=0.929, lr=0.000238] Steps: 12%|█▏ | 120/1000 [08:18<27:24, 1.87s/it, loss=0.929, lr=0.000238] Steps: 12%|█▏ | 120/1000 [08:18<27:24, 1.87s/it, loss=0.908, lr=0.00024] Steps: 12%|█▏ | 121/1000 [08:26<53:04, 3.62s/it, loss=0.908, lr=0.00024] Steps: 12%|█▏ | 121/1000 [08:26<53:04, 3.62s/it, loss=0.81, lr=0.000242] Steps: 12%|█▏ | 122/1000 [08:28<45:18, 3.10s/it, loss=0.81, lr=0.000242] Steps: 12%|█▏ | 122/1000 [08:28<45:18, 3.10s/it, loss=0.315, lr=0.000244] Steps: 12%|█▏ | 123/1000 [08:29<39:51, 2.73s/it, loss=0.315, lr=0.000244] Steps: 12%|█▏ | 123/1000 [08:29<39:51, 2.73s/it, loss=0.311, lr=0.000246] Steps: 12%|█▏ | 124/1000 [08:31<36:03, 2.47s/it, loss=0.311, lr=0.000246] Steps: 12%|█▏ | 124/1000 [08:31<36:03, 2.47s/it, loss=0.634, lr=0.000248] Steps: 12%|█▎ | 125/1000 [08:33<33:20, 2.29s/it, loss=0.634, lr=0.000248] Steps: 12%|█▎ | 125/1000 [08:33<33:20, 2.29s/it, loss=0.728, lr=0.00025] Steps: 13%|█▎ | 126/1000 [08:35<31:28, 2.16s/it, loss=0.728, lr=0.00025] Steps: 13%|█▎ | 126/1000 [08:35<31:28, 2.16s/it, loss=0.38, lr=0.000252] Steps: 13%|█▎ | 127/1000 [08:37<30:09, 2.07s/it, loss=0.38, lr=0.000252] Steps: 13%|█▎ | 127/1000 [08:37<30:09, 2.07s/it, loss=0.335, lr=0.000254] Steps: 13%|█▎ | 128/1000 [08:39<29:14, 2.01s/it, loss=0.335, lr=0.000254] Steps: 13%|█▎ | 128/1000 [08:39<29:14, 2.01s/it, loss=0.41, lr=0.000256] Steps: 13%|█▎ | 129/1000 [08:41<28:36, 1.97s/it, loss=0.41, lr=0.000256] Steps: 13%|█▎ | 129/1000 [08:41<28:36, 1.97s/it, loss=0.336, lr=0.000258] Steps: 13%|█▎ | 130/1000 [08:43<28:06, 1.94s/it, loss=0.336, lr=0.000258] Steps: 13%|█▎ | 130/1000 [08:43<28:06, 1.94s/it, loss=0.8, lr=0.00026] Steps: 13%|█▎ | 131/1000 [08:44<27:46, 1.92s/it, loss=0.8, lr=0.00026] Steps: 13%|█▎ | 131/1000 [08:44<27:46, 1.92s/it, loss=0.97, lr=0.000262] Steps: 13%|█▎ | 132/1000 [08:46<27:31, 1.90s/it, loss=0.97, lr=0.000262] Steps: 13%|█▎ | 132/1000 [08:46<27:31, 1.90s/it, loss=0.688, lr=0.000264] Steps: 13%|█▎ | 133/1000 [08:48<27:21, 1.89s/it, loss=0.688, lr=0.000264] Steps: 13%|█▎ | 133/1000 [08:48<27:21, 1.89s/it, loss=0.557, lr=0.000266] Steps: 13%|█▎ | 134/1000 [08:50<27:13, 1.89s/it, loss=0.557, lr=0.000266] Steps: 13%|█▎ | 134/1000 [08:50<27:13, 1.89s/it, loss=0.548, lr=0.000268] Steps: 14%|█▎ | 135/1000 [08:52<27:07, 1.88s/it, loss=0.548, lr=0.000268] Steps: 14%|█▎ | 135/1000 [08:52<27:07, 1.88s/it, loss=0.355, lr=0.00027] Steps: 14%|█▎ | 136/1000 [08:54<27:04, 1.88s/it, loss=0.355, lr=0.00027] Steps: 14%|█▎ | 136/1000 [08:54<27:04, 1.88s/it, loss=0.873, lr=0.000272] Steps: 14%|█▎ | 137/1000 [08:56<26:59, 1.88s/it, loss=0.873, lr=0.000272] Steps: 14%|█▎ | 137/1000 [08:56<26:59, 1.88s/it, loss=0.217, lr=0.000274] Steps: 14%|█▍ | 138/1000 [08:57<26:55, 1.87s/it, loss=0.217, lr=0.000274] Steps: 14%|█▍ | 138/1000 [08:57<26:55, 1.87s/it, loss=0.332, lr=0.000276] Steps: 14%|█▍ | 139/1000 [08:59<26:52, 1.87s/it, loss=0.332, lr=0.000276] Steps: 14%|█▍ | 139/1000 [08:59<26:52, 1.87s/it, loss=0.547, lr=0.000278] Steps: 14%|█▍ | 140/1000 [09:01<26:49, 1.87s/it, loss=0.547, lr=0.000278] Steps: 14%|█▍ | 140/1000 [09:01<26:49, 1.87s/it, loss=0.644, lr=0.00028] Steps: 14%|█▍ | 141/1000 [09:03<26:47, 1.87s/it, loss=0.644, lr=0.00028] Steps: 14%|█▍ | 141/1000 [09:03<26:47, 1.87s/it, loss=0.493, lr=0.000282] Steps: 14%|█▍ | 142/1000 [09:05<26:45, 1.87s/it, loss=0.493, lr=0.000282] Steps: 14%|█▍ | 142/1000 [09:05<26:45, 1.87s/it, loss=0.339, lr=0.000284] Steps: 14%|█▍ | 143/1000 [09:07<26:42, 1.87s/it, loss=0.339, lr=0.000284] Steps: 14%|█▍ | 143/1000 [09:07<26:42, 1.87s/it, loss=0.47, lr=0.000286] Steps: 14%|█▍ | 144/1000 [09:09<26:41, 1.87s/it, loss=0.47, lr=0.000286] Steps: 14%|█▍ | 144/1000 [09:09<26:41, 1.87s/it, loss=0.236, lr=0.000288] Steps: 14%|█▍ | 145/1000 [09:11<26:39, 1.87s/it, loss=0.236, lr=0.000288] Steps: 14%|█▍ | 145/1000 [09:11<26:39, 1.87s/it, loss=0.722, lr=0.00029] Steps: 15%|█▍ | 146/1000 [09:12<26:36, 1.87s/it, loss=0.722, lr=0.00029] Steps: 15%|█▍ | 146/1000 [09:12<26:36, 1.87s/it, loss=0.636, lr=0.000292] Steps: 15%|█▍ | 147/1000 [09:14<26:34, 1.87s/it, loss=0.636, lr=0.000292] Steps: 15%|█▍ | 147/1000 [09:14<26:34, 1.87s/it, loss=0.563, lr=0.000294] Steps: 15%|█▍ | 148/1000 [09:16<26:32, 1.87s/it, loss=0.563, lr=0.000294] Steps: 15%|█▍ | 148/1000 [09:16<26:32, 1.87s/it, loss=0.534, lr=0.000296] Steps: 15%|█▍ | 149/1000 [09:18<26:31, 1.87s/it, loss=0.534, lr=0.000296] Steps: 15%|█▍ | 149/1000 [09:18<26:31, 1.87s/it, loss=0.71, lr=0.000298] Steps: 15%|█▌ | 150/1000 [09:20<26:30, 1.87s/it, loss=0.71, lr=0.000298] Steps: 15%|█▌ | 150/1000 [09:20<26:30, 1.87s/it, loss=0.825, lr=0.0003] Steps: 15%|█▌ | 151/1000 [09:28<51:04, 3.61s/it, loss=0.825, lr=0.0003] Steps: 15%|█▌ | 151/1000 [09:28<51:04, 3.61s/it, loss=0.336, lr=0.000302] Steps: 15%|█▌ | 152/1000 [09:29<43:37, 3.09s/it, loss=0.336, lr=0.000302] Steps: 15%|█▌ | 152/1000 [09:29<43:37, 3.09s/it, loss=0.331, lr=0.000304] Steps: 15%|█▌ | 153/1000 [09:31<38:24, 2.72s/it, loss=0.331, lr=0.000304] Steps: 15%|█▌ | 153/1000 [09:31<38:24, 2.72s/it, loss=0.313, lr=0.000306] Steps: 15%|█▌ | 154/1000 [09:33<34:45, 2.46s/it, loss=0.313, lr=0.000306] Steps: 15%|█▌ | 154/1000 [09:33<34:45, 2.46s/it, loss=0.345, lr=0.000308] Steps: 16%|█▌ | 155/1000 [09:35<32:11, 2.29s/it, loss=0.345, lr=0.000308] Steps: 16%|█▌ | 155/1000 [09:35<32:11, 2.29s/it, loss=0.606, lr=0.00031] Steps: 16%|█▌ | 156/1000 [09:37<30:24, 2.16s/it, loss=0.606, lr=0.00031] Steps: 16%|█▌ | 156/1000 [09:37<30:24, 2.16s/it, loss=0.288, lr=0.000312] Steps: 16%|█▌ | 157/1000 [09:39<29:08, 2.07s/it, loss=0.288, lr=0.000312] Steps: 16%|█▌ | 157/1000 [09:39<29:08, 2.07s/it, loss=0.866, lr=0.000314] Steps: 16%|█▌ | 158/1000 [09:41<28:14, 2.01s/it, loss=0.866, lr=0.000314] Steps: 16%|█▌ | 158/1000 [09:41<28:14, 2.01s/it, loss=0.418, lr=0.000316] Steps: 16%|█▌ | 159/1000 [09:43<27:36, 1.97s/it, loss=0.418, lr=0.000316] Steps: 16%|█▌ | 159/1000 [09:43<27:36, 1.97s/it, loss=0.55, lr=0.000318] Steps: 16%|█▌ | 160/1000 [09:44<27:09, 1.94s/it, loss=0.55, lr=0.000318] Steps: 16%|█▌ | 160/1000 [09:44<27:09, 1.94s/it, loss=0.516, lr=0.00032] Steps: 16%|█▌ | 161/1000 [09:46<26:50, 1.92s/it, loss=0.516, lr=0.00032] Steps: 16%|█▌ | 161/1000 [09:46<26:50, 1.92s/it, loss=0.978, lr=0.000322] Steps: 16%|█▌ | 162/1000 [09:48<26:35, 1.90s/it, loss=0.978, lr=0.000322] Steps: 16%|█▌ | 162/1000 [09:48<26:35, 1.90s/it, loss=0.323, lr=0.000324] Steps: 16%|█▋ | 163/1000 [09:50<26:25, 1.89s/it, loss=0.323, lr=0.000324] Steps: 16%|█▋ | 163/1000 [09:50<26:25, 1.89s/it, loss=0.346, lr=0.000326] Steps: 16%|█▋ | 164/1000 [09:52<26:16, 1.89s/it, loss=0.346, lr=0.000326] Steps: 16%|█▋ | 164/1000 [09:52<26:16, 1.89s/it, loss=0.55, lr=0.000328] Steps: 16%|█▋ | 165/1000 [09:54<26:11, 1.88s/it, loss=0.55, lr=0.000328] Steps: 16%|█▋ | 165/1000 [09:54<26:11, 1.88s/it, loss=0.918, lr=0.00033] Steps: 17%|█▋ | 166/1000 [09:56<26:06, 1.88s/it, loss=0.918, lr=0.00033] Steps: 17%|█▋ | 166/1000 [09:56<26:06, 1.88s/it, loss=0.73, lr=0.000332] Steps: 17%|█▋ | 167/1000 [09:57<26:02, 1.88s/it, loss=0.73, lr=0.000332] Steps: 17%|█▋ | 167/1000 [09:58<26:02, 1.88s/it, loss=0.521, lr=0.000334] Steps: 17%|█▋ | 168/1000 [09:59<25:59, 1.87s/it, loss=0.521, lr=0.000334] Steps: 17%|█▋ | 168/1000 [09:59<25:59, 1.87s/it, loss=0.319, lr=0.000336] Steps: 17%|█▋ | 169/1000 [10:01<25:56, 1.87s/it, loss=0.319, lr=0.000336] Steps: 17%|█▋ | 169/1000 [10:01<25:56, 1.87s/it, loss=0.307, lr=0.000338] Steps: 17%|█▋ | 170/1000 [10:03<25:51, 1.87s/it, loss=0.307, lr=0.000338] Steps: 17%|█▋ | 170/1000 [10:03<25:51, 1.87s/it, loss=0.336, lr=0.00034] Steps: 17%|█▋ | 171/1000 [10:05<25:48, 1.87s/it, loss=0.336, lr=0.00034] Steps: 17%|█▋ | 171/1000 [10:05<25:48, 1.87s/it, loss=0.472, lr=0.000342] Steps: 17%|█▋ | 172/1000 [10:07<25:48, 1.87s/it, loss=0.472, lr=0.000342] Steps: 17%|█▋ | 172/1000 [10:07<25:48, 1.87s/it, loss=0.364, lr=0.000344] Steps: 17%|█▋ | 173/1000 [10:09<25:46, 1.87s/it, loss=0.364, lr=0.000344] Steps: 17%|█▋ | 173/1000 [10:09<25:46, 1.87s/it, loss=0.311, lr=0.000346] Steps: 17%|█▋ | 174/1000 [10:11<25:44, 1.87s/it, loss=0.311, lr=0.000346] Steps: 17%|█▋ | 174/1000 [10:11<25:44, 1.87s/it, loss=0.228, lr=0.000348] Steps: 18%|█▊ | 175/1000 [10:12<25:41, 1.87s/it, loss=0.228, lr=0.000348] Steps: 18%|█▊ | 175/1000 [10:12<25:41, 1.87s/it, loss=0.406, lr=0.00035] Steps: 18%|█▊ | 176/1000 [10:14<25:38, 1.87s/it, loss=0.406, lr=0.00035] Steps: 18%|█▊ | 176/1000 [10:14<25:38, 1.87s/it, loss=0.322, lr=0.000352] Steps: 18%|█▊ | 177/1000 [10:16<25:36, 1.87s/it, loss=0.322, lr=0.000352] Steps: 18%|█▊ | 177/1000 [10:16<25:36, 1.87s/it, loss=0.417, lr=0.000354] Steps: 18%|█▊ | 178/1000 [10:18<25:35, 1.87s/it, loss=0.417, lr=0.000354] Steps: 18%|█▊ | 178/1000 [10:18<25:35, 1.87s/it, loss=0.71, lr=0.000356] Steps: 18%|█▊ | 179/1000 [10:20<25:34, 1.87s/it, loss=0.71, lr=0.000356] Steps: 18%|█▊ | 179/1000 [10:20<25:34, 1.87s/it, loss=0.443, lr=0.000358] Steps: 18%|█▊ | 180/1000 [10:22<25:33, 1.87s/it, loss=0.443, lr=0.000358] Steps: 18%|█▊ | 180/1000 [10:22<25:33, 1.87s/it, loss=0.893, lr=0.00036] Steps: 18%|█▊ | 181/1000 [10:29<49:23, 3.62s/it, loss=0.893, lr=0.00036] Steps: 18%|█▊ | 181/1000 [10:29<49:23, 3.62s/it, loss=0.798, lr=0.000362] Steps: 18%|█▊ | 182/1000 [10:31<42:10, 3.09s/it, loss=0.798, lr=0.000362] Steps: 18%|█▊ | 182/1000 [10:31<42:10, 3.09s/it, loss=1.03, lr=0.000364] Steps: 18%|█▊ | 183/1000 [10:33<37:06, 2.73s/it, loss=1.03, lr=0.000364] Steps: 18%|█▊ | 183/1000 [10:33<37:06, 2.73s/it, loss=0.711, lr=0.000366] Steps: 18%|█▊ | 184/1000 [10:35<33:33, 2.47s/it, loss=0.711, lr=0.000366] Steps: 18%|█▊ | 184/1000 [10:35<33:33, 2.47s/it, loss=0.311, lr=0.000368] Steps: 18%|█▊ | 185/1000 [10:37<31:05, 2.29s/it, loss=0.311, lr=0.000368] Steps: 18%|█▊ | 185/1000 [10:37<31:05, 2.29s/it, loss=1.05, lr=0.00037] Steps: 19%|█▊ | 186/1000 [10:39<29:20, 2.16s/it, loss=1.05, lr=0.00037] Steps: 19%|█▊ | 186/1000 [10:39<29:20, 2.16s/it, loss=0.781, lr=0.000372] Steps: 19%|█▊ | 187/1000 [10:41<28:07, 2.08s/it, loss=0.781, lr=0.000372] Steps: 19%|█▊ | 187/1000 [10:41<28:07, 2.08s/it, loss=0.506, lr=0.000374] Steps: 19%|█▉ | 188/1000 [10:43<27:14, 2.01s/it, loss=0.506, lr=0.000374] Steps: 19%|█▉ | 188/1000 [10:43<27:14, 2.01s/it, loss=0.415, lr=0.000376] Steps: 19%|█▉ | 189/1000 [10:44<26:38, 1.97s/it, loss=0.415, lr=0.000376] Steps: 19%|█▉ | 189/1000 [10:44<26:38, 1.97s/it, loss=0.37, lr=0.000378] Steps: 19%|█▉ | 190/1000 [10:46<26:10, 1.94s/it, loss=0.37, lr=0.000378] Steps: 19%|█▉ | 190/1000 [10:46<26:10, 1.94s/it, loss=0.327, lr=0.00038] Steps: 19%|█▉ | 191/1000 [10:48<25:51, 1.92s/it, loss=0.327, lr=0.00038] Steps: 19%|█▉ | 191/1000 [10:48<25:51, 1.92s/it, loss=0.883, lr=0.000382] Steps: 19%|█▉ | 192/1000 [10:50<25:38, 1.90s/it, loss=0.883, lr=0.000382] Steps: 19%|█▉ | 192/1000 [10:50<25:38, 1.90s/it, loss=0.868, lr=0.000384] Steps: 19%|█▉ | 193/1000 [10:52<25:29, 1.89s/it, loss=0.868, lr=0.000384] Steps: 19%|█▉ | 193/1000 [10:52<25:29, 1.89s/it, loss=0.294, lr=0.000386] Steps: 19%|█▉ | 194/1000 [10:54<25:21, 1.89s/it, loss=0.294, lr=0.000386] Steps: 19%|█▉ | 194/1000 [10:54<25:21, 1.89s/it, loss=0.529, lr=0.000388] Steps: 20%|█▉ | 195/1000 [10:56<25:15, 1.88s/it, loss=0.529, lr=0.000388] Steps: 20%|█▉ | 195/1000 [10:56<25:15, 1.88s/it, loss=0.343, lr=0.00039] Steps: 20%|█▉ | 196/1000 [10:58<25:10, 1.88s/it, loss=0.343, lr=0.00039] Steps: 20%|█▉ | 196/1000 [10:58<25:10, 1.88s/it, loss=0.996, lr=0.000392] Steps: 20%|█▉ | 197/1000 [10:59<25:06, 1.88s/it, loss=0.996, lr=0.000392] Steps: 20%|█▉ | 197/1000 [10:59<25:06, 1.88s/it, loss=0.36, lr=0.000394] Steps: 20%|█▉ | 198/1000 [11:01<25:02, 1.87s/it, loss=0.36, lr=0.000394] Steps: 20%|█▉ | 198/1000 [11:01<25:02, 1.87s/it, loss=0.869, lr=0.000396] Steps: 20%|█▉ | 199/1000 [11:03<25:00, 1.87s/it, loss=0.869, lr=0.000396] Steps: 20%|█▉ | 199/1000 [11:03<25:00, 1.87s/it, loss=1.02, lr=0.000398] Steps: 20%|██ | 200/1000 [11:05<24:58, 1.87s/it, loss=1.02, lr=0.000398] Steps: 20%|██ | 200/1000 [11:05<24:58, 1.87s/it, loss=0.336, lr=0.0004] Steps: 20%|██ | 201/1000 [11:07<24:54, 1.87s/it, loss=0.336, lr=0.0004] Steps: 20%|██ | 201/1000 [11:07<24:54, 1.87s/it, loss=0.51, lr=0.0004] Steps: 20%|██ | 202/1000 [11:09<24:52, 1.87s/it, loss=0.51, lr=0.0004] Steps: 20%|██ | 202/1000 [11:09<24:52, 1.87s/it, loss=0.543, lr=0.0004] Steps: 20%|██ | 203/1000 [11:11<24:51, 1.87s/it, loss=0.543, lr=0.0004] Steps: 20%|██ | 203/1000 [11:11<24:51, 1.87s/it, loss=1.08, lr=0.0004] Steps: 20%|██ | 204/1000 [11:12<24:48, 1.87s/it, loss=1.08, lr=0.0004] Steps: 20%|██ | 204/1000 [11:12<24:48, 1.87s/it, loss=0.29, lr=0.0004] Steps: 20%|██ | 205/1000 [11:14<24:47, 1.87s/it, loss=0.29, lr=0.0004] Steps: 20%|██ | 205/1000 [11:14<24:47, 1.87s/it, loss=0.432, lr=0.0004] Steps: 21%|██ | 206/1000 [11:16<24:45, 1.87s/it, loss=0.432, lr=0.0004] Steps: 21%|██ | 206/1000 [11:16<24:45, 1.87s/it, loss=0.486, lr=0.0004] Steps: 21%|██ | 207/1000 [11:18<24:44, 1.87s/it, loss=0.486, lr=0.0004] Steps: 21%|██ | 207/1000 [11:18<24:44, 1.87s/it, loss=0.376, lr=0.0004] Steps: 21%|██ | 208/1000 [11:20<24:43, 1.87s/it, loss=0.376, lr=0.0004] Steps: 21%|██ | 208/1000 [11:20<24:43, 1.87s/it, loss=1.03, lr=0.0004] Steps: 21%|██ | 209/1000 [11:22<24:40, 1.87s/it, loss=1.03, lr=0.0004] Steps: 21%|██ | 209/1000 [11:22<24:40, 1.87s/it, loss=0.757, lr=0.0004] Steps: 21%|██ | 210/1000 [11:24<24:37, 1.87s/it, loss=0.757, lr=0.0004] Steps: 21%|██ | 210/1000 [11:24<24:37, 1.87s/it, loss=0.469, lr=0.0004] Steps: 21%|██ | 211/1000 [11:31<47:37, 3.62s/it, loss=0.469, lr=0.0004] Steps: 21%|██ | 211/1000 [11:31<47:37, 3.62s/it, loss=0.361, lr=0.0004] Steps: 21%|██ | 212/1000 [11:33<40:39, 3.10s/it, loss=0.361, lr=0.0004] Steps: 21%|██ | 212/1000 [11:33<40:39, 3.10s/it, loss=0.325, lr=0.0004] Steps: 21%|██▏ | 213/1000 [11:35<35:45, 2.73s/it, loss=0.325, lr=0.0004] Steps: 21%|██▏ | 213/1000 [11:35<35:45, 2.73s/it, loss=0.449, lr=0.0004] Steps: 21%|██▏ | 214/1000 [11:37<32:20, 2.47s/it, loss=0.449, lr=0.0004] Steps: 21%|██▏ | 214/1000 [11:37<32:20, 2.47s/it, loss=0.918, lr=0.0004] Steps: 22%|██▏ | 215/1000 [11:39<29:56, 2.29s/it, loss=0.918, lr=0.0004] Steps: 22%|██▏ | 215/1000 [11:39<29:56, 2.29s/it, loss=0.51, lr=0.0004] Steps: 22%|██▏ | 216/1000 [11:41<28:15, 2.16s/it, loss=0.51, lr=0.0004] Steps: 22%|██▏ | 216/1000 [11:41<28:15, 2.16s/it, loss=0.909, lr=0.0004] Steps: 22%|██▏ | 217/1000 [11:43<27:05, 2.08s/it, loss=0.909, lr=0.0004] Steps: 22%|██▏ | 217/1000 [11:43<27:05, 2.08s/it, loss=0.676, lr=0.0004] Steps: 22%|██▏ | 218/1000 [11:45<26:14, 2.01s/it, loss=0.676, lr=0.0004] Steps: 22%|██▏ | 218/1000 [11:45<26:14, 2.01s/it, loss=0.345, lr=0.0004] Steps: 22%|██▏ | 219/1000 [11:46<25:40, 1.97s/it, loss=0.345, lr=0.0004] Steps: 22%|██▏ | 219/1000 [11:46<25:40, 1.97s/it, loss=0.619, lr=0.000399] Steps: 22%|██▏ | 220/1000 [11:48<25:14, 1.94s/it, loss=0.619, lr=0.000399] Steps: 22%|██▏ | 220/1000 [11:48<25:14, 1.94s/it, loss=0.333, lr=0.000399] Steps: 22%|██▏ | 221/1000 [11:50<24:55, 1.92s/it, loss=0.333, lr=0.000399] Steps: 22%|██▏ | 221/1000 [11:50<24:55, 1.92s/it, loss=0.915, lr=0.000399] Steps: 22%|██▏ | 222/1000 [11:52<24:41, 1.90s/it, loss=0.915, lr=0.000399] Steps: 22%|██▏ | 222/1000 [11:52<24:41, 1.90s/it, loss=0.36, lr=0.000399] Steps: 22%|██▏ | 223/1000 [11:54<24:31, 1.89s/it, loss=0.36, lr=0.000399] Steps: 22%|██▏ | 223/1000 [11:54<24:31, 1.89s/it, loss=0.39, lr=0.000399] Steps: 22%|██▏ | 224/1000 [11:56<24:24, 1.89s/it, loss=0.39, lr=0.000399] Steps: 22%|██▏ | 224/1000 [11:56<24:24, 1.89s/it, loss=1, lr=0.000399] Steps: 22%|██▎ | 225/1000 [11:58<24:19, 1.88s/it, loss=1, lr=0.000399] Steps: 22%|██▎ | 225/1000 [11:58<24:19, 1.88s/it, loss=0.49, lr=0.000399] Steps: 23%|██▎ | 226/1000 [11:59<24:13, 1.88s/it, loss=0.49, lr=0.000399] Steps: 23%|██▎ | 226/1000 [11:59<24:13, 1.88s/it, loss=0.729, lr=0.000399] Steps: 23%|██▎ | 227/1000 [12:01<24:10, 1.88s/it, loss=0.729, lr=0.000399] Steps: 23%|██▎ | 227/1000 [12:01<24:10, 1.88s/it, loss=0.512, lr=0.000399] Steps: 23%|██▎ | 228/1000 [12:03<24:07, 1.87s/it, loss=0.512, lr=0.000399] Steps: 23%|██▎ | 228/1000 [12:03<24:07, 1.87s/it, loss=0.311, lr=0.000399] Steps: 23%|██▎ | 229/1000 [12:05<24:04, 1.87s/it, loss=0.311, lr=0.000399] Steps: 23%|██▎ | 229/1000 [12:05<24:04, 1.87s/it, loss=0.6, lr=0.000399] Steps: 23%|██▎ | 230/1000 [12:07<24:00, 1.87s/it, loss=0.6, lr=0.000399] Steps: 23%|██▎ | 230/1000 [12:07<24:00, 1.87s/it, loss=0.635, lr=0.000399] Steps: 23%|██▎ | 231/1000 [12:09<24:00, 1.87s/it, loss=0.635, lr=0.000399] Steps: 23%|██▎ | 231/1000 [12:09<24:00, 1.87s/it, loss=0.945, lr=0.000399] Steps: 23%|██▎ | 232/1000 [12:11<23:59, 1.87s/it, loss=0.945, lr=0.000399] Steps: 23%|██▎ | 232/1000 [12:11<23:59, 1.87s/it, loss=0.644, lr=0.000398] Steps: 23%|██▎ | 233/1000 [12:13<23:56, 1.87s/it, loss=0.644, lr=0.000398] Steps: 23%|██▎ | 233/1000 [12:13<23:56, 1.87s/it, loss=0.553, lr=0.000398] Steps: 23%|██▎ | 234/1000 [12:14<23:54, 1.87s/it, loss=0.553, lr=0.000398] Steps: 23%|██▎ | 234/1000 [12:14<23:54, 1.87s/it, loss=0.975, lr=0.000398] Steps: 24%|██▎ | 235/1000 [12:16<23:51, 1.87s/it, loss=0.975, lr=0.000398] Steps: 24%|██▎ | 235/1000 [12:16<23:51, 1.87s/it, loss=0.839, lr=0.000398] Steps: 24%|██▎ | 236/1000 [12:18<23:50, 1.87s/it, loss=0.839, lr=0.000398] Steps: 24%|██▎ | 236/1000 [12:18<23:50, 1.87s/it, loss=0.346, lr=0.000398] Steps: 24%|██▎ | 237/1000 [12:20<23:48, 1.87s/it, loss=0.346, lr=0.000398] Steps: 24%|██▎ | 237/1000 [12:20<23:48, 1.87s/it, loss=0.325, lr=0.000398] Steps: 24%|██▍ | 238/1000 [12:22<23:46, 1.87s/it, loss=0.325, lr=0.000398] Steps: 24%|██▍ | 238/1000 [12:22<23:46, 1.87s/it, loss=0.562, lr=0.000398] Steps: 24%|██▍ | 239/1000 [12:24<23:44, 1.87s/it, loss=0.562, lr=0.000398] Steps: 24%|██▍ | 239/1000 [12:24<23:44, 1.87s/it, loss=0.508, lr=0.000398] Steps: 24%|██▍ | 240/1000 [12:26<23:42, 1.87s/it, loss=0.508, lr=0.000398] Steps: 24%|██▍ | 240/1000 [12:26<23:42, 1.87s/it, loss=0.486, lr=0.000398] Steps: 24%|██▍ | 241/1000 [12:33<45:44, 3.62s/it, loss=0.486, lr=0.000398] Steps: 24%|██▍ | 241/1000 [12:33<45:44, 3.62s/it, loss=0.593, lr=0.000397] Steps: 24%|██▍ | 242/1000 [12:35<39:01, 3.09s/it, loss=0.593, lr=0.000397] Steps: 24%|██▍ | 242/1000 [12:35<39:01, 3.09s/it, loss=0.567, lr=0.000397] Steps: 24%|██▍ | 243/1000 [12:37<34:20, 2.72s/it, loss=0.567, lr=0.000397] Steps: 24%|██▍ | 243/1000 [12:37<34:20, 2.72s/it, loss=0.515, lr=0.000397] Steps: 24%|██▍ | 244/1000 [12:39<31:03, 2.46s/it, loss=0.515, lr=0.000397] Steps: 24%|██▍ | 244/1000 [12:39<31:03, 2.46s/it, loss=0.465, lr=0.000397] Steps: 24%|██▍ | 245/1000 [12:41<28:46, 2.29s/it, loss=0.465, lr=0.000397] Steps: 24%|██▍ | 245/1000 [12:41<28:46, 2.29s/it, loss=1.02, lr=0.000397] Steps: 25%|██▍ | 246/1000 [12:43<27:09, 2.16s/it, loss=1.02, lr=0.000397] Steps: 25%|██▍ | 246/1000 [12:43<27:09, 2.16s/it, loss=0.31, lr=0.000397] Steps: 25%|██▍ | 247/1000 [12:45<26:01, 2.07s/it, loss=0.31, lr=0.000397] Steps: 25%|██▍ | 247/1000 [12:45<26:01, 2.07s/it, loss=0.84, lr=0.000397] Steps: 25%|██▍ | 248/1000 [12:46<25:13, 2.01s/it, loss=0.84, lr=0.000397] Steps: 25%|██▍ | 248/1000 [12:46<25:13, 2.01s/it, loss=0.425, lr=0.000396] Steps: 25%|██▍ | 249/1000 [12:48<24:39, 1.97s/it, loss=0.425, lr=0.000396] Steps: 25%|██▍ | 249/1000 [12:48<24:39, 1.97s/it, loss=0.586, lr=0.000396] Steps: 25%|██▌ | 250/1000 [12:50<24:14, 1.94s/it, loss=0.586, lr=0.000396] Steps: 25%|██▌ | 250/1000 [12:50<24:14, 1.94s/it, loss=0.319, lr=0.000396] Steps: 25%|██▌ | 251/1000 [12:52<23:55, 1.92s/it, loss=0.319, lr=0.000396] Steps: 25%|██▌ | 251/1000 [12:52<23:55, 1.92s/it, loss=0.498, lr=0.000396] Steps: 25%|██▌ | 252/1000 [12:54<23:41, 1.90s/it, loss=0.498, lr=0.000396] Steps: 25%|██▌ | 252/1000 [12:54<23:41, 1.90s/it, loss=0.296, lr=0.000396] Steps: 25%|██▌ | 253/1000 [12:56<23:30, 1.89s/it, loss=0.296, lr=0.000396] Steps: 25%|██▌ | 253/1000 [12:56<23:30, 1.89s/it, loss=0.635, lr=0.000396] Steps: 25%|██▌ | 254/1000 [12:58<23:23, 1.88s/it, loss=0.635, lr=0.000396] Steps: 25%|██▌ | 254/1000 [12:58<23:23, 1.88s/it, loss=0.294, lr=0.000396] Steps: 26%|██▌ | 255/1000 [12:59<23:18, 1.88s/it, loss=0.294, lr=0.000396] Steps: 26%|██▌ | 255/1000 [12:59<23:18, 1.88s/it, loss=1.02, lr=0.000395] Steps: 26%|██▌ | 256/1000 [13:01<23:13, 1.87s/it, loss=1.02, lr=0.000395] Steps: 26%|██▌ | 256/1000 [13:01<23:13, 1.87s/it, loss=0.376, lr=0.000395] Steps: 26%|██▌ | 257/1000 [13:03<23:09, 1.87s/it, loss=0.376, lr=0.000395] Steps: 26%|██▌ | 257/1000 [13:03<23:09, 1.87s/it, loss=0.251, lr=0.000395] Steps: 26%|██▌ | 258/1000 [13:05<23:06, 1.87s/it, loss=0.251, lr=0.000395] Steps: 26%|██▌ | 258/1000 [13:05<23:06, 1.87s/it, loss=0.311, lr=0.000395] Steps: 26%|██▌ | 259/1000 [13:07<23:02, 1.87s/it, loss=0.311, lr=0.000395] Steps: 26%|██▌ | 259/1000 [13:07<23:02, 1.87s/it, loss=0.36, lr=0.000395] Steps: 26%|██▌ | 260/1000 [13:09<22:59, 1.86s/it, loss=0.36, lr=0.000395] Steps: 26%|██▌ | 260/1000 [13:09<22:59, 1.86s/it, loss=0.892, lr=0.000394] Steps: 26%|██▌ | 261/1000 [13:11<22:56, 1.86s/it, loss=0.892, lr=0.000394] Steps: 26%|██▌ | 261/1000 [13:11<22:56, 1.86s/it, loss=1.02, lr=0.000394] Steps: 26%|██▌ | 262/1000 [13:13<22:56, 1.87s/it, loss=1.02, lr=0.000394] Steps: 26%|██▌ | 262/1000 [13:13<22:56, 1.87s/it, loss=0.481, lr=0.000394] Steps: 26%|██▋ | 263/1000 [13:14<22:53, 1.86s/it, loss=0.481, lr=0.000394] Steps: 26%|██▋ | 263/1000 [13:14<22:53, 1.86s/it, loss=1.03, lr=0.000394] Steps: 26%|██▋ | 264/1000 [13:16<22:50, 1.86s/it, loss=1.03, lr=0.000394] Steps: 26%|██▋ | 264/1000 [13:16<22:50, 1.86s/it, loss=0.393, lr=0.000394] Steps: 26%|██▋ | 265/1000 [13:18<22:49, 1.86s/it, loss=0.393, lr=0.000394] Steps: 26%|██▋ | 265/1000 [13:18<22:49, 1.86s/it, loss=0.546, lr=0.000394] Steps: 27%|██▋ | 266/1000 [13:20<22:47, 1.86s/it, loss=0.546, lr=0.000394] Steps: 27%|██▋ | 266/1000 [13:20<22:47, 1.86s/it, loss=0.786, lr=0.000393] Steps: 27%|██▋ | 267/1000 [13:22<22:45, 1.86s/it, loss=0.786, lr=0.000393] Steps: 27%|██▋ | 267/1000 [13:22<22:45, 1.86s/it, loss=0.431, lr=0.000393] Steps: 27%|██▋ | 268/1000 [13:24<22:43, 1.86s/it, loss=0.431, lr=0.000393] Steps: 27%|██▋ | 268/1000 [13:24<22:43, 1.86s/it, loss=0.815, lr=0.000393] Steps: 27%|██▋ | 269/1000 [13:26<22:41, 1.86s/it, loss=0.815, lr=0.000393] Steps: 27%|██▋ | 269/1000 [13:26<22:41, 1.86s/it, loss=0.551, lr=0.000393] Steps: 27%|██▋ | 270/1000 [13:27<22:40, 1.86s/it, loss=0.551, lr=0.000393] Steps: 27%|██▋ | 270/1000 [13:27<22:40, 1.86s/it, loss=0.948, lr=0.000392] Steps: 27%|██▋ | 271/1000 [13:35<43:52, 3.61s/it, loss=0.948, lr=0.000392] Steps: 27%|██▋ | 271/1000 [13:35<43:52, 3.61s/it, loss=0.387, lr=0.000392] Steps: 27%|██▋ | 272/1000 [13:37<37:28, 3.09s/it, loss=0.387, lr=0.000392] Steps: 27%|██▋ | 272/1000 [13:37<37:28, 3.09s/it, loss=0.634, lr=0.000392] Steps: 27%|██▋ | 273/1000 [13:39<32:58, 2.72s/it, loss=0.634, lr=0.000392] Steps: 27%|██▋ | 273/1000 [13:39<32:58, 2.72s/it, loss=0.463, lr=0.000392] Steps: 27%|██▋ | 274/1000 [13:41<29:50, 2.47s/it, loss=0.463, lr=0.000392] Steps: 27%|██▋ | 274/1000 [13:41<29:50, 2.47s/it, loss=0.27, lr=0.000392] Steps: 28%|██▊ | 275/1000 [13:43<27:38, 2.29s/it, loss=0.27, lr=0.000392] Steps: 28%|██▊ | 275/1000 [13:43<27:38, 2.29s/it, loss=0.49, lr=0.000391] Steps: 28%|██▊ | 276/1000 [13:44<26:05, 2.16s/it, loss=0.49, lr=0.000391] Steps: 28%|██▊ | 276/1000 [13:44<26:05, 2.16s/it, loss=0.532, lr=0.000391] Steps: 28%|██▊ | 277/1000 [13:46<24:59, 2.07s/it, loss=0.532, lr=0.000391] Steps: 28%|██▊ | 277/1000 [13:46<24:59, 2.07s/it, loss=0.567, lr=0.000391] Steps: 28%|██▊ | 278/1000 [13:48<24:13, 2.01s/it, loss=0.567, lr=0.000391] Steps: 28%|██▊ | 278/1000 [13:48<24:13, 2.01s/it, loss=0.58, lr=0.000391] Steps: 28%|██▊ | 279/1000 [13:50<23:40, 1.97s/it, loss=0.58, lr=0.000391] Steps: 28%|██▊ | 279/1000 [13:50<23:40, 1.97s/it, loss=0.46, lr=0.00039] Steps: 28%|██▊ | 280/1000 [13:52<23:17, 1.94s/it, loss=0.46, lr=0.00039] Steps: 28%|██▊ | 280/1000 [13:52<23:17, 1.94s/it, loss=0.31, lr=0.00039] Steps: 28%|██▊ | 281/1000 [13:54<23:00, 1.92s/it, loss=0.31, lr=0.00039] Steps: 28%|██▊ | 281/1000 [13:54<23:00, 1.92s/it, loss=0.328, lr=0.00039] Steps: 28%|██▊ | 282/1000 [13:56<22:47, 1.90s/it, loss=0.328, lr=0.00039] Steps: 28%|██▊ | 282/1000 [13:56<22:47, 1.90s/it, loss=0.712, lr=0.00039] Steps: 28%|██▊ | 283/1000 [13:58<22:37, 1.89s/it, loss=0.712, lr=0.00039] Steps: 28%|██▊ | 283/1000 [13:58<22:37, 1.89s/it, loss=0.335, lr=0.000389] Steps: 28%|██▊ | 284/1000 [13:59<22:31, 1.89s/it, loss=0.335, lr=0.000389] Steps: 28%|██▊ | 284/1000 [13:59<22:31, 1.89s/it, loss=0.621, lr=0.000389] Steps: 28%|██▊ | 285/1000 [14:01<22:25, 1.88s/it, loss=0.621, lr=0.000389] Steps: 28%|██▊ | 285/1000 [14:01<22:25, 1.88s/it, loss=0.368, lr=0.000389] Steps: 29%|██▊ | 286/1000 [14:03<22:21, 1.88s/it, loss=0.368, lr=0.000389] Steps: 29%|██▊ | 286/1000 [14:03<22:21, 1.88s/it, loss=0.709, lr=0.000389] Steps: 29%|██▊ | 287/1000 [14:05<22:17, 1.88s/it, loss=0.709, lr=0.000389] Steps: 29%|██▊ | 287/1000 [14:05<22:17, 1.88s/it, loss=0.947, lr=0.000388] Steps: 29%|██▉ | 288/1000 [14:07<22:13, 1.87s/it, loss=0.947, lr=0.000388] Steps: 29%|██▉ | 288/1000 [14:07<22:13, 1.87s/it, loss=0.336, lr=0.000388] Steps: 29%|██▉ | 289/1000 [14:09<22:10, 1.87s/it, loss=0.336, lr=0.000388] Steps: 29%|██▉ | 289/1000 [14:09<22:10, 1.87s/it, loss=1.03, lr=0.000388] Steps: 29%|██▉ | 290/1000 [14:11<22:08, 1.87s/it, loss=1.03, lr=0.000388] Steps: 29%|██▉ | 290/1000 [14:11<22:08, 1.87s/it, loss=0.524, lr=0.000388] Steps: 29%|██▉ | 291/1000 [14:13<22:06, 1.87s/it, loss=0.524, lr=0.000388] Steps: 29%|██▉ | 291/1000 [14:13<22:06, 1.87s/it, loss=0.304, lr=0.000387] Steps: 29%|██▉ | 292/1000 [14:14<22:05, 1.87s/it, loss=0.304, lr=0.000387] Steps: 29%|██▉ | 292/1000 [14:14<22:05, 1.87s/it, loss=0.303, lr=0.000387] Steps: 29%|██▉ | 293/1000 [14:16<22:03, 1.87s/it, loss=0.303, lr=0.000387] Steps: 29%|██▉ | 293/1000 [14:16<22:03, 1.87s/it, loss=0.492, lr=0.000387] Steps: 29%|██▉ | 294/1000 [14:18<22:00, 1.87s/it, loss=0.492, lr=0.000387] Steps: 29%|██▉ | 294/1000 [14:18<22:00, 1.87s/it, loss=0.545, lr=0.000387] Steps: 30%|██▉ | 295/1000 [14:20<21:58, 1.87s/it, loss=0.545, lr=0.000387] Steps: 30%|██▉ | 295/1000 [14:20<21:58, 1.87s/it, loss=0.984, lr=0.000386] Steps: 30%|██▉ | 296/1000 [14:22<21:56, 1.87s/it, loss=0.984, lr=0.000386] Steps: 30%|██▉ | 296/1000 [14:22<21:56, 1.87s/it, loss=0.821, lr=0.000386] Steps: 30%|██▉ | 297/1000 [14:24<21:55, 1.87s/it, loss=0.821, lr=0.000386] Steps: 30%|██▉ | 297/1000 [14:24<21:55, 1.87s/it, loss=0.346, lr=0.000386] Steps: 30%|██▉ | 298/1000 [14:26<21:55, 1.87s/it, loss=0.346, lr=0.000386] Steps: 30%|██▉ | 298/1000 [14:26<21:55, 1.87s/it, loss=0.297, lr=0.000385] Steps: 30%|██▉ | 299/1000 [14:27<21:51, 1.87s/it, loss=0.297, lr=0.000385] Steps: 30%|██▉ | 299/1000 [14:27<21:51, 1.87s/it, loss=0.665, lr=0.000385] Steps: 30%|███ | 300/1000 [14:29<21:49, 1.87s/it, loss=0.665, lr=0.000385] Steps: 30%|███ | 300/1000 [14:29<21:49, 1.87s/it, loss=0.433, lr=0.000385] Steps: 30%|███ | 301/1000 [14:37<42:03, 3.61s/it, loss=0.433, lr=0.000385] Steps: 30%|███ | 301/1000 [14:37<42:03, 3.61s/it, loss=0.369, lr=0.000384] Steps: 30%|███ | 302/1000 [14:39<35:55, 3.09s/it, loss=0.369, lr=0.000384] Steps: 30%|███ | 302/1000 [14:39<35:55, 3.09s/it, loss=0.543, lr=0.000384] Steps: 30%|███ | 303/1000 [14:41<31:37, 2.72s/it, loss=0.543, lr=0.000384] Steps: 30%|███ | 303/1000 [14:41<31:37, 2.72s/it, loss=0.327, lr=0.000384] Steps: 30%|███ | 304/1000 [14:43<28:35, 2.46s/it, loss=0.327, lr=0.000384] Steps: 30%|███ | 304/1000 [14:43<28:35, 2.46s/it, loss=0.959, lr=0.000384] Steps: 30%|███ | 305/1000 [14:44<26:28, 2.29s/it, loss=0.959, lr=0.000384] Steps: 30%|███ | 305/1000 [14:44<26:28, 2.29s/it, loss=0.281, lr=0.000383] Steps: 31%|███ | 306/1000 [14:46<24:59, 2.16s/it, loss=0.281, lr=0.000383] Steps: 31%|███ | 306/1000 [14:46<24:59, 2.16s/it, loss=0.432, lr=0.000383] Steps: 31%|███ | 307/1000 [14:48<23:56, 2.07s/it, loss=0.432, lr=0.000383] Steps: 31%|███ | 307/1000 [14:48<23:56, 2.07s/it, loss=0.563, lr=0.000383] Steps: 31%|███ | 308/1000 [14:50<23:10, 2.01s/it, loss=0.563, lr=0.000383] Steps: 31%|███ | 308/1000 [14:50<23:10, 2.01s/it, loss=0.529, lr=0.000382] Steps: 31%|███ | 309/1000 [14:52<22:40, 1.97s/it, loss=0.529, lr=0.000382] Steps: 31%|███ | 309/1000 [14:52<22:40, 1.97s/it, loss=0.73, lr=0.000382] Steps: 31%|███ | 310/1000 [14:54<22:17, 1.94s/it, loss=0.73, lr=0.000382] Steps: 31%|███ | 310/1000 [14:54<22:17, 1.94s/it, loss=0.317, lr=0.000382] Steps: 31%|███ | 311/1000 [14:56<22:01, 1.92s/it, loss=0.317, lr=0.000382] Steps: 31%|███ | 311/1000 [14:56<22:01, 1.92s/it, loss=0.406, lr=0.000381] Steps: 31%|███ | 312/1000 [14:58<21:50, 1.90s/it, loss=0.406, lr=0.000381] Steps: 31%|███ | 312/1000 [14:58<21:50, 1.90s/it, loss=0.944, lr=0.000381] Steps: 31%|███▏ | 313/1000 [14:59<21:41, 1.89s/it, loss=0.944, lr=0.000381] Steps: 31%|███▏ | 313/1000 [14:59<21:41, 1.89s/it, loss=1.06, lr=0.000381] Steps: 31%|███▏ | 314/1000 [15:01<21:33, 1.89s/it, loss=1.06, lr=0.000381] Steps: 31%|███▏ | 314/1000 [15:01<21:33, 1.89s/it, loss=0.557, lr=0.00038] Steps: 32%|███▏ | 315/1000 [15:03<21:28, 1.88s/it, loss=0.557, lr=0.00038] Steps: 32%|███▏ | 315/1000 [15:03<21:28, 1.88s/it, loss=0.632, lr=0.00038] Steps: 32%|███▏ | 316/1000 [15:05<21:24, 1.88s/it, loss=0.632, lr=0.00038] Steps: 32%|███▏ | 316/1000 [15:05<21:24, 1.88s/it, loss=0.384, lr=0.00038] Steps: 32%|███▏ | 317/1000 [15:07<21:21, 1.88s/it, loss=0.384, lr=0.00038] Steps: 32%|███▏ | 317/1000 [15:07<21:21, 1.88s/it, loss=0.725, lr=0.000379] Steps: 32%|███▏ | 318/1000 [15:09<21:18, 1.87s/it, loss=0.725, lr=0.000379] Steps: 32%|███▏ | 318/1000 [15:09<21:18, 1.87s/it, loss=1.03, lr=0.000379] Steps: 32%|███▏ | 319/1000 [15:11<21:16, 1.87s/it, loss=1.03, lr=0.000379] Steps: 32%|███▏ | 319/1000 [15:11<21:16, 1.87s/it, loss=0.48, lr=0.000379] Steps: 32%|███▏ | 320/1000 [15:13<21:14, 1.87s/it, loss=0.48, lr=0.000379] Steps: 32%|███▏ | 320/1000 [15:13<21:14, 1.87s/it, loss=0.702, lr=0.000378] Steps: 32%|███▏ | 321/1000 [15:14<21:11, 1.87s/it, loss=0.702, lr=0.000378] Steps: 32%|███▏ | 321/1000 [15:14<21:11, 1.87s/it, loss=0.453, lr=0.000378] Steps: 32%|███▏ | 322/1000 [15:16<21:10, 1.87s/it, loss=0.453, lr=0.000378] Steps: 32%|███▏ | 322/1000 [15:16<21:10, 1.87s/it, loss=0.384, lr=0.000377] Steps: 32%|███▏ | 323/1000 [15:18<21:06, 1.87s/it, loss=0.384, lr=0.000377] Steps: 32%|███▏ | 323/1000 [15:18<21:06, 1.87s/it, loss=0.349, lr=0.000377] Steps: 32%|███▏ | 324/1000 [15:20<21:02, 1.87s/it, loss=0.349, lr=0.000377] Steps: 32%|███▏ | 324/1000 [15:20<21:02, 1.87s/it, loss=0.612, lr=0.000377] Steps: 32%|███▎ | 325/1000 [15:22<21:00, 1.87s/it, loss=0.612, lr=0.000377] Steps: 32%|███▎ | 325/1000 [15:22<21:00, 1.87s/it, loss=0.6, lr=0.000376] Steps: 33%|███▎ | 326/1000 [15:24<20:59, 1.87s/it, loss=0.6, lr=0.000376] Steps: 33%|███▎ | 326/1000 [15:24<20:59, 1.87s/it, loss=0.39, lr=0.000376] Steps: 33%|███▎ | 327/1000 [15:26<20:56, 1.87s/it, loss=0.39, lr=0.000376] Steps: 33%|███▎ | 327/1000 [15:26<20:56, 1.87s/it, loss=0.709, lr=0.000376] Steps: 33%|███▎ | 328/1000 [15:27<20:55, 1.87s/it, loss=0.709, lr=0.000376] Steps: 33%|███▎ | 328/1000 [15:27<20:55, 1.87s/it, loss=0.313, lr=0.000375] Steps: 33%|███▎ | 329/1000 [15:29<20:51, 1.86s/it, loss=0.313, lr=0.000375] Steps: 33%|███▎ | 329/1000 [15:29<20:51, 1.86s/it, loss=0.695, lr=0.000375] Steps: 33%|███▎ | 330/1000 [15:31<20:49, 1.86s/it, loss=0.695, lr=0.000375] Steps: 33%|███▎ | 330/1000 [15:31<20:49, 1.86s/it, loss=0.548, lr=0.000374] Steps: 33%|███▎ | 331/1000 [15:39<40:19, 3.62s/it, loss=0.548, lr=0.000374] Steps: 33%|███▎ | 331/1000 [15:39<40:19, 3.62s/it, loss=0.915, lr=0.000374] Steps: 33%|███▎ | 332/1000 [15:41<34:25, 3.09s/it, loss=0.915, lr=0.000374] Steps: 33%|███▎ | 332/1000 [15:41<34:25, 3.09s/it, loss=0.617, lr=0.000374] Steps: 33%|███▎ | 333/1000 [15:43<30:18, 2.73s/it, loss=0.617, lr=0.000374] Steps: 33%|███▎ | 333/1000 [15:43<30:18, 2.73s/it, loss=0.328, lr=0.000373] Steps: 33%|███▎ | 334/1000 [15:45<27:25, 2.47s/it, loss=0.328, lr=0.000373] Steps: 33%|███▎ | 334/1000 [15:45<27:25, 2.47s/it, loss=0.745, lr=0.000373] Steps: 34%|███▎ | 335/1000 [15:46<25:22, 2.29s/it, loss=0.745, lr=0.000373] Steps: 34%|███▎ | 335/1000 [15:46<25:22, 2.29s/it, loss=0.752, lr=0.000373] Steps: 34%|███▎ | 336/1000 [15:48<23:56, 2.16s/it, loss=0.752, lr=0.000373] Steps: 34%|███▎ | 336/1000 [15:48<23:56, 2.16s/it, loss=0.307, lr=0.000372] Steps: 34%|███▎ | 337/1000 [15:50<22:55, 2.08s/it, loss=0.307, lr=0.000372] Steps: 34%|███▎ | 337/1000 [15:50<22:55, 2.08s/it, loss=0.995, lr=0.000372] Steps: 34%|███▍ | 338/1000 [15:52<22:13, 2.01s/it, loss=0.995, lr=0.000372] Steps: 34%|███▍ | 338/1000 [15:52<22:13, 2.01s/it, loss=0.637, lr=0.000371] Steps: 34%|███▍ | 339/1000 [15:54<21:42, 1.97s/it, loss=0.637, lr=0.000371] Steps: 34%|███▍ | 339/1000 [15:54<21:42, 1.97s/it, loss=1.02, lr=0.000371] Steps: 34%|███▍ | 340/1000 [15:56<21:20, 1.94s/it, loss=1.02, lr=0.000371] Steps: 34%|███▍ | 340/1000 [15:56<21:20, 1.94s/it, loss=0.464, lr=0.000371] Steps: 34%|███▍ | 341/1000 [15:58<21:05, 1.92s/it, loss=0.464, lr=0.000371] Steps: 34%|███▍ | 341/1000 [15:58<21:05, 1.92s/it, loss=0.321, lr=0.00037] Steps: 34%|███▍ | 342/1000 [15:59<20:53, 1.91s/it, loss=0.321, lr=0.00037] Steps: 34%|███▍ | 342/1000 [15:59<20:53, 1.91s/it, loss=0.649, lr=0.00037] Steps: 34%|███▍ | 343/1000 [16:01<20:45, 1.90s/it, loss=0.649, lr=0.00037] Steps: 34%|███▍ | 343/1000 [16:01<20:45, 1.90s/it, loss=0.569, lr=0.000369] Steps: 34%|███▍ | 344/1000 [16:03<20:38, 1.89s/it, loss=0.569, lr=0.000369] Steps: 34%|███▍ | 344/1000 [16:03<20:38, 1.89s/it, loss=0.286, lr=0.000369] Steps: 34%|███▍ | 345/1000 [16:05<20:32, 1.88s/it, loss=0.286, lr=0.000369] Steps: 34%|███▍ | 345/1000 [16:05<20:32, 1.88s/it, loss=0.714, lr=0.000368] Steps: 35%|███▍ | 346/1000 [16:07<20:27, 1.88s/it, loss=0.714, lr=0.000368] Steps: 35%|███▍ | 346/1000 [16:07<20:27, 1.88s/it, loss=0.395, lr=0.000368] Steps: 35%|███▍ | 347/1000 [16:09<20:23, 1.87s/it, loss=0.395, lr=0.000368] Steps: 35%|███▍ | 347/1000 [16:09<20:23, 1.87s/it, loss=0.835, lr=0.000368] Steps: 35%|███▍ | 348/1000 [16:11<20:19, 1.87s/it, loss=0.835, lr=0.000368] Steps: 35%|███▍ | 348/1000 [16:11<20:19, 1.87s/it, loss=0.386, lr=0.000367] Steps: 35%|███▍ | 349/1000 [16:13<20:17, 1.87s/it, loss=0.386, lr=0.000367] Steps: 35%|███▍ | 349/1000 [16:13<20:17, 1.87s/it, loss=0.482, lr=0.000367] Steps: 35%|███▌ | 350/1000 [16:14<20:14, 1.87s/it, loss=0.482, lr=0.000367] Steps: 35%|███▌ | 350/1000 [16:14<20:14, 1.87s/it, loss=1.06, lr=0.000366] Steps: 35%|███▌ | 351/1000 [16:16<20:11, 1.87s/it, loss=1.06, lr=0.000366] Steps: 35%|███▌ | 351/1000 [16:16<20:11, 1.87s/it, loss=0.54, lr=0.000366] Steps: 35%|███▌ | 352/1000 [16:18<20:10, 1.87s/it, loss=0.54, lr=0.000366] Steps: 35%|███▌ | 352/1000 [16:18<20:10, 1.87s/it, loss=1.04, lr=0.000365] Steps: 35%|███▌ | 353/1000 [16:20<20:08, 1.87s/it, loss=1.04, lr=0.000365] Steps: 35%|███▌ | 353/1000 [16:20<20:08, 1.87s/it, loss=0.389, lr=0.000365] Steps: 35%|███▌ | 354/1000 [16:22<20:05, 1.87s/it, loss=0.389, lr=0.000365] Steps: 35%|███▌ | 354/1000 [16:22<20:05, 1.87s/it, loss=0.695, lr=0.000365] Steps: 36%|███▌ | 355/1000 [16:24<20:03, 1.87s/it, loss=0.695, lr=0.000365] Steps: 36%|███▌ | 355/1000 [16:24<20:03, 1.87s/it, loss=0.45, lr=0.000364] Steps: 36%|███▌ | 356/1000 [16:26<20:01, 1.87s/it, loss=0.45, lr=0.000364] Steps: 36%|███▌ | 356/1000 [16:26<20:01, 1.87s/it, loss=0.875, lr=0.000364] Steps: 36%|███▌ | 357/1000 [16:27<20:00, 1.87s/it, loss=0.875, lr=0.000364] Steps: 36%|███▌ | 357/1000 [16:27<20:00, 1.87s/it, loss=0.711, lr=0.000363] Steps: 36%|███▌ | 358/1000 [16:29<19:58, 1.87s/it, loss=0.711, lr=0.000363] Steps: 36%|███▌ | 358/1000 [16:29<19:58, 1.87s/it, loss=0.635, lr=0.000363] Steps: 36%|███▌ | 359/1000 [16:31<19:54, 1.86s/it, loss=0.635, lr=0.000363] Steps: 36%|███▌ | 359/1000 [16:31<19:54, 1.86s/it, loss=0.983, lr=0.000362] Steps: 36%|███▌ | 360/1000 [16:33<19:53, 1.86s/it, loss=0.983, lr=0.000362] Steps: 36%|███▌ | 360/1000 [16:33<19:53, 1.86s/it, loss=0.776, lr=0.000362] Steps: 36%|███▌ | 361/1000 [16:41<38:48, 3.64s/it, loss=0.776, lr=0.000362] Steps: 36%|███▌ | 361/1000 [16:41<38:48, 3.64s/it, loss=0.335, lr=0.000361] Steps: 36%|███▌ | 362/1000 [16:43<33:05, 3.11s/it, loss=0.335, lr=0.000361] Steps: 36%|███▌ | 362/1000 [16:43<33:05, 3.11s/it, loss=0.319, lr=0.000361] Steps: 36%|███▋ | 363/1000 [16:45<29:05, 2.74s/it, loss=0.319, lr=0.000361] Steps: 36%|███▋ | 363/1000 [16:45<29:05, 2.74s/it, loss=0.497, lr=0.00036] Steps: 36%|███▋ | 364/1000 [16:46<26:16, 2.48s/it, loss=0.497, lr=0.00036] Steps: 36%|███▋ | 364/1000 [16:46<26:16, 2.48s/it, loss=0.38, lr=0.00036] Steps: 36%|███▋ | 365/1000 [16:48<24:18, 2.30s/it, loss=0.38, lr=0.00036] Steps: 36%|███▋ | 365/1000 [16:48<24:18, 2.30s/it, loss=0.281, lr=0.000359] Steps: 37%|███▋ | 366/1000 [16:50<22:54, 2.17s/it, loss=0.281, lr=0.000359] Steps: 37%|███▋ | 366/1000 [16:50<22:54, 2.17s/it, loss=0.668, lr=0.000359] Steps: 37%|███▋ | 367/1000 [16:52<21:56, 2.08s/it, loss=0.668, lr=0.000359] Steps: 37%|███▋ | 367/1000 [16:52<21:56, 2.08s/it, loss=0.576, lr=0.000359] Steps: 37%|███▋ | 368/1000 [16:54<21:13, 2.02s/it, loss=0.576, lr=0.000359] Steps: 37%|███▋ | 368/1000 [16:54<21:13, 2.02s/it, loss=0.352, lr=0.000358] Steps: 37%|███▋ | 369/1000 [16:56<20:45, 1.97s/it, loss=0.352, lr=0.000358] Steps: 37%|███▋ | 369/1000 [16:56<20:45, 1.97s/it, loss=0.295, lr=0.000358] Steps: 37%|███▋ | 370/1000 [16:58<20:23, 1.94s/it, loss=0.295, lr=0.000358] Steps: 37%|███▋ | 370/1000 [16:58<20:23, 1.94s/it, loss=0.324, lr=0.000357] Steps: 37%|███▋ | 371/1000 [17:00<20:07, 1.92s/it, loss=0.324, lr=0.000357] Steps: 37%|███▋ | 371/1000 [17:00<20:07, 1.92s/it, loss=0.819, lr=0.000357] Steps: 37%|███▋ | 372/1000 [17:01<19:56, 1.90s/it, loss=0.819, lr=0.000357] Steps: 37%|███▋ | 372/1000 [17:01<19:56, 1.90s/it, loss=0.616, lr=0.000356] Steps: 37%|███▋ | 373/1000 [17:03<19:47, 1.89s/it, loss=0.616, lr=0.000356] Steps: 37%|███▋ | 373/1000 [17:03<19:47, 1.89s/it, loss=0.496, lr=0.000356] Steps: 37%|███▋ | 374/1000 [17:05<19:42, 1.89s/it, loss=0.496, lr=0.000356] Steps: 37%|███▋ | 374/1000 [17:05<19:42, 1.89s/it, loss=1.04, lr=0.000355] Steps: 38%|███▊ | 375/1000 [17:07<19:35, 1.88s/it, loss=1.04, lr=0.000355] Steps: 38%|███▊ | 375/1000 [17:07<19:35, 1.88s/it, loss=0.9, lr=0.000355] Steps: 38%|███▊ | 376/1000 [17:09<19:31, 1.88s/it, loss=0.9, lr=0.000355] Steps: 38%|███▊ | 376/1000 [17:09<19:31, 1.88s/it, loss=0.34, lr=0.000354] Steps: 38%|███▊ | 377/1000 [17:11<19:26, 1.87s/it, loss=0.34, lr=0.000354] Steps: 38%|███▊ | 377/1000 [17:11<19:26, 1.87s/it, loss=0.779, lr=0.000354] Steps: 38%|███▊ | 378/1000 [17:13<19:23, 1.87s/it, loss=0.779, lr=0.000354] Steps: 38%|███▊ | 378/1000 [17:13<19:23, 1.87s/it, loss=0.889, lr=0.000353] Steps: 38%|███▊ | 379/1000 [17:15<19:21, 1.87s/it, loss=0.889, lr=0.000353] Steps: 38%|███▊ | 379/1000 [17:15<19:21, 1.87s/it, loss=0.66, lr=0.000353] Steps: 38%|███▊ | 380/1000 [17:16<19:18, 1.87s/it, loss=0.66, lr=0.000353] Steps: 38%|███▊ | 380/1000 [17:16<19:18, 1.87s/it, loss=1.02, lr=0.000352] Steps: 38%|███▊ | 381/1000 [17:18<19:16, 1.87s/it, loss=1.02, lr=0.000352] Steps: 38%|███▊ | 381/1000 [17:18<19:16, 1.87s/it, loss=0.313, lr=0.000352] Steps: 38%|███▊ | 382/1000 [17:20<19:14, 1.87s/it, loss=0.313, lr=0.000352] Steps: 38%|███▊ | 382/1000 [17:20<19:14, 1.87s/it, loss=0.447, lr=0.000351] Steps: 38%|███▊ | 383/1000 [17:22<19:12, 1.87s/it, loss=0.447, lr=0.000351] Steps: 38%|███▊ | 383/1000 [17:22<19:12, 1.87s/it, loss=0.36, lr=0.000351] Steps: 38%|███▊ | 384/1000 [17:24<19:10, 1.87s/it, loss=0.36, lr=0.000351] Steps: 38%|███▊ | 384/1000 [17:24<19:10, 1.87s/it, loss=0.428, lr=0.00035] Steps: 38%|███▊ | 385/1000 [17:26<19:08, 1.87s/it, loss=0.428, lr=0.00035] Steps: 38%|███▊ | 385/1000 [17:26<19:08, 1.87s/it, loss=0.344, lr=0.00035] Steps: 39%|███▊ | 386/1000 [17:28<19:06, 1.87s/it, loss=0.344, lr=0.00035] Steps: 39%|███▊ | 386/1000 [17:28<19:06, 1.87s/it, loss=0.449, lr=0.000349] Steps: 39%|███▊ | 387/1000 [17:29<19:03, 1.87s/it, loss=0.449, lr=0.000349] Steps: 39%|███▊ | 387/1000 [17:29<19:03, 1.87s/it, loss=0.58, lr=0.000348] Steps: 39%|███▉ | 388/1000 [17:31<19:02, 1.87s/it, loss=0.58, lr=0.000348] Steps: 39%|███▉ | 388/1000 [17:31<19:02, 1.87s/it, loss=0.29, lr=0.000348] Steps: 39%|███▉ | 389/1000 [17:33<19:00, 1.87s/it, loss=0.29, lr=0.000348] Steps: 39%|███▉ | 389/1000 [17:33<19:00, 1.87s/it, loss=0.411, lr=0.000347] Steps: 39%|███▉ | 390/1000 [17:35<18:58, 1.87s/it, loss=0.411, lr=0.000347] Steps: 39%|███▉ | 390/1000 [17:35<18:58, 1.87s/it, loss=0.536, lr=0.000347] Steps: 39%|███▉ | 391/1000 [17:43<36:41, 3.61s/it, loss=0.536, lr=0.000347] Steps: 39%|███▉ | 391/1000 [17:43<36:41, 3.61s/it, loss=0.541, lr=0.000346] Steps: 39%|███▉ | 392/1000 [17:45<31:18, 3.09s/it, loss=0.541, lr=0.000346] Steps: 39%|███▉ | 392/1000 [17:45<31:18, 3.09s/it, loss=0.529, lr=0.000346] Steps: 39%|███▉ | 393/1000 [17:46<27:33, 2.72s/it, loss=0.529, lr=0.000346] Steps: 39%|███▉ | 393/1000 [17:46<27:33, 2.72s/it, loss=0.554, lr=0.000345] Steps: 39%|███▉ | 394/1000 [17:48<24:56, 2.47s/it, loss=0.554, lr=0.000345] Steps: 39%|███▉ | 394/1000 [17:48<24:56, 2.47s/it, loss=1.02, lr=0.000345] Steps: 40%|███▉ | 395/1000 [17:50<23:04, 2.29s/it, loss=1.02, lr=0.000345] Steps: 40%|███▉ | 395/1000 [17:50<23:04, 2.29s/it, loss=0.525, lr=0.000344] Steps: 40%|███▉ | 396/1000 [17:52<21:46, 2.16s/it, loss=0.525, lr=0.000344] Steps: 40%|███▉ | 396/1000 [17:52<21:46, 2.16s/it, loss=0.454, lr=0.000344] Steps: 40%|███▉ | 397/1000 [17:54<20:52, 2.08s/it, loss=0.454, lr=0.000344] Steps: 40%|███▉ | 397/1000 [17:54<20:52, 2.08s/it, loss=0.691, lr=0.000343] Steps: 40%|███▉ | 398/1000 [17:56<20:13, 2.02s/it, loss=0.691, lr=0.000343] Steps: 40%|███▉ | 398/1000 [17:56<20:13, 2.02s/it, loss=0.404, lr=0.000343] Steps: 40%|███▉ | 399/1000 [17:58<19:45, 1.97s/it, loss=0.404, lr=0.000343] Steps: 40%|███▉ | 399/1000 [17:58<19:45, 1.97s/it, loss=0.413, lr=0.000342] Steps: 40%|████ | 400/1000 [18:00<19:25, 1.94s/it, loss=0.413, lr=0.000342] Steps: 40%|████ | 400/1000 [18:00<19:25, 1.94s/it, loss=0.82, lr=0.000341] Steps: 40%|████ | 401/1000 [18:01<19:11, 1.92s/it, loss=0.82, lr=0.000341] Steps: 40%|████ | 401/1000 [18:01<19:11, 1.92s/it, loss=0.883, lr=0.000341] Steps: 40%|████ | 402/1000 [18:03<19:00, 1.91s/it, loss=0.883, lr=0.000341] Steps: 40%|████ | 402/1000 [18:03<19:00, 1.91s/it, loss=0.634, lr=0.00034] Steps: 40%|████ | 403/1000 [18:05<18:53, 1.90s/it, loss=0.634, lr=0.00034] Steps: 40%|████ | 403/1000 [18:05<18:53, 1.90s/it, loss=1.03, lr=0.00034] Steps: 40%|████ | 404/1000 [18:07<18:46, 1.89s/it, loss=1.03, lr=0.00034] Steps: 40%|████ | 404/1000 [18:07<18:46, 1.89s/it, loss=0.291, lr=0.000339] Steps: 40%|████ | 405/1000 [18:09<18:41, 1.89s/it, loss=0.291, lr=0.000339] Steps: 40%|████ | 405/1000 [18:09<18:41, 1.89s/it, loss=0.596, lr=0.000339] Steps: 41%|████ | 406/1000 [18:11<18:37, 1.88s/it, loss=0.596, lr=0.000339] Steps: 41%|████ | 406/1000 [18:11<18:37, 1.88s/it, loss=1.03, lr=0.000338] Steps: 41%|████ | 407/1000 [18:13<18:33, 1.88s/it, loss=1.03, lr=0.000338] Steps: 41%|████ | 407/1000 [18:13<18:33, 1.88s/it, loss=0.419, lr=0.000337] Steps: 41%|████ | 408/1000 [18:15<18:30, 1.88s/it, loss=0.419, lr=0.000337] Steps: 41%|████ | 408/1000 [18:15<18:30, 1.88s/it, loss=0.664, lr=0.000337] Steps: 41%|████ | 409/1000 [18:16<18:27, 1.87s/it, loss=0.664, lr=0.000337] Steps: 41%|████ | 409/1000 [18:16<18:27, 1.87s/it, loss=0.341, lr=0.000336] Steps: 41%|████ | 410/1000 [18:18<18:25, 1.87s/it, loss=0.341, lr=0.000336] Steps: 41%|████ | 410/1000 [18:18<18:25, 1.87s/it, loss=0.517, lr=0.000336] Steps: 41%|████ | 411/1000 [18:20<18:23, 1.87s/it, loss=0.517, lr=0.000336] Steps: 41%|████ | 411/1000 [18:20<18:23, 1.87s/it, loss=0.818, lr=0.000335] Steps: 41%|████ | 412/1000 [18:22<18:21, 1.87s/it, loss=0.818, lr=0.000335] Steps: 41%|████ | 412/1000 [18:22<18:21, 1.87s/it, loss=0.305, lr=0.000335] Steps: 41%|████▏ | 413/1000 [18:24<18:19, 1.87s/it, loss=0.305, lr=0.000335] Steps: 41%|████▏ | 413/1000 [18:24<18:19, 1.87s/it, loss=0.62, lr=0.000334] Steps: 41%|████▏ | 414/1000 [18:26<18:16, 1.87s/it, loss=0.62, lr=0.000334] Steps: 41%|████▏ | 414/1000 [18:26<18:16, 1.87s/it, loss=0.43, lr=0.000333] Steps: 42%|████▏ | 415/1000 [18:28<18:14, 1.87s/it, loss=0.43, lr=0.000333] Steps: 42%|████▏ | 415/1000 [18:28<18:14, 1.87s/it, loss=0.332, lr=0.000333] Steps: 42%|████▏ | 416/1000 [18:30<18:13, 1.87s/it, loss=0.332, lr=0.000333] Steps: 42%|████▏ | 416/1000 [18:30<18:13, 1.87s/it, loss=0.773, lr=0.000332] Steps: 42%|████▏ | 417/1000 [18:31<18:11, 1.87s/it, loss=0.773, lr=0.000332] Steps: 42%|████▏ | 417/1000 [18:31<18:11, 1.87s/it, loss=0.324, lr=0.000332] Steps: 42%|████▏ | 418/1000 [18:33<18:09, 1.87s/it, loss=0.324, lr=0.000332] Steps: 42%|████▏ | 418/1000 [18:33<18:09, 1.87s/it, loss=0.291, lr=0.000331] Steps: 42%|████▏ | 419/1000 [18:35<18:07, 1.87s/it, loss=0.291, lr=0.000331] Steps: 42%|████▏ | 419/1000 [18:35<18:07, 1.87s/it, loss=0.362, lr=0.00033] Steps: 42%|████▏ | 420/1000 [18:37<18:06, 1.87s/it, loss=0.362, lr=0.00033] Steps: 42%|████▏ | 420/1000 [18:37<18:06, 1.87s/it, loss=0.663, lr=0.00033] Steps: 42%|████▏ | 421/1000 [18:45<35:04, 3.64s/it, loss=0.663, lr=0.00033] Steps: 42%|████▏ | 421/1000 [18:45<35:04, 3.64s/it, loss=0.36, lr=0.000329] Steps: 42%|████▏ | 422/1000 [18:47<29:54, 3.11s/it, loss=0.36, lr=0.000329] Steps: 42%|████▏ | 422/1000 [18:47<29:54, 3.11s/it, loss=0.394, lr=0.000329] Steps: 42%|████▏ | 423/1000 [18:49<26:18, 2.74s/it, loss=0.394, lr=0.000329] Steps: 42%|████▏ | 423/1000 [18:49<26:18, 2.74s/it, loss=0.761, lr=0.000328] Steps: 42%|████▏ | 424/1000 [18:50<23:45, 2.48s/it, loss=0.761, lr=0.000328] Steps: 42%|████▏ | 424/1000 [18:50<23:45, 2.48s/it, loss=0.279, lr=0.000327] Steps: 42%|████▎ | 425/1000 [18:52<21:58, 2.29s/it, loss=0.279, lr=0.000327] Steps: 42%|████▎ | 425/1000 [18:52<21:58, 2.29s/it, loss=0.701, lr=0.000327] Steps: 43%|████▎ | 426/1000 [18:54<20:43, 2.17s/it, loss=0.701, lr=0.000327] Steps: 43%|████▎ | 426/1000 [18:54<20:43, 2.17s/it, loss=0.773, lr=0.000326] Steps: 43%|████▎ | 427/1000 [18:56<19:50, 2.08s/it, loss=0.773, lr=0.000326] Steps: 43%|████▎ | 427/1000 [18:56<19:50, 2.08s/it, loss=0.868, lr=0.000326] Steps: 43%|████▎ | 428/1000 [18:58<19:12, 2.01s/it, loss=0.868, lr=0.000326] Steps: 43%|████▎ | 428/1000 [18:58<19:12, 2.01s/it, loss=0.979, lr=0.000325] Steps: 43%|████▎ | 429/1000 [19:00<18:45, 1.97s/it, loss=0.979, lr=0.000325] Steps: 43%|████▎ | 429/1000 [19:00<18:45, 1.97s/it, loss=0.295, lr=0.000324] Steps: 43%|████▎ | 430/1000 [19:02<18:26, 1.94s/it, loss=0.295, lr=0.000324] Steps: 43%|████▎ | 430/1000 [19:02<18:26, 1.94s/it, loss=0.541, lr=0.000324] Steps: 43%|████▎ | 431/1000 [19:03<18:12, 1.92s/it, loss=0.541, lr=0.000324] Steps: 43%|████▎ | 431/1000 [19:03<18:12, 1.92s/it, loss=0.57, lr=0.000323] Steps: 43%|████▎ | 432/1000 [19:05<18:02, 1.91s/it, loss=0.57, lr=0.000323] Steps: 43%|████▎ | 432/1000 [19:05<18:02, 1.91s/it, loss=0.794, lr=0.000323] Steps: 43%|████▎ | 433/1000 [19:07<17:53, 1.89s/it, loss=0.794, lr=0.000323] Steps: 43%|████▎ | 433/1000 [19:07<17:53, 1.89s/it, loss=0.327, lr=0.000322] Steps: 43%|████▎ | 434/1000 [19:09<17:48, 1.89s/it, loss=0.327, lr=0.000322] Steps: 43%|████▎ | 434/1000 [19:09<17:48, 1.89s/it, loss=0.489, lr=0.000321] Steps: 44%|████▎ | 435/1000 [19:11<17:44, 1.88s/it, loss=0.489, lr=0.000321] Steps: 44%|████▎ | 435/1000 [19:11<17:44, 1.88s/it, loss=0.361, lr=0.000321] Steps: 44%|████▎ | 436/1000 [19:13<17:40, 1.88s/it, loss=0.361, lr=0.000321] Steps: 44%|████▎ | 436/1000 [19:13<17:40, 1.88s/it, loss=0.355, lr=0.00032] Steps: 44%|████▎ | 437/1000 [19:15<17:37, 1.88s/it, loss=0.355, lr=0.00032] Steps: 44%|████▎ | 437/1000 [19:15<17:37, 1.88s/it, loss=0.725, lr=0.000319] Steps: 44%|████▍ | 438/1000 [19:17<17:34, 1.88s/it, loss=0.725, lr=0.000319] Steps: 44%|████▍ | 438/1000 [19:17<17:34, 1.88s/it, loss=0.472, lr=0.000319] Steps: 44%|████▍ | 439/1000 [19:18<17:32, 1.88s/it, loss=0.472, lr=0.000319] Steps: 44%|████▍ | 439/1000 [19:18<17:32, 1.88s/it, loss=0.376, lr=0.000318] Steps: 44%|████▍ | 440/1000 [19:20<17:29, 1.87s/it, loss=0.376, lr=0.000318] Steps: 44%|████▍ | 440/1000 [19:20<17:29, 1.87s/it, loss=0.329, lr=0.000318] Steps: 44%|████▍ | 441/1000 [19:22<17:27, 1.87s/it, loss=0.329, lr=0.000318] Steps: 44%|████▍ | 441/1000 [19:22<17:27, 1.87s/it, loss=0.439, lr=0.000317] Steps: 44%|████▍ | 442/1000 [19:24<17:26, 1.87s/it, loss=0.439, lr=0.000317] Steps: 44%|████▍ | 442/1000 [19:24<17:26, 1.87s/it, loss=0.386, lr=0.000316] Steps: 44%|████▍ | 443/1000 [19:26<17:23, 1.87s/it, loss=0.386, lr=0.000316] Steps: 44%|████▍ | 443/1000 [19:26<17:23, 1.87s/it, loss=0.462, lr=0.000316] Steps: 44%|████▍ | 444/1000 [19:28<17:21, 1.87s/it, loss=0.462, lr=0.000316] Steps: 44%|████▍ | 444/1000 [19:28<17:21, 1.87s/it, loss=0.255, lr=0.000315] Steps: 44%|████▍ | 445/1000 [19:30<17:19, 1.87s/it, loss=0.255, lr=0.000315] Steps: 44%|████▍ | 445/1000 [19:30<17:19, 1.87s/it, loss=0.503, lr=0.000314] Steps: 45%|████▍ | 446/1000 [19:32<17:17, 1.87s/it, loss=0.503, lr=0.000314] Steps: 45%|████▍ | 446/1000 [19:32<17:17, 1.87s/it, loss=0.824, lr=0.000314] Steps: 45%|████▍ | 447/1000 [19:33<17:15, 1.87s/it, loss=0.824, lr=0.000314] Steps: 45%|████▍ | 447/1000 [19:33<17:15, 1.87s/it, loss=0.623, lr=0.000313] Steps: 45%|████▍ | 448/1000 [19:35<17:13, 1.87s/it, loss=0.623, lr=0.000313] Steps: 45%|████▍ | 448/1000 [19:35<17:13, 1.87s/it, loss=0.3, lr=0.000312] Steps: 45%|████▍ | 449/1000 [19:37<17:12, 1.87s/it, loss=0.3, lr=0.000312] Steps: 45%|████▍ | 449/1000 [19:37<17:12, 1.87s/it, loss=0.368, lr=0.000312] Steps: 45%|████▌ | 450/1000 [19:39<17:09, 1.87s/it, loss=0.368, lr=0.000312] Steps: 45%|████▌ | 450/1000 [19:39<17:09, 1.87s/it, loss=0.449, lr=0.000311] Steps: 45%|████▌ | 451/1000 [19:47<33:20, 3.64s/it, loss=0.449, lr=0.000311] Steps: 45%|████▌ | 451/1000 [19:47<33:20, 3.64s/it, loss=0.314, lr=0.00031] Steps: 45%|████▌ | 452/1000 [19:49<28:25, 3.11s/it, loss=0.314, lr=0.00031] Steps: 45%|████▌ | 452/1000 [19:49<28:25, 3.11s/it, loss=0.31, lr=0.00031] Steps: 45%|████▌ | 453/1000 [19:51<24:58, 2.74s/it, loss=0.31, lr=0.00031] Steps: 45%|████▌ | 453/1000 [19:51<24:58, 2.74s/it, loss=0.85, lr=0.000309] Steps: 45%|████▌ | 454/1000 [19:52<22:33, 2.48s/it, loss=0.85, lr=0.000309] Steps: 45%|████▌ | 454/1000 [19:52<22:33, 2.48s/it, loss=0.582, lr=0.000308] Steps: 46%|████▌ | 455/1000 [19:54<20:51, 2.30s/it, loss=0.582, lr=0.000308] Steps: 46%|████▌ | 455/1000 [19:54<20:51, 2.30s/it, loss=0.394, lr=0.000308] Steps: 46%|████▌ | 456/1000 [19:56<19:39, 2.17s/it, loss=0.394, lr=0.000308] Steps: 46%|████▌ | 456/1000 [19:56<19:39, 2.17s/it, loss=0.563, lr=0.000307] Steps: 46%|████▌ | 457/1000 [19:58<18:48, 2.08s/it, loss=0.563, lr=0.000307] Steps: 46%|████▌ | 457/1000 [19:58<18:48, 2.08s/it, loss=0.714, lr=0.000307] Steps: 46%|████▌ | 458/1000 [20:00<18:13, 2.02s/it, loss=0.714, lr=0.000307] Steps: 46%|████▌ | 458/1000 [20:00<18:13, 2.02s/it, loss=0.468, lr=0.000306] Steps: 46%|████▌ | 459/1000 [20:02<17:47, 1.97s/it, loss=0.468, lr=0.000306] Steps: 46%|████▌ | 459/1000 [20:02<17:47, 1.97s/it, loss=0.883, lr=0.000305] Steps: 46%|████▌ | 460/1000 [20:04<17:28, 1.94s/it, loss=0.883, lr=0.000305] Steps: 46%|████▌ | 460/1000 [20:04<17:28, 1.94s/it, loss=0.721, lr=0.000304] Steps: 46%|████▌ | 461/1000 [20:06<17:14, 1.92s/it, loss=0.721, lr=0.000304] Steps: 46%|████▌ | 461/1000 [20:06<17:14, 1.92s/it, loss=0.321, lr=0.000304] Steps: 46%|████▌ | 462/1000 [20:07<17:05, 1.91s/it, loss=0.321, lr=0.000304] Steps: 46%|████▌ | 462/1000 [20:07<17:05, 1.91s/it, loss=0.527, lr=0.000303] Steps: 46%|████▋ | 463/1000 [20:09<16:57, 1.90s/it, loss=0.527, lr=0.000303] Steps: 46%|████▋ | 463/1000 [20:09<16:57, 1.90s/it, loss=0.29, lr=0.000302] Steps: 46%|████▋ | 464/1000 [20:11<16:52, 1.89s/it, loss=0.29, lr=0.000302] Steps: 46%|████▋ | 464/1000 [20:11<16:52, 1.89s/it, loss=0.279, lr=0.000302] Steps: 46%|████▋ | 465/1000 [20:13<16:48, 1.88s/it, loss=0.279, lr=0.000302] Steps: 46%|████▋ | 465/1000 [20:13<16:48, 1.88s/it, loss=0.475, lr=0.000301] Steps: 47%|████▋ | 466/1000 [20:15<16:44, 1.88s/it, loss=0.475, lr=0.000301] Steps: 47%|████▋ | 466/1000 [20:15<16:44, 1.88s/it, loss=0.343, lr=0.0003] Steps: 47%|████▋ | 467/1000 [20:17<16:41, 1.88s/it, loss=0.343, lr=0.0003] Steps: 47%|████▋ | 467/1000 [20:17<16:41, 1.88s/it, loss=0.299, lr=0.0003] Steps: 47%|████▋ | 468/1000 [20:19<16:38, 1.88s/it, loss=0.299, lr=0.0003] Steps: 47%|████▋ | 468/1000 [20:19<16:38, 1.88s/it, loss=0.336, lr=0.000299] Steps: 47%|████▋ | 469/1000 [20:21<16:35, 1.87s/it, loss=0.336, lr=0.000299] Steps: 47%|████▋ | 469/1000 [20:21<16:35, 1.87s/it, loss=1.01, lr=0.000298] Steps: 47%|████▋ | 470/1000 [20:22<16:33, 1.87s/it, loss=1.01, lr=0.000298] Steps: 47%|████▋ | 470/1000 [20:22<16:33, 1.87s/it, loss=0.577, lr=0.000298] Steps: 47%|████▋ | 471/1000 [20:24<16:30, 1.87s/it, loss=0.577, lr=0.000298] Steps: 47%|████▋ | 471/1000 [20:24<16:30, 1.87s/it, loss=0.366, lr=0.000297] Steps: 47%|████▋ | 472/1000 [20:26<16:28, 1.87s/it, loss=0.366, lr=0.000297] Steps: 47%|████▋ | 472/1000 [20:26<16:28, 1.87s/it, loss=0.912, lr=0.000296] Steps: 47%|████▋ | 473/1000 [20:28<16:26, 1.87s/it, loss=0.912, lr=0.000296] Steps: 47%|████▋ | 473/1000 [20:28<16:26, 1.87s/it, loss=0.422, lr=0.000296] Steps: 47%|████▋ | 474/1000 [20:30<16:24, 1.87s/it, loss=0.422, lr=0.000296] Steps: 47%|████▋ | 474/1000 [20:30<16:24, 1.87s/it, loss=0.437, lr=0.000295] Steps: 48%|████▊ | 475/1000 [20:32<16:21, 1.87s/it, loss=0.437, lr=0.000295] Steps: 48%|████▊ | 475/1000 [20:32<16:21, 1.87s/it, loss=0.517, lr=0.000294] Steps: 48%|████▊ | 476/1000 [20:34<16:20, 1.87s/it, loss=0.517, lr=0.000294] Steps: 48%|████▊ | 476/1000 [20:34<16:20, 1.87s/it, loss=0.304, lr=0.000294] Steps: 48%|████▊ | 477/1000 [20:35<16:18, 1.87s/it, loss=0.304, lr=0.000294] Steps: 48%|████▊ | 477/1000 [20:35<16:18, 1.87s/it, loss=0.668, lr=0.000293] Steps: 48%|████▊ | 478/1000 [20:37<16:17, 1.87s/it, loss=0.668, lr=0.000293] Steps: 48%|████▊ | 478/1000 [20:37<16:17, 1.87s/it, loss=0.745, lr=0.000292] Steps: 48%|████▊ | 479/1000 [20:39<16:15, 1.87s/it, loss=0.745, lr=0.000292] Steps: 48%|████▊ | 479/1000 [20:39<16:15, 1.87s/it, loss=0.335, lr=0.000291] Steps: 48%|████▊ | 480/1000 [20:41<16:13, 1.87s/it, loss=0.335, lr=0.000291] Steps: 48%|████▊ | 480/1000 [20:41<16:13, 1.87s/it, loss=0.358, lr=0.000291] Steps: 48%|████▊ | 481/1000 [20:49<31:21, 3.62s/it, loss=0.358, lr=0.000291] Steps: 48%|████▊ | 481/1000 [20:49<31:21, 3.62s/it, loss=0.715, lr=0.00029] Steps: 48%|████▊ | 482/1000 [20:51<26:44, 3.10s/it, loss=0.715, lr=0.00029] Steps: 48%|████▊ | 482/1000 [20:51<26:44, 3.10s/it, loss=1.03, lr=0.000289] Steps: 48%|████▊ | 483/1000 [20:53<23:31, 2.73s/it, loss=1.03, lr=0.000289] Steps: 48%|████▊ | 483/1000 [20:53<23:31, 2.73s/it, loss=0.355, lr=0.000289] Steps: 48%|████▊ | 484/1000 [20:54<21:16, 2.47s/it, loss=0.355, lr=0.000289] Steps: 48%|████▊ | 484/1000 [20:54<21:16, 2.47s/it, loss=0.276, lr=0.000288] Steps: 48%|████▊ | 485/1000 [20:56<19:40, 2.29s/it, loss=0.276, lr=0.000288] Steps: 48%|████▊ | 485/1000 [20:56<19:40, 2.29s/it, loss=0.664, lr=0.000287] Steps: 49%|████▊ | 486/1000 [20:58<18:33, 2.17s/it, loss=0.664, lr=0.000287] Steps: 49%|████▊ | 486/1000 [20:58<18:33, 2.17s/it, loss=0.294, lr=0.000287] Steps: 49%|████▊ | 487/1000 [21:00<17:46, 2.08s/it, loss=0.294, lr=0.000287] Steps: 49%|████▊ | 487/1000 [21:00<17:46, 2.08s/it, loss=0.327, lr=0.000286] Steps: 49%|████▉ | 488/1000 [21:02<17:13, 2.02s/it, loss=0.327, lr=0.000286] Steps: 49%|████▉ | 488/1000 [21:02<17:13, 2.02s/it, loss=0.493, lr=0.000285] Steps: 49%|████▉ | 489/1000 [21:04<16:48, 1.97s/it, loss=0.493, lr=0.000285] Steps: 49%|████▉ | 489/1000 [21:04<16:48, 1.97s/it, loss=0.294, lr=0.000284] Steps: 49%|████▉ | 490/1000 [21:06<16:31, 1.94s/it, loss=0.294, lr=0.000284] Steps: 49%|████▉ | 490/1000 [21:06<16:31, 1.94s/it, loss=0.385, lr=0.000284] Steps: 49%|████▉ | 491/1000 [21:08<16:18, 1.92s/it, loss=0.385, lr=0.000284] Steps: 49%|████▉ | 491/1000 [21:08<16:18, 1.92s/it, loss=0.769, lr=0.000283] Steps: 49%|████▉ | 492/1000 [21:09<16:08, 1.91s/it, loss=0.769, lr=0.000283] Steps: 49%|████▉ | 492/1000 [21:09<16:08, 1.91s/it, loss=0.481, lr=0.000282] Steps: 49%|████▉ | 493/1000 [21:11<16:00, 1.89s/it, loss=0.481, lr=0.000282] Steps: 49%|████▉ | 493/1000 [21:11<16:00, 1.89s/it, loss=0.504, lr=0.000282] Steps: 49%|████▉ | 494/1000 [21:13<15:55, 1.89s/it, loss=0.504, lr=0.000282] Steps: 49%|████▉ | 494/1000 [21:13<15:55, 1.89s/it, loss=0.78, lr=0.000281] Steps: 50%|████▉ | 495/1000 [21:15<15:51, 1.88s/it, loss=0.78, lr=0.000281] Steps: 50%|████▉ | 495/1000 [21:15<15:51, 1.88s/it, loss=0.375, lr=0.00028] Steps: 50%|████▉ | 496/1000 [21:17<15:47, 1.88s/it, loss=0.375, lr=0.00028] Steps: 50%|████▉ | 496/1000 [21:17<15:47, 1.88s/it, loss=0.553, lr=0.000279] Steps: 50%|████▉ | 497/1000 [21:19<15:45, 1.88s/it, loss=0.553, lr=0.000279] Steps: 50%|████▉ | 497/1000 [21:19<15:45, 1.88s/it, loss=0.602, lr=0.000279] Steps: 50%|████▉ | 498/1000 [21:21<15:41, 1.88s/it, loss=0.602, lr=0.000279] Steps: 50%|████▉ | 498/1000 [21:21<15:41, 1.88s/it, loss=0.305, lr=0.000278] Steps: 50%|████▉ | 499/1000 [21:23<15:38, 1.87s/it, loss=0.305, lr=0.000278] Steps: 50%|████▉ | 499/1000 [21:23<15:38, 1.87s/it, loss=0.806, lr=0.000277] Steps: 50%|█████ | 500/1000 [21:24<15:37, 1.87s/it, loss=0.806, lr=0.000277] Steps: 50%|█████ | 500/1000 [21:24<15:37, 1.87s/it, loss=0.926, lr=0.000277] Steps: 50%|█████ | 501/1000 [21:26<15:35, 1.87s/it, loss=0.926, lr=0.000277] Steps: 50%|█████ | 501/1000 [21:26<15:35, 1.87s/it, loss=0.813, lr=0.000276] Steps: 50%|█████ | 502/1000 [21:28<15:33, 1.87s/it, loss=0.813, lr=0.000276] Steps: 50%|█████ | 502/1000 [21:28<15:33, 1.87s/it, loss=0.582, lr=0.000275] Steps: 50%|█████ | 503/1000 [21:30<15:30, 1.87s/it, loss=0.582, lr=0.000275] Steps: 50%|█████ | 503/1000 [21:30<15:30, 1.87s/it, loss=0.995, lr=0.000274] Steps: 50%|█████ | 504/1000 [21:32<15:27, 1.87s/it, loss=0.995, lr=0.000274] Steps: 50%|█████ | 504/1000 [21:32<15:27, 1.87s/it, loss=0.305, lr=0.000274] Steps: 50%|█████ | 505/1000 [21:34<15:26, 1.87s/it, loss=0.305, lr=0.000274] Steps: 50%|█████ | 505/1000 [21:34<15:26, 1.87s/it, loss=0.632, lr=0.000273] Steps: 51%|█████ | 506/1000 [21:36<15:23, 1.87s/it, loss=0.632, lr=0.000273] Steps: 51%|█████ | 506/1000 [21:36<15:23, 1.87s/it, loss=0.632, lr=0.000272] Steps: 51%|█████ | 507/1000 [21:37<15:21, 1.87s/it, loss=0.632, lr=0.000272] Steps: 51%|█████ | 507/1000 [21:37<15:21, 1.87s/it, loss=0.711, lr=0.000271] Steps: 51%|█████ | 508/1000 [21:39<15:18, 1.87s/it, loss=0.711, lr=0.000271] Steps: 51%|█████ | 508/1000 [21:39<15:18, 1.87s/it, loss=0.43, lr=0.000271] Steps: 51%|█████ | 509/1000 [21:41<15:17, 1.87s/it, loss=0.43, lr=0.000271] Steps: 51%|█████ | 509/1000 [21:41<15:17, 1.87s/it, loss=0.368, lr=0.00027] Steps: 51%|█████ | 510/1000 [21:43<15:15, 1.87s/it, loss=0.368, lr=0.00027] Steps: 51%|█████ | 510/1000 [21:43<15:15, 1.87s/it, loss=0.375, lr=0.000269] Steps: 51%|█████ | 511/1000 [21:51<29:36, 3.63s/it, loss=0.375, lr=0.000269] Steps: 51%|█████ | 511/1000 [21:51<29:36, 3.63s/it, loss=1.01, lr=0.000268] Steps: 51%|█████ | 512/1000 [21:53<25:14, 3.10s/it, loss=1.01, lr=0.000268] Steps: 51%|█████ | 512/1000 [21:53<25:14, 3.10s/it, loss=0.322, lr=0.000268] Steps: 51%|█████▏ | 513/1000 [21:55<22:10, 2.73s/it, loss=0.322, lr=0.000268] Steps: 51%|█████▏ | 513/1000 [21:55<22:10, 2.73s/it, loss=0.47, lr=0.000267] Steps: 51%|█████▏ | 514/1000 [21:56<20:03, 2.48s/it, loss=0.47, lr=0.000267] Steps: 51%|█████▏ | 514/1000 [21:56<20:03, 2.48s/it, loss=0.292, lr=0.000266] Steps: 52%|█████▏ | 515/1000 [21:58<18:32, 2.29s/it, loss=0.292, lr=0.000266] Steps: 52%|█████▏ | 515/1000 [21:58<18:32, 2.29s/it, loss=0.704, lr=0.000266] Steps: 52%|█████▏ | 516/1000 [22:00<17:28, 2.17s/it, loss=0.704, lr=0.000266] Steps: 52%|█████▏ | 516/1000 [22:00<17:28, 2.17s/it, loss=0.439, lr=0.000265] Steps: 52%|█████▏ | 517/1000 [22:02<16:44, 2.08s/it, loss=0.439, lr=0.000265] Steps: 52%|█████▏ | 517/1000 [22:02<16:44, 2.08s/it, loss=0.626, lr=0.000264] Steps: 52%|█████▏ | 518/1000 [22:04<16:12, 2.02s/it, loss=0.626, lr=0.000264] Steps: 52%|█████▏ | 518/1000 [22:04<16:12, 2.02s/it, loss=0.579, lr=0.000263] Steps: 52%|█████▏ | 519/1000 [22:06<15:49, 1.97s/it, loss=0.579, lr=0.000263] Steps: 52%|█████▏ | 519/1000 [22:06<15:49, 1.97s/it, loss=0.284, lr=0.000263] Steps: 52%|█████▏ | 520/1000 [22:08<15:31, 1.94s/it, loss=0.284, lr=0.000263] Steps: 52%|█████▏ | 520/1000 [22:08<15:31, 1.94s/it, loss=0.961, lr=0.000262] Steps: 52%|█████▏ | 521/1000 [22:10<15:20, 1.92s/it, loss=0.961, lr=0.000262] Steps: 52%|█████▏ | 521/1000 [22:10<15:20, 1.92s/it, loss=1.02, lr=0.000261] Steps: 52%|█████▏ | 522/1000 [22:11<15:11, 1.91s/it, loss=1.02, lr=0.000261] Steps: 52%|█████▏ | 522/1000 [22:11<15:11, 1.91s/it, loss=0.494, lr=0.00026] Steps: 52%|█████▏ | 523/1000 [22:13<15:04, 1.90s/it, loss=0.494, lr=0.00026] Steps: 52%|█████▏ | 523/1000 [22:13<15:04, 1.90s/it, loss=0.594, lr=0.00026] Steps: 52%|█████▏ | 524/1000 [22:15<14:59, 1.89s/it, loss=0.594, lr=0.00026] Steps: 52%|█████▏ | 524/1000 [22:15<14:59, 1.89s/it, loss=0.322, lr=0.000259] Steps: 52%|█████▎ | 525/1000 [22:17<14:55, 1.88s/it, loss=0.322, lr=0.000259] Steps: 52%|█████▎ | 525/1000 [22:17<14:55, 1.88s/it, loss=0.674, lr=0.000258] Steps: 53%|█████▎ | 526/1000 [22:19<14:51, 1.88s/it, loss=0.674, lr=0.000258] Steps: 53%|█████▎ | 526/1000 [22:19<14:51, 1.88s/it, loss=0.353, lr=0.000257] Steps: 53%|█████▎ | 527/1000 [22:21<14:48, 1.88s/it, loss=0.353, lr=0.000257] Steps: 53%|█████▎ | 527/1000 [22:21<14:48, 1.88s/it, loss=0.218, lr=0.000257] Steps: 53%|█████▎ | 528/1000 [22:23<14:45, 1.88s/it, loss=0.218, lr=0.000257] Steps: 53%|█████▎ | 528/1000 [22:23<14:45, 1.88s/it, loss=0.551, lr=0.000256] Steps: 53%|█████▎ | 529/1000 [22:25<14:42, 1.87s/it, loss=0.551, lr=0.000256] Steps: 53%|█████▎ | 529/1000 [22:25<14:42, 1.87s/it, loss=0.606, lr=0.000255] Steps: 53%|█████▎ | 530/1000 [22:26<14:40, 1.87s/it, loss=0.606, lr=0.000255] Steps: 53%|█████▎ | 530/1000 [22:26<14:40, 1.87s/it, loss=0.932, lr=0.000254] Steps: 53%|█████▎ | 531/1000 [22:28<14:38, 1.87s/it, loss=0.932, lr=0.000254] Steps: 53%|█████▎ | 531/1000 [22:28<14:38, 1.87s/it, loss=0.52, lr=0.000254] Steps: 53%|█████▎ | 532/1000 [22:30<14:36, 1.87s/it, loss=0.52, lr=0.000254] Steps: 53%|█████▎ | 532/1000 [22:30<14:36, 1.87s/it, loss=0.558, lr=0.000253] Steps: 53%|█████▎ | 533/1000 [22:32<14:34, 1.87s/it, loss=0.558, lr=0.000253] Steps: 53%|█████▎ | 533/1000 [22:32<14:34, 1.87s/it, loss=0.606, lr=0.000252] Steps: 53%|█████▎ | 534/1000 [22:34<14:32, 1.87s/it, loss=0.606, lr=0.000252] Steps: 53%|█████▎ | 534/1000 [22:34<14:32, 1.87s/it, loss=0.358, lr=0.000251] Steps: 54%|█████▎ | 535/1000 [22:36<14:31, 1.87s/it, loss=0.358, lr=0.000251] Steps: 54%|█████▎ | 535/1000 [22:36<14:31, 1.87s/it, loss=0.323, lr=0.00025] Steps: 54%|█████▎ | 536/1000 [22:38<14:29, 1.87s/it, loss=0.323, lr=0.00025] Steps: 54%|█████▎ | 536/1000 [22:38<14:29, 1.87s/it, loss=0.517, lr=0.00025] Steps: 54%|█████▎ | 537/1000 [22:39<14:27, 1.87s/it, loss=0.517, lr=0.00025] Steps: 54%|█████▎ | 537/1000 [22:40<14:27, 1.87s/it, loss=0.376, lr=0.000249] Steps: 54%|█████▍ | 538/1000 [22:41<14:25, 1.87s/it, loss=0.376, lr=0.000249] Steps: 54%|█████▍ | 538/1000 [22:41<14:25, 1.87s/it, loss=0.298, lr=0.000248] Steps: 54%|█████▍ | 539/1000 [22:43<14:22, 1.87s/it, loss=0.298, lr=0.000248] Steps: 54%|█████▍ | 539/1000 [22:43<14:22, 1.87s/it, loss=0.557, lr=0.000247] Steps: 54%|█████▍ | 540/1000 [22:45<14:21, 1.87s/it, loss=0.557, lr=0.000247] Steps: 54%|█████▍ | 540/1000 [22:45<14:21, 1.87s/it, loss=0.401, lr=0.000247] Steps: 54%|█████▍ | 541/1000 [22:53<27:46, 3.63s/it, loss=0.401, lr=0.000247] Steps: 54%|█████▍ | 541/1000 [22:53<27:46, 3.63s/it, loss=0.696, lr=0.000246] Steps: 54%|█████▍ | 542/1000 [22:55<23:40, 3.10s/it, loss=0.696, lr=0.000246] Steps: 54%|█████▍ | 542/1000 [22:55<23:40, 3.10s/it, loss=0.533, lr=0.000245] Steps: 54%|█████▍ | 543/1000 [22:57<20:49, 2.73s/it, loss=0.533, lr=0.000245] Steps: 54%|█████▍ | 543/1000 [22:57<20:49, 2.73s/it, loss=0.759, lr=0.000244] Steps: 54%|█████▍ | 544/1000 [22:58<18:47, 2.47s/it, loss=0.759, lr=0.000244] Steps: 54%|█████▍ | 544/1000 [22:58<18:47, 2.47s/it, loss=0.702, lr=0.000244] Steps: 55%|█████▍ | 545/1000 [23:00<17:23, 2.29s/it, loss=0.702, lr=0.000244] Steps: 55%|█████▍ | 545/1000 [23:00<17:23, 2.29s/it, loss=0.356, lr=0.000243] Steps: 55%|█████▍ | 546/1000 [23:02<16:23, 2.17s/it, loss=0.356, lr=0.000243] Steps: 55%|█████▍ | 546/1000 [23:02<16:23, 2.17s/it, loss=0.828, lr=0.000242] Steps: 55%|█████▍ | 547/1000 [23:04<15:40, 2.08s/it, loss=0.828, lr=0.000242] Steps: 55%|█████▍ | 547/1000 [23:04<15:40, 2.08s/it, loss=0.483, lr=0.000241] Steps: 55%|█████▍ | 548/1000 [23:06<15:10, 2.01s/it, loss=0.483, lr=0.000241] Steps: 55%|█████▍ | 548/1000 [23:06<15:10, 2.01s/it, loss=0.418, lr=0.000241] Steps: 55%|█████▍ | 549/1000 [23:08<14:49, 1.97s/it, loss=0.418, lr=0.000241] Steps: 55%|█████▍ | 549/1000 [23:08<14:49, 1.97s/it, loss=0.678, lr=0.00024] Steps: 55%|█████▌ | 550/1000 [23:10<14:34, 1.94s/it, loss=0.678, lr=0.00024] Steps: 55%|█████▌ | 550/1000 [23:10<14:34, 1.94s/it, loss=0.363, lr=0.000239] Steps: 55%|█████▌ | 551/1000 [23:12<14:21, 1.92s/it, loss=0.363, lr=0.000239] Steps: 55%|█████▌ | 551/1000 [23:12<14:21, 1.92s/it, loss=0.89, lr=0.000238] Steps: 55%|█████▌ | 552/1000 [23:13<14:13, 1.91s/it, loss=0.89, lr=0.000238] Steps: 55%|█████▌ | 552/1000 [23:13<14:13, 1.91s/it, loss=0.366, lr=0.000237] Steps: 55%|█████▌ | 553/1000 [23:15<14:07, 1.89s/it, loss=0.366, lr=0.000237] Steps: 55%|█████▌ | 553/1000 [23:15<14:07, 1.89s/it, loss=0.379, lr=0.000237] Steps: 55%|█████▌ | 554/1000 [23:17<14:01, 1.89s/it, loss=0.379, lr=0.000237] Steps: 55%|█████▌ | 554/1000 [23:17<14:01, 1.89s/it, loss=0.333, lr=0.000236] Steps: 56%|█████▌ | 555/1000 [23:19<13:57, 1.88s/it, loss=0.333, lr=0.000236] Steps: 56%|█████▌ | 555/1000 [23:19<13:57, 1.88s/it, loss=0.532, lr=0.000235] Steps: 56%|█████▌ | 556/1000 [23:21<13:54, 1.88s/it, loss=0.532, lr=0.000235] Steps: 56%|█████▌ | 556/1000 [23:21<13:54, 1.88s/it, loss=0.584, lr=0.000234] Steps: 56%|█████▌ | 557/1000 [23:23<13:51, 1.88s/it, loss=0.584, lr=0.000234] Steps: 56%|█████▌ | 557/1000 [23:23<13:51, 1.88s/it, loss=0.409, lr=0.000234] Steps: 56%|█████▌ | 558/1000 [23:25<13:49, 1.88s/it, loss=0.409, lr=0.000234] Steps: 56%|█████▌ | 558/1000 [23:25<13:49, 1.88s/it, loss=0.335, lr=0.000233] Steps: 56%|█████▌ | 559/1000 [23:27<13:46, 1.87s/it, loss=0.335, lr=0.000233] Steps: 56%|█████▌ | 559/1000 [23:27<13:46, 1.87s/it, loss=0.624, lr=0.000232] Steps: 56%|█████▌ | 560/1000 [23:28<13:44, 1.87s/it, loss=0.624, lr=0.000232] Steps: 56%|█████▌ | 560/1000 [23:28<13:44, 1.87s/it, loss=1.03, lr=0.000231] Steps: 56%|█████▌ | 561/1000 [23:30<13:42, 1.87s/it, loss=1.03, lr=0.000231] Steps: 56%|█████▌ | 561/1000 [23:30<13:42, 1.87s/it, loss=0.635, lr=0.000231] Steps: 56%|█████▌ | 562/1000 [23:32<13:40, 1.87s/it, loss=0.635, lr=0.000231] Steps: 56%|█████▌ | 562/1000 [23:32<13:40, 1.87s/it, loss=0.686, lr=0.00023] Steps: 56%|█████▋ | 563/1000 [23:34<13:38, 1.87s/it, loss=0.686, lr=0.00023] Steps: 56%|█████▋ | 563/1000 [23:34<13:38, 1.87s/it, loss=0.336, lr=0.000229] Steps: 56%|█████▋ | 564/1000 [23:36<13:36, 1.87s/it, loss=0.336, lr=0.000229] Steps: 56%|█████▋ | 564/1000 [23:36<13:36, 1.87s/it, loss=0.67, lr=0.000228] Steps: 56%|█████▋ | 565/1000 [23:41<21:23, 2.95s/it, loss=0.67, lr=0.000228] Steps: 56%|█████▋ | 565/1000 [23:41<21:23, 2.95s/it, loss=0.439, lr=0.000227] Steps: 57%|█████▋ | 566/1000 [23:43<18:57, 2.62s/it, loss=0.439, lr=0.000227] Steps: 57%|█████▋ | 566/1000 [23:43<18:57, 2.62s/it, loss=0.308, lr=0.000227] Steps: 57%|█████▋ | 567/1000 [23:45<17:17, 2.40s/it, loss=0.308, lr=0.000227] Steps: 57%|█████▋ | 567/1000 [23:45<17:17, 2.40s/it, loss=0.57, lr=0.000226] Steps: 57%|█████▋ | 568/1000 [23:47<16:07, 2.24s/it, loss=0.57, lr=0.000226] Steps: 57%|█████▋ | 568/1000 [23:47<16:07, 2.24s/it, loss=0.34, lr=0.000225] Steps: 57%|█████▋ | 569/1000 [23:49<15:17, 2.13s/it, loss=0.34, lr=0.000225] Steps: 57%|█████▋ | 569/1000 [23:49<15:17, 2.13s/it, loss=0.604, lr=0.000224] Steps: 57%|█████▋ | 570/1000 [23:51<14:42, 2.05s/it, loss=0.604, lr=0.000224] Steps: 57%|█████▋ | 570/1000 [23:51<14:42, 2.05s/it, loss=0.75, lr=0.000224] Steps: 57%|█████▋ | 571/1000 [23:58<26:56, 3.77s/it, loss=0.75, lr=0.000224] Steps: 57%|█████▋ | 571/1000 [23:58<26:56, 3.77s/it, loss=0.399, lr=0.000223] Steps: 57%|█████▋ | 572/1000 [24:00<22:48, 3.20s/it, loss=0.399, lr=0.000223] Steps: 57%|█████▋ | 572/1000 [24:00<22:48, 3.20s/it, loss=0.568, lr=0.000222] Steps: 57%|█████▋ | 573/1000 [24:02<19:54, 2.80s/it, loss=0.568, lr=0.000222] Steps: 57%|█████▋ | 573/1000 [24:02<19:54, 2.80s/it, loss=0.318, lr=0.000221] Steps: 57%|█████▋ | 574/1000 [24:04<17:53, 2.52s/it, loss=0.318, lr=0.000221] Steps: 57%|█████▋ | 574/1000 [24:04<17:53, 2.52s/it, loss=0.267, lr=0.00022] Steps: 57%|█████▊ | 575/1000 [24:06<16:27, 2.32s/it, loss=0.267, lr=0.00022] Steps: 57%|█████▊ | 575/1000 [24:06<16:27, 2.32s/it, loss=0.557, lr=0.00022] Steps: 58%|█████▊ | 576/1000 [24:08<15:27, 2.19s/it, loss=0.557, lr=0.00022] Steps: 58%|█████▊ | 576/1000 [24:08<15:27, 2.19s/it, loss=0.874, lr=0.000219] Steps: 58%|█████▊ | 577/1000 [24:10<14:45, 2.09s/it, loss=0.874, lr=0.000219] Steps: 58%|█████▊ | 577/1000 [24:10<14:45, 2.09s/it, loss=0.478, lr=0.000218] Steps: 58%|█████▊ | 578/1000 [24:12<14:15, 2.03s/it, loss=0.478, lr=0.000218] Steps: 58%|█████▊ | 578/1000 [24:12<14:15, 2.03s/it, loss=1.02, lr=0.000217] Steps: 58%|█████▊ | 579/1000 [24:13<13:54, 1.98s/it, loss=1.02, lr=0.000217] Steps: 58%|█████▊ | 579/1000 [24:13<13:54, 1.98s/it, loss=0.802, lr=0.000216] Steps: 58%|█████▊ | 580/1000 [24:15<13:38, 1.95s/it, loss=0.802, lr=0.000216] Steps: 58%|█████▊ | 580/1000 [24:15<13:38, 1.95s/it, loss=0.638, lr=0.000216] Steps: 58%|█████▊ | 581/1000 [24:17<13:26, 1.93s/it, loss=0.638, lr=0.000216] Steps: 58%|█████▊ | 581/1000 [24:17<13:26, 1.93s/it, loss=0.338, lr=0.000215] Steps: 58%|█████▊ | 582/1000 [24:19<13:17, 1.91s/it, loss=0.338, lr=0.000215] Steps: 58%|█████▊ | 582/1000 [24:19<13:17, 1.91s/it, loss=0.299, lr=0.000214] Steps: 58%|█████▊ | 583/1000 [24:21<13:11, 1.90s/it, loss=0.299, lr=0.000214] Steps: 58%|█████▊ | 583/1000 [24:21<13:11, 1.90s/it, loss=0.568, lr=0.000213] Steps: 58%|█████▊ | 584/1000 [24:23<13:05, 1.89s/it, loss=0.568, lr=0.000213] Steps: 58%|█████▊ | 584/1000 [24:23<13:05, 1.89s/it, loss=0.278, lr=0.000213] Steps: 58%|█████▊ | 585/1000 [24:25<13:01, 1.88s/it, loss=0.278, lr=0.000213] Steps: 58%|█████▊ | 585/1000 [24:25<13:01, 1.88s/it, loss=0.555, lr=0.000212] Steps: 59%|█████▊ | 586/1000 [24:27<12:59, 1.88s/it, loss=0.555, lr=0.000212] Steps: 59%|█████▊ | 586/1000 [24:27<12:59, 1.88s/it, loss=0.329, lr=0.000211] Steps: 59%|█████▊ | 587/1000 [24:28<12:56, 1.88s/it, loss=0.329, lr=0.000211] Steps: 59%|█████▊ | 587/1000 [24:28<12:56, 1.88s/it, loss=0.315, lr=0.00021] Steps: 59%|█████▉ | 588/1000 [24:30<12:52, 1.88s/it, loss=0.315, lr=0.00021] Steps: 59%|█████▉ | 588/1000 [24:30<12:52, 1.88s/it, loss=0.919, lr=0.000209] Steps: 59%|█████▉ | 589/1000 [24:32<12:49, 1.87s/it, loss=0.919, lr=0.000209] Steps: 59%|█████▉ | 589/1000 [24:32<12:49, 1.87s/it, loss=0.477, lr=0.000209] Steps: 59%|█████▉ | 590/1000 [24:34<12:48, 1.87s/it, loss=0.477, lr=0.000209] Steps: 59%|█████▉ | 590/1000 [24:34<12:48, 1.87s/it, loss=0.706, lr=0.000208] Steps: 59%|█████▉ | 591/1000 [24:36<12:45, 1.87s/it, loss=0.706, lr=0.000208] Steps: 59%|█████▉ | 591/1000 [24:36<12:45, 1.87s/it, loss=0.396, lr=0.000207] Steps: 59%|█████▉ | 592/1000 [24:38<12:44, 1.87s/it, loss=0.396, lr=0.000207] Steps: 59%|█████▉ | 592/1000 [24:38<12:44, 1.87s/it, loss=0.563, lr=0.000206] Steps: 59%|█████▉ | 593/1000 [24:40<12:42, 1.87s/it, loss=0.563, lr=0.000206] Steps: 59%|█████▉ | 593/1000 [24:40<12:42, 1.87s/it, loss=0.316, lr=0.000205] Steps: 59%|█████▉ | 594/1000 [24:41<12:39, 1.87s/it, loss=0.316, lr=0.000205] Steps: 59%|█████▉ | 594/1000 [24:42<12:39, 1.87s/it, loss=0.312, lr=0.000205] Steps: 60%|█████▉ | 595/1000 [24:43<12:37, 1.87s/it, loss=0.312, lr=0.000205] Steps: 60%|█████▉ | 595/1000 [24:43<12:37, 1.87s/it, loss=1.03, lr=0.000204] Steps: 60%|█████▉ | 596/1000 [24:45<12:35, 1.87s/it, loss=1.03, lr=0.000204] Steps: 60%|█████▉ | 596/1000 [24:45<12:35, 1.87s/it, loss=0.493, lr=0.000203] Steps: 60%|█████▉ | 597/1000 [24:47<12:33, 1.87s/it, loss=0.493, lr=0.000203] Steps: 60%|█████▉ | 597/1000 [24:47<12:33, 1.87s/it, loss=0.395, lr=0.000202] Steps: 60%|█████▉ | 598/1000 [24:49<12:31, 1.87s/it, loss=0.395, lr=0.000202] Steps: 60%|█████▉ | 598/1000 [24:49<12:31, 1.87s/it, loss=0.312, lr=0.000202] Steps: 60%|█████▉ | 599/1000 [24:51<12:31, 1.87s/it, loss=0.312, lr=0.000202] Steps: 60%|█████▉ | 599/1000 [24:51<12:31, 1.87s/it, loss=0.5, lr=0.000201] Steps: 60%|██████ | 600/1000 [24:53<12:28, 1.87s/it, loss=0.5, lr=0.000201] Steps: 60%|██████ | 600/1000 [24:53<12:28, 1.87s/it, loss=0.747, lr=0.0002] Steps: 60%|██████ | 601/1000 [25:01<24:21, 3.66s/it, loss=0.747, lr=0.0002] Steps: 60%|██████ | 601/1000 [25:01<24:21, 3.66s/it, loss=0.944, lr=0.000199] Steps: 60%|██████ | 602/1000 [25:02<20:43, 3.12s/it, loss=0.944, lr=0.000199] Steps: 60%|██████ | 602/1000 [25:02<20:43, 3.12s/it, loss=0.316, lr=0.000198] Steps: 60%|██████ | 603/1000 [25:04<18:11, 2.75s/it, loss=0.316, lr=0.000198] Steps: 60%|██████ | 603/1000 [25:04<18:11, 2.75s/it, loss=0.32, lr=0.000198] Steps: 60%|██████ | 604/1000 [25:06<16:24, 2.49s/it, loss=0.32, lr=0.000198] Steps: 60%|██████ | 604/1000 [25:06<16:24, 2.49s/it, loss=0.311, lr=0.000197] Steps: 60%|██████ | 605/1000 [25:08<15:08, 2.30s/it, loss=0.311, lr=0.000197] Steps: 60%|██████ | 605/1000 [25:08<15:08, 2.30s/it, loss=0.359, lr=0.000196] Steps: 61%|██████ | 606/1000 [25:10<14:15, 2.17s/it, loss=0.359, lr=0.000196] Steps: 61%|██████ | 606/1000 [25:10<14:15, 2.17s/it, loss=0.376, lr=0.000195] Steps: 61%|██████ | 607/1000 [25:12<13:37, 2.08s/it, loss=0.376, lr=0.000195] Steps: 61%|██████ | 607/1000 [25:12<13:37, 2.08s/it, loss=0.972, lr=0.000195] Steps: 61%|██████ | 608/1000 [25:14<13:10, 2.02s/it, loss=0.972, lr=0.000195] Steps: 61%|██████ | 608/1000 [25:14<13:10, 2.02s/it, loss=0.953, lr=0.000194] Steps: 61%|██████ | 609/1000 [25:16<12:51, 1.97s/it, loss=0.953, lr=0.000194] Steps: 61%|██████ | 609/1000 [25:16<12:51, 1.97s/it, loss=0.748, lr=0.000193] Steps: 61%|██████ | 610/1000 [25:17<12:37, 1.94s/it, loss=0.748, lr=0.000193] Steps: 61%|██████ | 610/1000 [25:17<12:37, 1.94s/it, loss=0.962, lr=0.000192] Steps: 61%|██████ | 611/1000 [25:19<12:28, 1.92s/it, loss=0.962, lr=0.000192] Steps: 61%|██████ | 611/1000 [25:19<12:28, 1.92s/it, loss=0.633, lr=0.000191] Steps: 61%|██████ | 612/1000 [25:21<12:20, 1.91s/it, loss=0.633, lr=0.000191] Steps: 61%|██████ | 612/1000 [25:21<12:20, 1.91s/it, loss=0.295, lr=0.000191] Steps: 61%|██████▏ | 613/1000 [25:23<12:14, 1.90s/it, loss=0.295, lr=0.000191] Steps: 61%|██████▏ | 613/1000 [25:23<12:14, 1.90s/it, loss=0.616, lr=0.00019] Steps: 61%|██████▏ | 614/1000 [25:25<12:09, 1.89s/it, loss=0.616, lr=0.00019] Steps: 61%|██████▏ | 614/1000 [25:25<12:09, 1.89s/it, loss=0.307, lr=0.000189] Steps: 62%|██████▏ | 615/1000 [25:27<12:05, 1.89s/it, loss=0.307, lr=0.000189] Steps: 62%|██████▏ | 615/1000 [25:27<12:05, 1.89s/it, loss=0.786, lr=0.000188] Steps: 62%|██████▏ | 616/1000 [25:29<12:02, 1.88s/it, loss=0.786, lr=0.000188] Steps: 62%|██████▏ | 616/1000 [25:29<12:02, 1.88s/it, loss=0.587, lr=0.000187] Steps: 62%|██████▏ | 617/1000 [25:31<11:59, 1.88s/it, loss=0.587, lr=0.000187] Steps: 62%|██████▏ | 617/1000 [25:31<11:59, 1.88s/it, loss=0.473, lr=0.000187] Steps: 62%|██████▏ | 618/1000 [25:32<11:56, 1.88s/it, loss=0.473, lr=0.000187] Steps: 62%|██████▏ | 618/1000 [25:32<11:56, 1.88s/it, loss=0.537, lr=0.000186] Steps: 62%|██████▏ | 619/1000 [25:34<11:54, 1.87s/it, loss=0.537, lr=0.000186] Steps: 62%|██████▏ | 619/1000 [25:34<11:54, 1.87s/it, loss=0.884, lr=0.000185] Steps: 62%|██████▏ | 620/1000 [25:36<11:51, 1.87s/it, loss=0.884, lr=0.000185] Steps: 62%|██████▏ | 620/1000 [25:36<11:51, 1.87s/it, loss=0.388, lr=0.000184] Steps: 62%|██████▏ | 621/1000 [25:38<11:49, 1.87s/it, loss=0.388, lr=0.000184] Steps: 62%|██████▏ | 621/1000 [25:38<11:49, 1.87s/it, loss=0.436, lr=0.000184] Steps: 62%|██████▏ | 622/1000 [25:40<11:48, 1.87s/it, loss=0.436, lr=0.000184] Steps: 62%|██████▏ | 622/1000 [25:40<11:48, 1.87s/it, loss=0.293, lr=0.000183] Steps: 62%|██████▏ | 623/1000 [25:42<11:46, 1.87s/it, loss=0.293, lr=0.000183] Steps: 62%|██████▏ | 623/1000 [25:42<11:46, 1.87s/it, loss=0.54, lr=0.000182] Steps: 62%|██████▏ | 624/1000 [25:44<11:44, 1.87s/it, loss=0.54, lr=0.000182] Steps: 62%|██████▏ | 624/1000 [25:44<11:44, 1.87s/it, loss=0.226, lr=0.000181] Steps: 62%|██████▎ | 625/1000 [25:45<11:42, 1.87s/it, loss=0.226, lr=0.000181] Steps: 62%|██████▎ | 625/1000 [25:45<11:42, 1.87s/it, loss=0.816, lr=0.00018] Steps: 63%|██████▎ | 626/1000 [25:47<11:40, 1.87s/it, loss=0.816, lr=0.00018] Steps: 63%|██████▎ | 626/1000 [25:47<11:40, 1.87s/it, loss=0.36, lr=0.00018] Steps: 63%|██████▎ | 627/1000 [25:49<11:38, 1.87s/it, loss=0.36, lr=0.00018] Steps: 63%|██████▎ | 627/1000 [25:49<11:38, 1.87s/it, loss=0.569, lr=0.000179] Steps: 63%|██████▎ | 628/1000 [25:51<11:36, 1.87s/it, loss=0.569, lr=0.000179] Steps: 63%|██████▎ | 628/1000 [25:51<11:36, 1.87s/it, loss=0.617, lr=0.000178] Steps: 63%|██████▎ | 629/1000 [25:53<11:34, 1.87s/it, loss=0.617, lr=0.000178] Steps: 63%|██████▎ | 629/1000 [25:53<11:34, 1.87s/it, loss=0.592, lr=0.000177] Steps: 63%|██████▎ | 630/1000 [25:55<11:32, 1.87s/it, loss=0.592, lr=0.000177] Steps: 63%|██████▎ | 630/1000 [25:55<11:32, 1.87s/it, loss=0.288, lr=0.000176] Steps: 63%|██████▎ | 631/1000 [26:02<22:01, 3.58s/it, loss=0.288, lr=0.000176] Steps: 63%|██████▎ | 631/1000 [26:02<22:01, 3.58s/it, loss=0.517, lr=0.000176] Steps: 63%|██████▎ | 632/1000 [26:04<18:49, 3.07s/it, loss=0.517, lr=0.000176] Steps: 63%|██████▎ | 632/1000 [26:04<18:49, 3.07s/it, loss=0.57, lr=0.000175] Steps: 63%|██████▎ | 633/1000 [26:06<16:33, 2.71s/it, loss=0.57, lr=0.000175] Steps: 63%|██████▎ | 633/1000 [26:06<16:33, 2.71s/it, loss=1.01, lr=0.000174] Steps: 63%|██████▎ | 634/1000 [26:08<14:58, 2.46s/it, loss=1.01, lr=0.000174] Steps: 63%|██████▎ | 634/1000 [26:08<14:58, 2.46s/it, loss=0.282, lr=0.000173] Steps: 64%|██████▎ | 635/1000 [26:10<13:51, 2.28s/it, loss=0.282, lr=0.000173] Steps: 64%|██████▎ | 635/1000 [26:10<13:51, 2.28s/it, loss=0.437, lr=0.000173] Steps: 64%|██████▎ | 636/1000 [26:12<13:05, 2.16s/it, loss=0.437, lr=0.000173] Steps: 64%|██████▎ | 636/1000 [26:12<13:05, 2.16s/it, loss=0.302, lr=0.000172] Steps: 64%|██████▎ | 637/1000 [26:14<12:32, 2.07s/it, loss=0.302, lr=0.000172] Steps: 64%|██████▎ | 637/1000 [26:14<12:32, 2.07s/it, loss=0.353, lr=0.000171] Steps: 64%|██████▍ | 638/1000 [26:16<12:07, 2.01s/it, loss=0.353, lr=0.000171] Steps: 64%|██████▍ | 638/1000 [26:16<12:07, 2.01s/it, loss=0.327, lr=0.00017] Steps: 64%|██████▍ | 639/1000 [26:17<11:50, 1.97s/it, loss=0.327, lr=0.00017] Steps: 64%|██████▍ | 639/1000 [26:17<11:50, 1.97s/it, loss=0.421, lr=0.000169] Steps: 64%|██████▍ | 640/1000 [26:19<11:37, 1.94s/it, loss=0.421, lr=0.000169] Steps: 64%|██████▍ | 640/1000 [26:19<11:37, 1.94s/it, loss=0.428, lr=0.000169] Steps: 64%|██████▍ | 641/1000 [26:21<11:28, 1.92s/it, loss=0.428, lr=0.000169] Steps: 64%|██████▍ | 641/1000 [26:21<11:28, 1.92s/it, loss=0.291, lr=0.000168] Steps: 64%|██████▍ | 642/1000 [26:23<11:21, 1.90s/it, loss=0.291, lr=0.000168] Steps: 64%|██████▍ | 642/1000 [26:23<11:21, 1.90s/it, loss=0.631, lr=0.000167] Steps: 64%|██████▍ | 643/1000 [26:25<11:16, 1.89s/it, loss=0.631, lr=0.000167] Steps: 64%|██████▍ | 643/1000 [26:25<11:16, 1.89s/it, loss=0.315, lr=0.000166] Steps: 64%|██████▍ | 644/1000 [26:27<11:12, 1.89s/it, loss=0.315, lr=0.000166] Steps: 64%|██████▍ | 644/1000 [26:27<11:12, 1.89s/it, loss=0.345, lr=0.000166] Steps: 64%|██████▍ | 645/1000 [26:29<11:08, 1.88s/it, loss=0.345, lr=0.000166] Steps: 64%|██████▍ | 645/1000 [26:29<11:08, 1.88s/it, loss=0.714, lr=0.000165] Steps: 65%|██████▍ | 646/1000 [26:30<11:05, 1.88s/it, loss=0.714, lr=0.000165] Steps: 65%|██████▍ | 646/1000 [26:30<11:05, 1.88s/it, loss=1.03, lr=0.000164] Steps: 65%|██████▍ | 647/1000 [26:32<11:02, 1.88s/it, loss=1.03, lr=0.000164] Steps: 65%|██████▍ | 647/1000 [26:32<11:02, 1.88s/it, loss=0.933, lr=0.000163] Steps: 65%|██████▍ | 648/1000 [26:34<10:59, 1.87s/it, loss=0.933, lr=0.000163] Steps: 65%|██████▍ | 648/1000 [26:34<10:59, 1.87s/it, loss=0.308, lr=0.000163] Steps: 65%|██████▍ | 649/1000 [26:36<10:57, 1.87s/it, loss=0.308, lr=0.000163] Steps: 65%|██████▍ | 649/1000 [26:36<10:57, 1.87s/it, loss=0.408, lr=0.000162] Steps: 65%|██████▌ | 650/1000 [26:38<10:55, 1.87s/it, loss=0.408, lr=0.000162] Steps: 65%|██████▌ | 650/1000 [26:38<10:55, 1.87s/it, loss=0.374, lr=0.000161] Steps: 65%|██████▌ | 651/1000 [26:40<10:53, 1.87s/it, loss=0.374, lr=0.000161] Steps: 65%|██████▌ | 651/1000 [26:40<10:53, 1.87s/it, loss=0.295, lr=0.00016] Steps: 65%|██████▌ | 652/1000 [26:42<10:51, 1.87s/it, loss=0.295, lr=0.00016] Steps: 65%|██████▌ | 652/1000 [26:42<10:51, 1.87s/it, loss=0.458, lr=0.000159] Steps: 65%|██████▌ | 653/1000 [26:44<10:49, 1.87s/it, loss=0.458, lr=0.000159] Steps: 65%|██████▌ | 653/1000 [26:44<10:49, 1.87s/it, loss=0.286, lr=0.000159] Steps: 65%|██████▌ | 654/1000 [26:45<10:47, 1.87s/it, loss=0.286, lr=0.000159] Steps: 65%|██████▌ | 654/1000 [26:45<10:47, 1.87s/it, loss=0.394, lr=0.000158] Steps: 66%|██████▌ | 655/1000 [26:47<10:46, 1.87s/it, loss=0.394, lr=0.000158] Steps: 66%|██████▌ | 655/1000 [26:47<10:46, 1.87s/it, loss=0.894, lr=0.000157] Steps: 66%|██████▌ | 656/1000 [26:49<10:44, 1.87s/it, loss=0.894, lr=0.000157] Steps: 66%|██████▌ | 656/1000 [26:49<10:44, 1.87s/it, loss=0.28, lr=0.000156] Steps: 66%|██████▌ | 657/1000 [26:51<10:41, 1.87s/it, loss=0.28, lr=0.000156] Steps: 66%|██████▌ | 657/1000 [26:51<10:41, 1.87s/it, loss=0.316, lr=0.000156] Steps: 66%|██████▌ | 658/1000 [26:53<10:40, 1.87s/it, loss=0.316, lr=0.000156] Steps: 66%|██████▌ | 658/1000 [26:53<10:40, 1.87s/it, loss=0.992, lr=0.000155] Steps: 66%|██████▌ | 659/1000 [26:55<10:38, 1.87s/it, loss=0.992, lr=0.000155] Steps: 66%|██████▌ | 659/1000 [26:55<10:38, 1.87s/it, loss=0.338, lr=0.000154] Steps: 66%|██████▌ | 660/1000 [26:57<10:36, 1.87s/it, loss=0.338, lr=0.000154] Steps: 66%|██████▌ | 660/1000 [26:57<10:36, 1.87s/it, loss=0.535, lr=0.000153] Steps: 66%|██████▌ | 661/1000 [27:04<20:18, 3.59s/it, loss=0.535, lr=0.000153] Steps: 66%|██████▌ | 661/1000 [27:04<20:18, 3.59s/it, loss=0.435, lr=0.000153] Steps: 66%|██████▌ | 662/1000 [27:06<17:19, 3.08s/it, loss=0.435, lr=0.000153] Steps: 66%|██████▌ | 662/1000 [27:06<17:19, 3.08s/it, loss=0.683, lr=0.000152] Steps: 66%|██████▋ | 663/1000 [27:08<15:14, 2.71s/it, loss=0.683, lr=0.000152] Steps: 66%|██████▋ | 663/1000 [27:08<15:14, 2.71s/it, loss=0.694, lr=0.000151] Steps: 66%|██████▋ | 664/1000 [27:10<13:47, 2.46s/it, loss=0.694, lr=0.000151] Steps: 66%|██████▋ | 664/1000 [27:10<13:47, 2.46s/it, loss=0.385, lr=0.00015] Steps: 66%|██████▋ | 665/1000 [27:12<12:45, 2.28s/it, loss=0.385, lr=0.00015] Steps: 66%|██████▋ | 665/1000 [27:12<12:45, 2.28s/it, loss=0.316, lr=0.00015] Steps: 67%|██████▋ | 666/1000 [27:14<12:01, 2.16s/it, loss=0.316, lr=0.00015] Steps: 67%|██████▋ | 666/1000 [27:14<12:01, 2.16s/it, loss=0.866, lr=0.000149] Steps: 67%|██████▋ | 667/1000 [27:16<11:30, 2.07s/it, loss=0.866, lr=0.000149] Steps: 67%|██████▋ | 667/1000 [27:16<11:30, 2.07s/it, loss=0.656, lr=0.000148] Steps: 67%|██████▋ | 668/1000 [27:17<11:07, 2.01s/it, loss=0.656, lr=0.000148] Steps: 67%|██████▋ | 668/1000 [27:17<11:07, 2.01s/it, loss=0.43, lr=0.000147] Steps: 67%|██████▋ | 669/1000 [27:19<10:52, 1.97s/it, loss=0.43, lr=0.000147] Steps: 67%|██████▋ | 669/1000 [27:19<10:52, 1.97s/it, loss=1.02, lr=0.000146] Steps: 67%|██████▋ | 670/1000 [27:21<10:40, 1.94s/it, loss=1.02, lr=0.000146] Steps: 67%|██████▋ | 670/1000 [27:21<10:40, 1.94s/it, loss=0.334, lr=0.000146] Steps: 67%|██████▋ | 671/1000 [27:23<10:31, 1.92s/it, loss=0.334, lr=0.000146] Steps: 67%|██████▋ | 671/1000 [27:23<10:31, 1.92s/it, loss=0.28, lr=0.000145] Steps: 67%|██████▋ | 672/1000 [27:25<10:24, 1.91s/it, loss=0.28, lr=0.000145] Steps: 67%|██████▋ | 672/1000 [27:25<10:24, 1.91s/it, loss=0.327, lr=0.000144] Steps: 67%|██████▋ | 673/1000 [27:27<10:19, 1.89s/it, loss=0.327, lr=0.000144] Steps: 67%|██████▋ | 673/1000 [27:27<10:19, 1.89s/it, loss=1, lr=0.000143] Steps: 67%|██████▋ | 674/1000 [27:29<10:15, 1.89s/it, loss=1, lr=0.000143] Steps: 67%|██████▋ | 674/1000 [27:29<10:15, 1.89s/it, loss=0.946, lr=0.000143] Steps: 68%|██████▊ | 675/1000 [27:30<10:11, 1.88s/it, loss=0.946, lr=0.000143] Steps: 68%|██████▊ | 675/1000 [27:30<10:11, 1.88s/it, loss=0.582, lr=0.000142] Steps: 68%|██████▊ | 676/1000 [27:32<10:09, 1.88s/it, loss=0.582, lr=0.000142] Steps: 68%|██████▊ | 676/1000 [27:32<10:09, 1.88s/it, loss=0.33, lr=0.000141] Steps: 68%|██████▊ | 677/1000 [27:34<10:06, 1.88s/it, loss=0.33, lr=0.000141] Steps: 68%|██████▊ | 677/1000 [27:34<10:06, 1.88s/it, loss=0.237, lr=0.00014] Steps: 68%|██████▊ | 678/1000 [27:36<10:04, 1.88s/it, loss=0.237, lr=0.00014] Steps: 68%|██████▊ | 678/1000 [27:36<10:04, 1.88s/it, loss=0.393, lr=0.00014] Steps: 68%|██████▊ | 679/1000 [27:38<10:02, 1.88s/it, loss=0.393, lr=0.00014] Steps: 68%|██████▊ | 679/1000 [27:38<10:02, 1.88s/it, loss=0.812, lr=0.000139] Steps: 68%|██████▊ | 680/1000 [27:40<10:00, 1.88s/it, loss=0.812, lr=0.000139] Steps: 68%|██████▊ | 680/1000 [27:40<10:00, 1.88s/it, loss=0.74, lr=0.000138] Steps: 68%|██████▊ | 681/1000 [27:42<09:57, 1.87s/it, loss=0.74, lr=0.000138] Steps: 68%|██████▊ | 681/1000 [27:42<09:57, 1.87s/it, loss=1.04, lr=0.000137] Steps: 68%|██████▊ | 682/1000 [27:44<09:55, 1.87s/it, loss=1.04, lr=0.000137] Steps: 68%|██████▊ | 682/1000 [27:44<09:55, 1.87s/it, loss=0.292, lr=0.000137] Steps: 68%|██████▊ | 683/1000 [27:45<09:53, 1.87s/it, loss=0.292, lr=0.000137] Steps: 68%|██████▊ | 683/1000 [27:45<09:53, 1.87s/it, loss=0.491, lr=0.000136] Steps: 68%|██████▊ | 684/1000 [27:47<09:50, 1.87s/it, loss=0.491, lr=0.000136] Steps: 68%|██████▊ | 684/1000 [27:47<09:50, 1.87s/it, loss=0.56, lr=0.000135] Steps: 68%|██████▊ | 685/1000 [27:49<09:49, 1.87s/it, loss=0.56, lr=0.000135] Steps: 68%|██████▊ | 685/1000 [27:49<09:49, 1.87s/it, loss=0.931, lr=0.000134] Steps: 69%|██████▊ | 686/1000 [27:51<09:47, 1.87s/it, loss=0.931, lr=0.000134] Steps: 69%|██████▊ | 686/1000 [27:51<09:47, 1.87s/it, loss=0.9, lr=0.000134] Steps: 69%|██████▊ | 687/1000 [27:53<09:45, 1.87s/it, loss=0.9, lr=0.000134] Steps: 69%|██████▊ | 687/1000 [27:53<09:45, 1.87s/it, loss=0.472, lr=0.000133] Steps: 69%|██████▉ | 688/1000 [27:55<09:44, 1.87s/it, loss=0.472, lr=0.000133] Steps: 69%|██████▉ | 688/1000 [27:55<09:44, 1.87s/it, loss=0.273, lr=0.000132] Steps: 69%|██████▉ | 689/1000 [27:57<09:42, 1.87s/it, loss=0.273, lr=0.000132] Steps: 69%|██████▉ | 689/1000 [27:57<09:42, 1.87s/it, loss=0.333, lr=0.000132] Steps: 69%|██████▉ | 690/1000 [27:59<09:40, 1.87s/it, loss=0.333, lr=0.000132] Steps: 69%|██████▉ | 690/1000 [27:59<09:40, 1.87s/it, loss=0.755, lr=0.000131] Steps: 69%|██████▉ | 691/1000 [28:06<18:42, 3.63s/it, loss=0.755, lr=0.000131] Steps: 69%|██████▉ | 691/1000 [28:06<18:42, 3.63s/it, loss=0.336, lr=0.00013] Steps: 69%|██████▉ | 692/1000 [28:08<15:55, 3.10s/it, loss=0.336, lr=0.00013] Steps: 69%|██████▉ | 692/1000 [28:08<15:55, 3.10s/it, loss=0.546, lr=0.000129] Steps: 69%|██████▉ | 693/1000 [28:10<13:59, 2.73s/it, loss=0.546, lr=0.000129] Steps: 69%|██████▉ | 693/1000 [28:10<13:59, 2.73s/it, loss=0.302, lr=0.000129] Steps: 69%|██████▉ | 694/1000 [28:12<12:37, 2.48s/it, loss=0.302, lr=0.000129] Steps: 69%|██████▉ | 694/1000 [28:12<12:37, 2.48s/it, loss=0.268, lr=0.000128] Steps: 70%|██████▉ | 695/1000 [28:14<11:39, 2.29s/it, loss=0.268, lr=0.000128] Steps: 70%|██████▉ | 695/1000 [28:14<11:39, 2.29s/it, loss=0.666, lr=0.000127] Steps: 70%|██████▉ | 696/1000 [28:16<10:58, 2.17s/it, loss=0.666, lr=0.000127] Steps: 70%|██████▉ | 696/1000 [28:16<10:58, 2.17s/it, loss=0.302, lr=0.000126] Steps: 70%|██████▉ | 697/1000 [28:18<10:29, 2.08s/it, loss=0.302, lr=0.000126] Steps: 70%|██████▉ | 697/1000 [28:18<10:29, 2.08s/it, loss=0.704, lr=0.000126] Steps: 70%|██████▉ | 698/1000 [28:19<10:08, 2.02s/it, loss=0.704, lr=0.000126] Steps: 70%|██████▉ | 698/1000 [28:19<10:08, 2.02s/it, loss=0.329, lr=0.000125] Steps: 70%|██████▉ | 699/1000 [28:21<09:53, 1.97s/it, loss=0.329, lr=0.000125] Steps: 70%|██████▉ | 699/1000 [28:21<09:53, 1.97s/it, loss=0.309, lr=0.000124] Steps: 70%|███████ | 700/1000 [28:23<09:42, 1.94s/it, loss=0.309, lr=0.000124] Steps: 70%|███████ | 700/1000 [28:23<09:42, 1.94s/it, loss=0.715, lr=0.000123] Steps: 70%|███████ | 701/1000 [28:25<09:34, 1.92s/it, loss=0.715, lr=0.000123] Steps: 70%|███████ | 701/1000 [28:25<09:34, 1.92s/it, loss=0.756, lr=0.000123] Steps: 70%|███████ | 702/1000 [28:27<09:27, 1.90s/it, loss=0.756, lr=0.000123] Steps: 70%|███████ | 702/1000 [28:27<09:27, 1.90s/it, loss=0.805, lr=0.000122] Steps: 70%|███████ | 703/1000 [28:29<09:22, 1.89s/it, loss=0.805, lr=0.000122] Steps: 70%|███████ | 703/1000 [28:29<09:22, 1.89s/it, loss=0.541, lr=0.000121] Steps: 70%|███████ | 704/1000 [28:31<09:18, 1.89s/it, loss=0.541, lr=0.000121] Steps: 70%|███████ | 704/1000 [28:31<09:18, 1.89s/it, loss=0.535, lr=0.000121] Steps: 70%|███████ | 705/1000 [28:32<09:15, 1.88s/it, loss=0.535, lr=0.000121] Steps: 70%|███████ | 705/1000 [28:32<09:15, 1.88s/it, loss=0.319, lr=0.00012] Steps: 71%|███████ | 706/1000 [28:34<09:12, 1.88s/it, loss=0.319, lr=0.00012] Steps: 71%|███████ | 706/1000 [28:34<09:12, 1.88s/it, loss=0.533, lr=0.000119] Steps: 71%|███████ | 707/1000 [28:36<09:09, 1.88s/it, loss=0.533, lr=0.000119] Steps: 71%|███████ | 707/1000 [28:36<09:09, 1.88s/it, loss=0.304, lr=0.000118] Steps: 71%|███████ | 708/1000 [28:38<09:07, 1.88s/it, loss=0.304, lr=0.000118] Steps: 71%|███████ | 708/1000 [28:38<09:07, 1.88s/it, loss=0.296, lr=0.000118] Steps: 71%|███████ | 709/1000 [28:40<09:04, 1.87s/it, loss=0.296, lr=0.000118] Steps: 71%|███████ | 709/1000 [28:40<09:04, 1.87s/it, loss=0.554, lr=0.000117] Steps: 71%|███████ | 710/1000 [28:42<09:02, 1.87s/it, loss=0.554, lr=0.000117] Steps: 71%|███████ | 710/1000 [28:42<09:02, 1.87s/it, loss=0.404, lr=0.000116] Steps: 71%|███████ | 711/1000 [28:44<09:01, 1.87s/it, loss=0.404, lr=0.000116] Steps: 71%|███████ | 711/1000 [28:44<09:01, 1.87s/it, loss=0.494, lr=0.000116] Steps: 71%|███████ | 712/1000 [28:46<08:59, 1.87s/it, loss=0.494, lr=0.000116] Steps: 71%|███████ | 712/1000 [28:46<08:59, 1.87s/it, loss=0.575, lr=0.000115] Steps: 71%|███████▏ | 713/1000 [28:47<08:57, 1.87s/it, loss=0.575, lr=0.000115] Steps: 71%|███████▏ | 713/1000 [28:47<08:57, 1.87s/it, loss=0.809, lr=0.000114] Steps: 71%|███████▏ | 714/1000 [28:49<08:55, 1.87s/it, loss=0.809, lr=0.000114] Steps: 71%|███████▏ | 714/1000 [28:49<08:55, 1.87s/it, loss=0.242, lr=0.000113] Steps: 72%|███████▏ | 715/1000 [28:51<08:53, 1.87s/it, loss=0.242, lr=0.000113] Steps: 72%|███████▏ | 715/1000 [28:51<08:53, 1.87s/it, loss=0.922, lr=0.000113] Steps: 72%|███████▏ | 716/1000 [28:53<08:51, 1.87s/it, loss=0.922, lr=0.000113] Steps: 72%|███████▏ | 716/1000 [28:53<08:51, 1.87s/it, loss=0.338, lr=0.000112] Steps: 72%|███████▏ | 717/1000 [28:55<08:49, 1.87s/it, loss=0.338, lr=0.000112] Steps: 72%|███████▏ | 717/1000 [28:55<08:49, 1.87s/it, loss=0.4, lr=0.000111] Steps: 72%|███████▏ | 718/1000 [28:57<08:47, 1.87s/it, loss=0.4, lr=0.000111] Steps: 72%|███████▏ | 718/1000 [28:57<08:47, 1.87s/it, loss=0.472, lr=0.000111] Steps: 72%|███████▏ | 719/1000 [28:59<08:45, 1.87s/it, loss=0.472, lr=0.000111] Steps: 72%|███████▏ | 719/1000 [28:59<08:45, 1.87s/it, loss=0.346, lr=0.00011] Steps: 72%|███████▏ | 720/1000 [29:01<08:44, 1.87s/it, loss=0.346, lr=0.00011] Steps: 72%|███████▏ | 720/1000 [29:01<08:44, 1.87s/it, loss=0.278, lr=0.000109] Steps: 72%|███████▏ | 721/1000 [29:08<16:45, 3.61s/it, loss=0.278, lr=0.000109] Steps: 72%|███████▏ | 721/1000 [29:08<16:45, 3.61s/it, loss=0.41, lr=0.000109] Steps: 72%|███████▏ | 722/1000 [29:10<14:17, 3.08s/it, loss=0.41, lr=0.000109] Steps: 72%|███████▏ | 722/1000 [29:10<14:17, 3.08s/it, loss=0.684, lr=0.000108] Steps: 72%|███████▏ | 723/1000 [29:12<12:33, 2.72s/it, loss=0.684, lr=0.000108] Steps: 72%|███████▏ | 723/1000 [29:12<12:33, 2.72s/it, loss=0.397, lr=0.000107] Steps: 72%|███████▏ | 724/1000 [29:14<11:20, 2.46s/it, loss=0.397, lr=0.000107] Steps: 72%|███████▏ | 724/1000 [29:14<11:20, 2.46s/it, loss=0.553, lr=0.000106] Steps: 72%|███████▎ | 725/1000 [29:16<10:28, 2.29s/it, loss=0.553, lr=0.000106] Steps: 72%|███████▎ | 725/1000 [29:16<10:28, 2.29s/it, loss=0.656, lr=0.000106] Steps: 73%|███████▎ | 726/1000 [29:18<09:52, 2.16s/it, loss=0.656, lr=0.000106] Steps: 73%|███████▎ | 726/1000 [29:18<09:52, 2.16s/it, loss=0.394, lr=0.000105] Steps: 73%|███████▎ | 727/1000 [29:19<09:25, 2.07s/it, loss=0.394, lr=0.000105] Steps: 73%|███████▎ | 727/1000 [29:19<09:25, 2.07s/it, loss=0.329, lr=0.000104] Steps: 73%|███████▎ | 728/1000 [29:21<09:07, 2.01s/it, loss=0.329, lr=0.000104] Steps: 73%|███████▎ | 728/1000 [29:21<09:07, 2.01s/it, loss=0.849, lr=0.000104] Steps: 73%|███████▎ | 729/1000 [29:23<08:53, 1.97s/it, loss=0.849, lr=0.000104] Steps: 73%|███████▎ | 729/1000 [29:23<08:53, 1.97s/it, loss=0.514, lr=0.000103] Steps: 73%|███████▎ | 730/1000 [29:25<08:43, 1.94s/it, loss=0.514, lr=0.000103] Steps: 73%|███████▎ | 730/1000 [29:25<08:43, 1.94s/it, loss=0.35, lr=0.000102] Steps: 73%|███████▎ | 731/1000 [29:27<08:36, 1.92s/it, loss=0.35, lr=0.000102] Steps: 73%|███████▎ | 731/1000 [29:27<08:36, 1.92s/it, loss=0.565, lr=0.000102] Steps: 73%|███████▎ | 732/1000 [29:29<08:30, 1.91s/it, loss=0.565, lr=0.000102] Steps: 73%|███████▎ | 732/1000 [29:29<08:30, 1.91s/it, loss=0.907, lr=0.000101] Steps: 73%|███████▎ | 733/1000 [29:31<08:25, 1.90s/it, loss=0.907, lr=0.000101] Steps: 73%|███████▎ | 733/1000 [29:31<08:25, 1.90s/it, loss=0.68, lr=0.0001] Steps: 73%|███████▎ | 734/1000 [29:32<08:21, 1.89s/it, loss=0.68, lr=0.0001] Steps: 73%|███████▎ | 734/1000 [29:33<08:21, 1.89s/it, loss=0.374, lr=9.95e-5] Steps: 74%|███████▎ | 735/1000 [29:34<08:18, 1.88s/it, loss=0.374, lr=9.95e-5] Steps: 74%|███████▎ | 735/1000 [29:34<08:18, 1.88s/it, loss=0.32, lr=9.89e-5] Steps: 74%|███████▎ | 736/1000 [29:36<08:15, 1.88s/it, loss=0.32, lr=9.89e-5] Steps: 74%|███████▎ | 736/1000 [29:36<08:15, 1.88s/it, loss=0.331, lr=9.82e-5] Steps: 74%|███████▎ | 737/1000 [29:38<08:12, 1.87s/it, loss=0.331, lr=9.82e-5] Steps: 74%|███████▎ | 737/1000 [29:38<08:12, 1.87s/it, loss=0.579, lr=9.75e-5] Steps: 74%|███████▍ | 738/1000 [29:40<08:10, 1.87s/it, loss=0.579, lr=9.75e-5] Steps: 74%|███████▍ | 738/1000 [29:40<08:10, 1.87s/it, loss=0.369, lr=9.68e-5] Steps: 74%|███████▍ | 739/1000 [29:42<08:07, 1.87s/it, loss=0.369, lr=9.68e-5] Steps: 74%|███████▍ | 739/1000 [29:42<08:07, 1.87s/it, loss=0.469, lr=9.62e-5] Steps: 74%|███████▍ | 740/1000 [29:44<08:05, 1.87s/it, loss=0.469, lr=9.62e-5] Steps: 74%|███████▍ | 740/1000 [29:44<08:05, 1.87s/it, loss=0.932, lr=9.55e-5] Steps: 74%|███████▍ | 741/1000 [29:46<08:03, 1.87s/it, loss=0.932, lr=9.55e-5] Steps: 74%|███████▍ | 741/1000 [29:46<08:03, 1.87s/it, loss=0.518, lr=9.48e-5] Steps: 74%|███████▍ | 742/1000 [29:47<08:01, 1.87s/it, loss=0.518, lr=9.48e-5] Steps: 74%|███████▍ | 742/1000 [29:47<08:01, 1.87s/it, loss=0.301, lr=9.42e-5] Steps: 74%|███████▍ | 743/1000 [29:49<07:59, 1.87s/it, loss=0.301, lr=9.42e-5] Steps: 74%|███████▍ | 743/1000 [29:49<07:59, 1.87s/it, loss=0.681, lr=9.35e-5] Steps: 74%|███████▍ | 744/1000 [29:51<07:57, 1.87s/it, loss=0.681, lr=9.35e-5] Steps: 74%|███████▍ | 744/1000 [29:51<07:57, 1.87s/it, loss=0.229, lr=9.28e-5] Steps: 74%|███████▍ | 745/1000 [29:53<07:56, 1.87s/it, loss=0.229, lr=9.28e-5] Steps: 74%|███████▍ | 745/1000 [29:53<07:56, 1.87s/it, loss=0.42, lr=9.22e-5] Steps: 75%|███████▍ | 746/1000 [29:55<07:54, 1.87s/it, loss=0.42, lr=9.22e-5] Steps: 75%|███████▍ | 746/1000 [29:55<07:54, 1.87s/it, loss=0.654, lr=9.15e-5] Steps: 75%|███████▍ | 747/1000 [29:57<07:52, 1.87s/it, loss=0.654, lr=9.15e-5] Steps: 75%|███████▍ | 747/1000 [29:57<07:52, 1.87s/it, loss=0.484, lr=9.09e-5] Steps: 75%|███████▍ | 748/1000 [29:59<07:50, 1.87s/it, loss=0.484, lr=9.09e-5] Steps: 75%|███████▍ | 748/1000 [29:59<07:50, 1.87s/it, loss=0.28, lr=9.02e-5] Steps: 75%|███████▍ | 749/1000 [30:00<07:48, 1.87s/it, loss=0.28, lr=9.02e-5] Steps: 75%|███████▍ | 749/1000 [30:00<07:48, 1.87s/it, loss=0.429, lr=8.95e-5] Steps: 75%|███████▌ | 750/1000 [30:02<07:46, 1.87s/it, loss=0.429, lr=8.95e-5] Steps: 75%|███████▌ | 750/1000 [30:02<07:46, 1.87s/it, loss=0.43, lr=8.89e-5] Steps: 75%|███████▌ | 751/1000 [30:10<14:55, 3.59s/it, loss=0.43, lr=8.89e-5] Steps: 75%|███████▌ | 751/1000 [30:10<14:55, 3.59s/it, loss=0.583, lr=8.82e-5] Steps: 75%|███████▌ | 752/1000 [30:12<12:43, 3.08s/it, loss=0.583, lr=8.82e-5] Steps: 75%|███████▌ | 752/1000 [30:12<12:43, 3.08s/it, loss=0.377, lr=8.76e-5] Steps: 75%|███████▌ | 753/1000 [30:14<11:10, 2.72s/it, loss=0.377, lr=8.76e-5] Steps: 75%|███████▌ | 753/1000 [30:14<11:10, 2.72s/it, loss=0.544, lr=8.69e-5] Steps: 75%|███████▌ | 754/1000 [30:16<10:05, 2.46s/it, loss=0.544, lr=8.69e-5] Steps: 75%|███████▌ | 754/1000 [30:16<10:05, 2.46s/it, loss=0.281, lr=8.63e-5] Steps: 76%|███████▌ | 755/1000 [30:17<09:19, 2.28s/it, loss=0.281, lr=8.63e-5] Steps: 76%|███████▌ | 755/1000 [30:17<09:19, 2.28s/it, loss=1.04, lr=8.56e-5] Steps: 76%|███████▌ | 756/1000 [30:19<08:46, 2.16s/it, loss=1.04, lr=8.56e-5] Steps: 76%|███████▌ | 756/1000 [30:19<08:46, 2.16s/it, loss=0.313, lr=8.5e-5] Steps: 76%|███████▌ | 757/1000 [30:21<08:23, 2.07s/it, loss=0.313, lr=8.5e-5] Steps: 76%|███████▌ | 757/1000 [30:21<08:23, 2.07s/it, loss=0.71, lr=8.44e-5] Steps: 76%|███████▌ | 758/1000 [30:23<08:07, 2.01s/it, loss=0.71, lr=8.44e-5] Steps: 76%|███████▌ | 758/1000 [30:23<08:07, 2.01s/it, loss=0.784, lr=8.37e-5] Steps: 76%|███████▌ | 759/1000 [30:25<07:55, 1.97s/it, loss=0.784, lr=8.37e-5] Steps: 76%|███████▌ | 759/1000 [30:25<07:55, 1.97s/it, loss=0.59, lr=8.31e-5] Steps: 76%|███████▌ | 760/1000 [30:27<07:45, 1.94s/it, loss=0.59, lr=8.31e-5] Steps: 76%|███████▌ | 760/1000 [30:27<07:45, 1.94s/it, loss=0.81, lr=8.24e-5] Steps: 76%|███████▌ | 761/1000 [30:29<07:38, 1.92s/it, loss=0.81, lr=8.24e-5] Steps: 76%|███████▌ | 761/1000 [30:29<07:38, 1.92s/it, loss=0.661, lr=8.18e-5] Steps: 76%|███████▌ | 762/1000 [30:31<07:33, 1.90s/it, loss=0.661, lr=8.18e-5] Steps: 76%|███████▌ | 762/1000 [30:31<07:33, 1.90s/it, loss=0.452, lr=8.12e-5] Steps: 76%|███████▋ | 763/1000 [30:32<07:28, 1.89s/it, loss=0.452, lr=8.12e-5] Steps: 76%|███████▋ | 763/1000 [30:32<07:28, 1.89s/it, loss=0.422, lr=8.05e-5] Steps: 76%|███████▋ | 764/1000 [30:34<07:25, 1.89s/it, loss=0.422, lr=8.05e-5] Steps: 76%|███████▋ | 764/1000 [30:34<07:25, 1.89s/it, loss=0.53, lr=7.99e-5] Steps: 76%|███████▋ | 765/1000 [30:36<07:22, 1.88s/it, loss=0.53, lr=7.99e-5] Steps: 76%|███████▋ | 765/1000 [30:36<07:22, 1.88s/it, loss=0.322, lr=7.93e-5] Steps: 77%|███████▋ | 766/1000 [30:38<07:19, 1.88s/it, loss=0.322, lr=7.93e-5] Steps: 77%|███████▋ | 766/1000 [30:38<07:19, 1.88s/it, loss=0.702, lr=7.87e-5] Steps: 77%|███████▋ | 767/1000 [30:40<07:17, 1.88s/it, loss=0.702, lr=7.87e-5] Steps: 77%|███████▋ | 767/1000 [30:40<07:17, 1.88s/it, loss=0.364, lr=7.8e-5] Steps: 77%|███████▋ | 768/1000 [30:42<07:15, 1.88s/it, loss=0.364, lr=7.8e-5] Steps: 77%|███████▋ | 768/1000 [30:42<07:15, 1.88s/it, loss=0.327, lr=7.74e-5] Steps: 77%|███████▋ | 769/1000 [30:44<07:13, 1.88s/it, loss=0.327, lr=7.74e-5] Steps: 77%|███████▋ | 769/1000 [30:44<07:13, 1.88s/it, loss=0.331, lr=7.68e-5] Steps: 77%|███████▋ | 770/1000 [30:46<07:11, 1.87s/it, loss=0.331, lr=7.68e-5] Steps: 77%|███████▋ | 770/1000 [30:46<07:11, 1.87s/it, loss=0.455, lr=7.62e-5] Steps: 77%|███████▋ | 771/1000 [30:47<07:09, 1.88s/it, loss=0.455, lr=7.62e-5] Steps: 77%|███████▋ | 771/1000 [30:47<07:09, 1.88s/it, loss=0.404, lr=7.56e-5] Steps: 77%|███████▋ | 772/1000 [30:49<07:07, 1.87s/it, loss=0.404, lr=7.56e-5] Steps: 77%|███████▋ | 772/1000 [30:49<07:07, 1.87s/it, loss=1.01, lr=7.5e-5] Steps: 77%|███████▋ | 773/1000 [30:51<07:05, 1.87s/it, loss=1.01, lr=7.5e-5] Steps: 77%|███████▋ | 773/1000 [30:51<07:05, 1.87s/it, loss=0.762, lr=7.43e-5] Steps: 77%|███████▋ | 774/1000 [30:53<07:03, 1.88s/it, loss=0.762, lr=7.43e-5] Steps: 77%|███████▋ | 774/1000 [30:53<07:03, 1.88s/it, loss=0.371, lr=7.37e-5] Steps: 78%|███████▊ | 775/1000 [30:55<07:01, 1.88s/it, loss=0.371, lr=7.37e-5] Steps: 78%|███████▊ | 775/1000 [30:55<07:01, 1.88s/it, loss=0.976, lr=7.31e-5] Steps: 78%|███████▊ | 776/1000 [30:57<06:59, 1.87s/it, loss=0.976, lr=7.31e-5] Steps: 78%|███████▊ | 776/1000 [30:57<06:59, 1.87s/it, loss=0.998, lr=7.25e-5] Steps: 78%|███████▊ | 777/1000 [30:59<06:57, 1.87s/it, loss=0.998, lr=7.25e-5] Steps: 78%|███████▊ | 777/1000 [30:59<06:57, 1.87s/it, loss=0.697, lr=7.19e-5] Steps: 78%|███████▊ | 778/1000 [31:01<06:55, 1.87s/it, loss=0.697, lr=7.19e-5] Steps: 78%|███████▊ | 778/1000 [31:01<06:55, 1.87s/it, loss=0.637, lr=7.13e-5] Steps: 78%|███████▊ | 779/1000 [31:02<06:53, 1.87s/it, loss=0.637, lr=7.13e-5] Steps: 78%|███████▊ | 779/1000 [31:02<06:53, 1.87s/it, loss=0.719, lr=7.07e-5] Steps: 78%|███████▊ | 780/1000 [31:04<06:52, 1.87s/it, loss=0.719, lr=7.07e-5] Steps: 78%|███████▊ | 780/1000 [31:04<06:52, 1.87s/it, loss=0.402, lr=7.01e-5] Steps: 78%|███████▊ | 781/1000 [31:12<13:11, 3.61s/it, loss=0.402, lr=7.01e-5] Steps: 78%|███████▊ | 781/1000 [31:12<13:11, 3.61s/it, loss=0.353, lr=6.95e-5] Steps: 78%|███████▊ | 782/1000 [31:14<11:13, 3.09s/it, loss=0.353, lr=6.95e-5] Steps: 78%|███████▊ | 782/1000 [31:14<11:13, 3.09s/it, loss=0.519, lr=6.89e-5] Steps: 78%|███████▊ | 783/1000 [31:16<09:51, 2.72s/it, loss=0.519, lr=6.89e-5] Steps: 78%|███████▊ | 783/1000 [31:16<09:51, 2.72s/it, loss=0.33, lr=6.83e-5] Steps: 78%|███████▊ | 784/1000 [31:18<08:53, 2.47s/it, loss=0.33, lr=6.83e-5] Steps: 78%|███████▊ | 784/1000 [31:18<08:53, 2.47s/it, loss=0.923, lr=6.77e-5] Steps: 78%|███████▊ | 785/1000 [31:19<08:12, 2.29s/it, loss=0.923, lr=6.77e-5] Steps: 78%|███████▊ | 785/1000 [31:19<08:12, 2.29s/it, loss=0.756, lr=6.71e-5] Steps: 79%|███████▊ | 786/1000 [31:21<07:42, 2.16s/it, loss=0.756, lr=6.71e-5] Steps: 79%|███████▊ | 786/1000 [31:21<07:42, 2.16s/it, loss=0.279, lr=6.66e-5] Steps: 79%|███████▊ | 787/1000 [31:23<07:22, 2.08s/it, loss=0.279, lr=6.66e-5] Steps: 79%|███████▊ | 787/1000 [31:23<07:22, 2.08s/it, loss=0.66, lr=6.6e-5] Steps: 79%|███████▉ | 788/1000 [31:25<07:07, 2.01s/it, loss=0.66, lr=6.6e-5] Steps: 79%|███████▉ | 788/1000 [31:25<07:07, 2.01s/it, loss=0.417, lr=6.54e-5] Steps: 79%|███████▉ | 789/1000 [31:27<06:56, 1.97s/it, loss=0.417, lr=6.54e-5] Steps: 79%|███████▉ | 789/1000 [31:27<06:56, 1.97s/it, loss=0.785, lr=6.48e-5] Steps: 79%|███████▉ | 790/1000 [31:29<06:48, 1.94s/it, loss=0.785, lr=6.48e-5] Steps: 79%|███████▉ | 790/1000 [31:29<06:48, 1.94s/it, loss=0.428, lr=6.42e-5] Steps: 79%|███████▉ | 791/1000 [31:31<06:42, 1.92s/it, loss=0.428, lr=6.42e-5] Steps: 79%|███████▉ | 791/1000 [31:31<06:42, 1.92s/it, loss=0.274, lr=6.37e-5] Steps: 79%|███████▉ | 792/1000 [31:33<06:36, 1.91s/it, loss=0.274, lr=6.37e-5] Steps: 79%|███████▉ | 792/1000 [31:33<06:36, 1.91s/it, loss=0.798, lr=6.31e-5] Steps: 79%|███████▉ | 793/1000 [31:34<06:32, 1.90s/it, loss=0.798, lr=6.31e-5] Steps: 79%|███████▉ | 793/1000 [31:34<06:32, 1.90s/it, loss=0.288, lr=6.25e-5] Steps: 79%|███████▉ | 794/1000 [31:36<06:28, 1.89s/it, loss=0.288, lr=6.25e-5] Steps: 79%|███████▉ | 794/1000 [31:36<06:28, 1.89s/it, loss=0.728, lr=6.19e-5] Steps: 80%|███████▉ | 795/1000 [31:38<06:26, 1.88s/it, loss=0.728, lr=6.19e-5] Steps: 80%|███████▉ | 795/1000 [31:38<06:26, 1.88s/it, loss=0.617, lr=6.14e-5] Steps: 80%|███████▉ | 796/1000 [31:40<06:23, 1.88s/it, loss=0.617, lr=6.14e-5] Steps: 80%|███████▉ | 796/1000 [31:40<06:23, 1.88s/it, loss=0.68, lr=6.08e-5] Steps: 80%|███████▉ | 797/1000 [31:42<06:21, 1.88s/it, loss=0.68, lr=6.08e-5] Steps: 80%|███████▉ | 797/1000 [31:42<06:21, 1.88s/it, loss=0.577, lr=6.03e-5] Steps: 80%|███████▉ | 798/1000 [31:44<06:19, 1.88s/it, loss=0.577, lr=6.03e-5] Steps: 80%|███████▉ | 798/1000 [31:44<06:19, 1.88s/it, loss=0.427, lr=5.97e-5] Steps: 80%|███████▉ | 799/1000 [31:46<06:17, 1.88s/it, loss=0.427, lr=5.97e-5] Steps: 80%|███████▉ | 799/1000 [31:46<06:17, 1.88s/it, loss=0.317, lr=5.91e-5] Steps: 80%|████████ | 800/1000 [31:48<06:15, 1.88s/it, loss=0.317, lr=5.91e-5] Steps: 80%|████████ | 800/1000 [31:48<06:15, 1.88s/it, loss=0.501, lr=5.86e-5] Steps: 80%|████████ | 801/1000 [31:49<06:13, 1.87s/it, loss=0.501, lr=5.86e-5] Steps: 80%|████████ | 801/1000 [31:49<06:13, 1.87s/it, loss=0.288, lr=5.8e-5] Steps: 80%|████████ | 802/1000 [31:51<06:11, 1.87s/it, loss=0.288, lr=5.8e-5] Steps: 80%|████████ | 802/1000 [31:51<06:11, 1.87s/it, loss=0.983, lr=5.75e-5] Steps: 80%|████████ | 803/1000 [31:53<06:09, 1.87s/it, loss=0.983, lr=5.75e-5] Steps: 80%|████████ | 803/1000 [31:53<06:09, 1.87s/it, loss=0.357, lr=5.69e-5] Steps: 80%|████████ | 804/1000 [31:55<06:07, 1.87s/it, loss=0.357, lr=5.69e-5] Steps: 80%|████████ | 804/1000 [31:55<06:07, 1.87s/it, loss=0.227, lr=5.64e-5] Steps: 80%|████████ | 805/1000 [31:57<06:05, 1.87s/it, loss=0.227, lr=5.64e-5] Steps: 80%|████████ | 805/1000 [31:57<06:05, 1.87s/it, loss=0.467, lr=5.58e-5] Steps: 81%|████████ | 806/1000 [31:59<06:03, 1.87s/it, loss=0.467, lr=5.58e-5] Steps: 81%|████████ | 806/1000 [31:59<06:03, 1.87s/it, loss=0.31, lr=5.53e-5] Steps: 81%|████████ | 807/1000 [32:01<06:01, 1.87s/it, loss=0.31, lr=5.53e-5] Steps: 81%|████████ | 807/1000 [32:01<06:01, 1.87s/it, loss=0.681, lr=5.47e-5] Steps: 81%|████████ | 808/1000 [32:02<05:59, 1.87s/it, loss=0.681, lr=5.47e-5] Steps: 81%|████████ | 808/1000 [32:03<05:59, 1.87s/it, loss=0.285, lr=5.42e-5] Steps: 81%|████████ | 809/1000 [32:04<05:58, 1.87s/it, loss=0.285, lr=5.42e-5] Steps: 81%|████████ | 809/1000 [32:04<05:58, 1.87s/it, loss=0.362, lr=5.37e-5] Steps: 81%|████████ | 810/1000 [32:06<05:56, 1.88s/it, loss=0.362, lr=5.37e-5] Steps: 81%|████████ | 810/1000 [32:06<05:56, 1.88s/it, loss=0.746, lr=5.31e-5] Steps: 81%|████████ | 811/1000 [32:14<11:29, 3.65s/it, loss=0.746, lr=5.31e-5] Steps: 81%|████████ | 811/1000 [32:14<11:29, 3.65s/it, loss=1.01, lr=5.26e-5] Steps: 81%|████████ | 812/1000 [32:16<09:45, 3.11s/it, loss=1.01, lr=5.26e-5] Steps: 81%|████████ | 812/1000 [32:16<09:45, 3.11s/it, loss=0.306, lr=5.21e-5] Steps: 81%|████████▏ | 813/1000 [32:18<08:32, 2.74s/it, loss=0.306, lr=5.21e-5] Steps: 81%|████████▏ | 813/1000 [32:18<08:32, 2.74s/it, loss=0.647, lr=5.15e-5] Steps: 81%|████████▏ | 814/1000 [32:20<07:41, 2.48s/it, loss=0.647, lr=5.15e-5] Steps: 81%|████████▏ | 814/1000 [32:20<07:41, 2.48s/it, loss=0.805, lr=5.1e-5] Steps: 82%|████████▏ | 815/1000 [32:22<07:04, 2.30s/it, loss=0.805, lr=5.1e-5] Steps: 82%|████████▏ | 815/1000 [32:22<07:04, 2.30s/it, loss=0.457, lr=5.05e-5] Steps: 82%|████████▏ | 816/1000 [32:23<06:39, 2.17s/it, loss=0.457, lr=5.05e-5] Steps: 82%|████████▏ | 816/1000 [32:23<06:39, 2.17s/it, loss=0.582, lr=5e-5] Steps: 82%|████████▏ | 817/1000 [32:25<06:20, 2.08s/it, loss=0.582, lr=5e-5] Steps: 82%|████████▏ | 817/1000 [32:25<06:20, 2.08s/it, loss=0.313, lr=4.95e-5] Steps: 82%|████████▏ | 818/1000 [32:27<06:07, 2.02s/it, loss=0.313, lr=4.95e-5] Steps: 82%|████████▏ | 818/1000 [32:27<06:07, 2.02s/it, loss=0.524, lr=4.89e-5] Steps: 82%|████████▏ | 819/1000 [32:29<05:57, 1.97s/it, loss=0.524, lr=4.89e-5] Steps: 82%|████████▏ | 819/1000 [32:29<05:57, 1.97s/it, loss=0.812, lr=4.84e-5] Steps: 82%|████████▏ | 820/1000 [32:31<05:49, 1.94s/it, loss=0.812, lr=4.84e-5] Steps: 82%|████████▏ | 820/1000 [32:31<05:49, 1.94s/it, loss=0.401, lr=4.79e-5] Steps: 82%|████████▏ | 821/1000 [32:33<05:43, 1.92s/it, loss=0.401, lr=4.79e-5] Steps: 82%|████████▏ | 821/1000 [32:33<05:43, 1.92s/it, loss=0.325, lr=4.74e-5] Steps: 82%|████████▏ | 822/1000 [32:35<05:39, 1.91s/it, loss=0.325, lr=4.74e-5] Steps: 82%|████████▏ | 822/1000 [32:35<05:39, 1.91s/it, loss=0.639, lr=4.69e-5] Steps: 82%|████████▏ | 823/1000 [32:36<05:35, 1.89s/it, loss=0.639, lr=4.69e-5] Steps: 82%|████████▏ | 823/1000 [32:36<05:35, 1.89s/it, loss=0.799, lr=4.64e-5] Steps: 82%|████████▏ | 824/1000 [32:38<05:32, 1.89s/it, loss=0.799, lr=4.64e-5] Steps: 82%|████████▏ | 824/1000 [32:38<05:32, 1.89s/it, loss=0.505, lr=4.59e-5] Steps: 82%|████████▎ | 825/1000 [32:40<05:29, 1.88s/it, loss=0.505, lr=4.59e-5] Steps: 82%|████████▎ | 825/1000 [32:40<05:29, 1.88s/it, loss=0.998, lr=4.54e-5] Steps: 83%|████████▎ | 826/1000 [32:42<05:27, 1.88s/it, loss=0.998, lr=4.54e-5] Steps: 83%|████████▎ | 826/1000 [32:42<05:27, 1.88s/it, loss=0.37, lr=4.49e-5] Steps: 83%|████████▎ | 827/1000 [32:44<05:24, 1.88s/it, loss=0.37, lr=4.49e-5] Steps: 83%|████████▎ | 827/1000 [32:44<05:24, 1.88s/it, loss=0.67, lr=4.44e-5] Steps: 83%|████████▎ | 828/1000 [32:46<05:22, 1.88s/it, loss=0.67, lr=4.44e-5] Steps: 83%|████████▎ | 828/1000 [32:46<05:22, 1.88s/it, loss=0.298, lr=4.39e-5] Steps: 83%|████████▎ | 829/1000 [32:48<05:20, 1.87s/it, loss=0.298, lr=4.39e-5] Steps: 83%|████████▎ | 829/1000 [32:48<05:20, 1.87s/it, loss=0.783, lr=4.34e-5] Steps: 83%|████████▎ | 830/1000 [32:50<05:18, 1.87s/it, loss=0.783, lr=4.34e-5] Steps: 83%|████████▎ | 830/1000 [32:50<05:18, 1.87s/it, loss=0.355, lr=4.29e-5] Steps: 83%|████████▎ | 831/1000 [32:51<05:16, 1.87s/it, loss=0.355, lr=4.29e-5] Steps: 83%|████████▎ | 831/1000 [32:51<05:16, 1.87s/it, loss=0.796, lr=4.25e-5] Steps: 83%|████████▎ | 832/1000 [32:53<05:14, 1.87s/it, loss=0.796, lr=4.25e-5] Steps: 83%|████████▎ | 832/1000 [32:53<05:14, 1.87s/it, loss=0.467, lr=4.2e-5] Steps: 83%|████████▎ | 833/1000 [32:55<05:12, 1.87s/it, loss=0.467, lr=4.2e-5] Steps: 83%|████████▎ | 833/1000 [32:55<05:12, 1.87s/it, loss=0.29, lr=4.15e-5] Steps: 83%|████████▎ | 834/1000 [32:57<05:10, 1.87s/it, loss=0.29, lr=4.15e-5] Steps: 83%|████████▎ | 834/1000 [32:57<05:10, 1.87s/it, loss=0.222, lr=4.1e-5] Steps: 84%|████████▎ | 835/1000 [32:59<05:08, 1.87s/it, loss=0.222, lr=4.1e-5] Steps: 84%|████████▎ | 835/1000 [32:59<05:08, 1.87s/it, loss=0.359, lr=4.05e-5] Steps: 84%|████████▎ | 836/1000 [33:01<05:06, 1.87s/it, loss=0.359, lr=4.05e-5] Steps: 84%|████████▎ | 836/1000 [33:01<05:06, 1.87s/it, loss=0.51, lr=4.01e-5] Steps: 84%|████████▎ | 837/1000 [33:03<05:04, 1.87s/it, loss=0.51, lr=4.01e-5] Steps: 84%|████████▎ | 837/1000 [33:03<05:04, 1.87s/it, loss=0.674, lr=3.96e-5] Steps: 84%|████████▍ | 838/1000 [33:05<05:02, 1.87s/it, loss=0.674, lr=3.96e-5] Steps: 84%|████████▍ | 838/1000 [33:05<05:02, 1.87s/it, loss=0.796, lr=3.91e-5] Steps: 84%|████████▍ | 839/1000 [33:06<05:00, 1.87s/it, loss=0.796, lr=3.91e-5] Steps: 84%|████████▍ | 839/1000 [33:06<05:00, 1.87s/it, loss=0.477, lr=3.87e-5] Steps: 84%|████████▍ | 840/1000 [33:08<04:58, 1.87s/it, loss=0.477, lr=3.87e-5] Steps: 84%|████████▍ | 840/1000 [33:08<04:58, 1.87s/it, loss=0.394, lr=3.82e-5] Steps: 84%|████████▍ | 841/1000 [33:16<09:41, 3.66s/it, loss=0.394, lr=3.82e-5] Steps: 84%|████████▍ | 841/1000 [33:16<09:41, 3.66s/it, loss=0.318, lr=3.77e-5] Steps: 84%|████████▍ | 842/1000 [33:18<08:12, 3.12s/it, loss=0.318, lr=3.77e-5] Steps: 84%|████████▍ | 842/1000 [33:18<08:12, 3.12s/it, loss=0.402, lr=3.73e-5] Steps: 84%|████████▍ | 843/1000 [33:20<07:11, 2.75s/it, loss=0.402, lr=3.73e-5] Steps: 84%|████████▍ | 843/1000 [33:20<07:11, 2.75s/it, loss=0.834, lr=3.68e-5] Steps: 84%|████████▍ | 844/1000 [33:22<06:27, 2.49s/it, loss=0.834, lr=3.68e-5] Steps: 84%|████████▍ | 844/1000 [33:22<06:27, 2.49s/it, loss=0.346, lr=3.64e-5] Steps: 84%|████████▍ | 845/1000 [33:24<05:56, 2.30s/it, loss=0.346, lr=3.64e-5] Steps: 84%|████████▍ | 845/1000 [33:24<05:56, 2.30s/it, loss=0.486, lr=3.59e-5] Steps: 85%|████████▍ | 846/1000 [33:25<05:34, 2.17s/it, loss=0.486, lr=3.59e-5] Steps: 85%|████████▍ | 846/1000 [33:25<05:34, 2.17s/it, loss=0.326, lr=3.55e-5] Steps: 85%|████████▍ | 847/1000 [33:27<05:18, 2.08s/it, loss=0.326, lr=3.55e-5] Steps: 85%|████████▍ | 847/1000 [33:27<05:18, 2.08s/it, loss=0.328, lr=3.5e-5] Steps: 85%|████████▍ | 848/1000 [33:29<05:06, 2.02s/it, loss=0.328, lr=3.5e-5] Steps: 85%|████████▍ | 848/1000 [33:29<05:06, 2.02s/it, loss=0.697, lr=3.46e-5] Steps: 85%|████████▍ | 849/1000 [33:31<04:58, 1.97s/it, loss=0.697, lr=3.46e-5] Steps: 85%|████████▍ | 849/1000 [33:31<04:58, 1.97s/it, loss=0.375, lr=3.41e-5] Steps: 85%|████████▌ | 850/1000 [33:33<04:51, 1.94s/it, loss=0.375, lr=3.41e-5] Steps: 85%|████████▌ | 850/1000 [33:33<04:51, 1.94s/it, loss=0.996, lr=3.37e-5] Steps: 85%|████████▌ | 851/1000 [33:35<04:46, 1.92s/it, loss=0.996, lr=3.37e-5] Steps: 85%|████████▌ | 851/1000 [33:35<04:46, 1.92s/it, loss=0.817, lr=3.33e-5] Steps: 85%|████████▌ | 852/1000 [33:37<04:42, 1.91s/it, loss=0.817, lr=3.33e-5] Steps: 85%|████████▌ | 852/1000 [33:37<04:42, 1.91s/it, loss=0.285, lr=3.28e-5] Steps: 85%|████████▌ | 853/1000 [33:39<04:38, 1.90s/it, loss=0.285, lr=3.28e-5] Steps: 85%|████████▌ | 853/1000 [33:39<04:38, 1.90s/it, loss=0.641, lr=3.24e-5] Steps: 85%|████████▌ | 854/1000 [33:40<04:35, 1.89s/it, loss=0.641, lr=3.24e-5] Steps: 85%|████████▌ | 854/1000 [33:40<04:35, 1.89s/it, loss=0.678, lr=3.2e-5] Steps: 86%|████████▌ | 855/1000 [33:42<04:33, 1.89s/it, loss=0.678, lr=3.2e-5] Steps: 86%|████████▌ | 855/1000 [33:42<04:33, 1.89s/it, loss=0.953, lr=3.16e-5] Steps: 86%|████████▌ | 856/1000 [33:44<04:31, 1.88s/it, loss=0.953, lr=3.16e-5] Steps: 86%|████████▌ | 856/1000 [33:44<04:31, 1.88s/it, loss=0.33, lr=3.11e-5] Steps: 86%|████████▌ | 857/1000 [33:46<04:28, 1.88s/it, loss=0.33, lr=3.11e-5] Steps: 86%|████████▌ | 857/1000 [33:46<04:28, 1.88s/it, loss=0.782, lr=3.07e-5] Steps: 86%|████████▌ | 858/1000 [33:48<04:26, 1.88s/it, loss=0.782, lr=3.07e-5] Steps: 86%|████████▌ | 858/1000 [33:48<04:26, 1.88s/it, loss=0.652, lr=3.03e-5] Steps: 86%|████████▌ | 859/1000 [33:50<04:24, 1.88s/it, loss=0.652, lr=3.03e-5] Steps: 86%|████████▌ | 859/1000 [33:50<04:24, 1.88s/it, loss=0.55, lr=2.99e-5] Steps: 86%|████████▌ | 860/1000 [33:52<04:22, 1.88s/it, loss=0.55, lr=2.99e-5] Steps: 86%|████████▌ | 860/1000 [33:52<04:22, 1.88s/it, loss=0.467, lr=2.95e-5] Steps: 86%|████████▌ | 861/1000 [33:54<04:20, 1.88s/it, loss=0.467, lr=2.95e-5] Steps: 86%|████████▌ | 861/1000 [33:54<04:20, 1.88s/it, loss=0.636, lr=2.91e-5] Steps: 86%|████████▌ | 862/1000 [33:55<04:18, 1.87s/it, loss=0.636, lr=2.91e-5] Steps: 86%|████████▌ | 862/1000 [33:55<04:18, 1.87s/it, loss=0.502, lr=2.87e-5] Steps: 86%|████████▋ | 863/1000 [33:57<04:16, 1.87s/it, loss=0.502, lr=2.87e-5] Steps: 86%|████████▋ | 863/1000 [33:57<04:16, 1.87s/it, loss=0.29, lr=2.83e-5] Steps: 86%|████████▋ | 864/1000 [33:59<04:14, 1.87s/it, loss=0.29, lr=2.83e-5] Steps: 86%|████████▋ | 864/1000 [33:59<04:14, 1.87s/it, loss=0.379, lr=2.79e-5] Steps: 86%|████████▋ | 865/1000 [34:01<04:12, 1.87s/it, loss=0.379, lr=2.79e-5] Steps: 86%|████████▋ | 865/1000 [34:01<04:12, 1.87s/it, loss=0.47, lr=2.75e-5] Steps: 87%|████████▋ | 866/1000 [34:03<04:10, 1.87s/it, loss=0.47, lr=2.75e-5] Steps: 87%|████████▋ | 866/1000 [34:03<04:10, 1.87s/it, loss=0.333, lr=2.71e-5] Steps: 87%|████████▋ | 867/1000 [34:05<04:09, 1.87s/it, loss=0.333, lr=2.71e-5] Steps: 87%|████████▋ | 867/1000 [34:05<04:09, 1.87s/it, loss=0.916, lr=2.67e-5] Steps: 87%|████████▋ | 868/1000 [34:07<04:07, 1.87s/it, loss=0.916, lr=2.67e-5] Steps: 87%|████████▋ | 868/1000 [34:07<04:07, 1.87s/it, loss=0.406, lr=2.63e-5] Steps: 87%|████████▋ | 869/1000 [34:09<04:05, 1.87s/it, loss=0.406, lr=2.63e-5] Steps: 87%|████████▋ | 869/1000 [34:09<04:05, 1.87s/it, loss=0.387, lr=2.59e-5] Steps: 87%|████████▋ | 870/1000 [34:10<04:03, 1.87s/it, loss=0.387, lr=2.59e-5] Steps: 87%|████████▋ | 870/1000 [34:10<04:03, 1.87s/it, loss=0.272, lr=2.55e-5] Steps: 87%|████████▋ | 871/1000 [34:18<07:45, 3.61s/it, loss=0.272, lr=2.55e-5] Steps: 87%|████████▋ | 871/1000 [34:18<07:45, 3.61s/it, loss=0.311, lr=2.51e-5] Steps: 87%|████████▋ | 872/1000 [34:20<06:35, 3.09s/it, loss=0.311, lr=2.51e-5] Steps: 87%|████████▋ | 872/1000 [34:20<06:35, 3.09s/it, loss=0.616, lr=2.47e-5] Steps: 87%|████████▋ | 873/1000 [34:22<05:45, 2.72s/it, loss=0.616, lr=2.47e-5] Steps: 87%|████████▋ | 873/1000 [34:22<05:45, 2.72s/it, loss=0.909, lr=2.44e-5] Steps: 87%|████████▋ | 874/1000 [34:24<05:10, 2.47s/it, loss=0.909, lr=2.44e-5] Steps: 87%|████████▋ | 874/1000 [34:24<05:10, 2.47s/it, loss=0.92, lr=2.4e-5] Steps: 88%|████████▊ | 875/1000 [34:26<04:45, 2.29s/it, loss=0.92, lr=2.4e-5] Steps: 88%|████████▊ | 875/1000 [34:26<04:45, 2.29s/it, loss=0.308, lr=2.36e-5] Steps: 88%|████████▊ | 876/1000 [34:27<04:28, 2.16s/it, loss=0.308, lr=2.36e-5] Steps: 88%|████████▊ | 876/1000 [34:27<04:28, 2.16s/it, loss=0.602, lr=2.32e-5] Steps: 88%|████████▊ | 877/1000 [34:29<04:15, 2.07s/it, loss=0.602, lr=2.32e-5] Steps: 88%|████████▊ | 877/1000 [34:29<04:15, 2.07s/it, loss=0.335, lr=2.29e-5] Steps: 88%|████████▊ | 878/1000 [34:31<04:05, 2.01s/it, loss=0.335, lr=2.29e-5] Steps: 88%|████████▊ | 878/1000 [34:31<04:05, 2.01s/it, loss=0.42, lr=2.25e-5] Steps: 88%|████████▊ | 879/1000 [34:33<03:58, 1.97s/it, loss=0.42, lr=2.25e-5] Steps: 88%|████████▊ | 879/1000 [34:33<03:58, 1.97s/it, loss=0.296, lr=2.22e-5] Steps: 88%|████████▊ | 880/1000 [34:35<03:52, 1.94s/it, loss=0.296, lr=2.22e-5] Steps: 88%|████████▊ | 880/1000 [34:35<03:52, 1.94s/it, loss=0.369, lr=2.18e-5] Steps: 88%|████████▊ | 881/1000 [34:37<03:48, 1.92s/it, loss=0.369, lr=2.18e-5] Steps: 88%|████████▊ | 881/1000 [34:37<03:48, 1.92s/it, loss=0.855, lr=2.14e-5] Steps: 88%|████████▊ | 882/1000 [34:39<03:44, 1.90s/it, loss=0.855, lr=2.14e-5] Steps: 88%|████████▊ | 882/1000 [34:39<03:44, 1.90s/it, loss=0.897, lr=2.11e-5] Steps: 88%|████████▊ | 883/1000 [34:41<03:41, 1.90s/it, loss=0.897, lr=2.11e-5] Steps: 88%|████████▊ | 883/1000 [34:41<03:41, 1.90s/it, loss=0.313, lr=2.07e-5] Steps: 88%|████████▊ | 884/1000 [34:42<03:39, 1.89s/it, loss=0.313, lr=2.07e-5] Steps: 88%|████████▊ | 884/1000 [34:42<03:39, 1.89s/it, loss=0.438, lr=2.04e-5] Steps: 88%|████████▊ | 885/1000 [34:44<03:36, 1.88s/it, loss=0.438, lr=2.04e-5] Steps: 88%|████████▊ | 885/1000 [34:44<03:36, 1.88s/it, loss=1.02, lr=2.01e-5] Steps: 89%|████████▊ | 886/1000 [34:46<03:34, 1.88s/it, loss=1.02, lr=2.01e-5] Steps: 89%|████████▊ | 886/1000 [34:46<03:34, 1.88s/it, loss=1.03, lr=1.97e-5] Steps: 89%|████████▊ | 887/1000 [34:48<03:32, 1.88s/it, loss=1.03, lr=1.97e-5] Steps: 89%|████████▊ | 887/1000 [34:48<03:32, 1.88s/it, loss=0.438, lr=1.94e-5] Steps: 89%|████████▉ | 888/1000 [34:50<03:30, 1.88s/it, loss=0.438, lr=1.94e-5] Steps: 89%|████████▉ | 888/1000 [34:50<03:30, 1.88s/it, loss=0.478, lr=1.9e-5] Steps: 89%|████████▉ | 889/1000 [34:52<03:28, 1.88s/it, loss=0.478, lr=1.9e-5] Steps: 89%|████████▉ | 889/1000 [34:52<03:28, 1.88s/it, loss=0.345, lr=1.87e-5] Steps: 89%|████████▉ | 890/1000 [34:54<03:26, 1.87s/it, loss=0.345, lr=1.87e-5] Steps: 89%|████████▉ | 890/1000 [34:54<03:26, 1.87s/it, loss=0.646, lr=1.84e-5] Steps: 89%|████████▉ | 891/1000 [34:55<03:24, 1.87s/it, loss=0.646, lr=1.84e-5] Steps: 89%|████████▉ | 891/1000 [34:56<03:24, 1.87s/it, loss=0.328, lr=1.8e-5] Steps: 89%|████████▉ | 892/1000 [34:57<03:22, 1.87s/it, loss=0.328, lr=1.8e-5] Steps: 89%|████████▉ | 892/1000 [34:57<03:22, 1.87s/it, loss=0.561, lr=1.77e-5] Steps: 89%|████████▉ | 893/1000 [34:59<03:20, 1.87s/it, loss=0.561, lr=1.77e-5] Steps: 89%|████████▉ | 893/1000 [34:59<03:20, 1.87s/it, loss=0.326, lr=1.74e-5] Steps: 89%|████████▉ | 894/1000 [35:01<03:18, 1.87s/it, loss=0.326, lr=1.74e-5] Steps: 89%|████████▉ | 894/1000 [35:01<03:18, 1.87s/it, loss=0.299, lr=1.71e-5] Steps: 90%|████████▉ | 895/1000 [35:03<03:16, 1.87s/it, loss=0.299, lr=1.71e-5] Steps: 90%|████████▉ | 895/1000 [35:03<03:16, 1.87s/it, loss=0.333, lr=1.68e-5] Steps: 90%|████████▉ | 896/1000 [35:05<03:14, 1.87s/it, loss=0.333, lr=1.68e-5] Steps: 90%|████████▉ | 896/1000 [35:05<03:14, 1.87s/it, loss=0.362, lr=1.64e-5] Steps: 90%|████████▉ | 897/1000 [35:07<03:12, 1.87s/it, loss=0.362, lr=1.64e-5] Steps: 90%|████████▉ | 897/1000 [35:07<03:12, 1.87s/it, loss=0.921, lr=1.61e-5] Steps: 90%|████████▉ | 898/1000 [35:09<03:11, 1.87s/it, loss=0.921, lr=1.61e-5] Steps: 90%|████████▉ | 898/1000 [35:09<03:11, 1.87s/it, loss=0.953, lr=1.58e-5] Steps: 90%|████████▉ | 899/1000 [35:10<03:09, 1.87s/it, loss=0.953, lr=1.58e-5] Steps: 90%|████████▉ | 899/1000 [35:10<03:09, 1.87s/it, loss=0.381, lr=1.55e-5] Steps: 90%|█████████ | 900/1000 [35:12<03:07, 1.87s/it, loss=0.381, lr=1.55e-5] Steps: 90%|█████████ | 900/1000 [35:12<03:07, 1.87s/it, loss=0.381, lr=1.52e-5] Steps: 90%|█████████ | 901/1000 [35:20<06:00, 3.64s/it, loss=0.381, lr=1.52e-5] Steps: 90%|█████████ | 901/1000 [35:20<06:00, 3.64s/it, loss=0.543, lr=1.49e-5] Steps: 90%|█████████ | 902/1000 [35:22<05:04, 3.11s/it, loss=0.543, lr=1.49e-5] Steps: 90%|█████████ | 902/1000 [35:22<05:04, 3.11s/it, loss=0.503, lr=1.46e-5] Steps: 90%|█████████ | 903/1000 [35:24<04:25, 2.74s/it, loss=0.503, lr=1.46e-5] Steps: 90%|█████████ | 903/1000 [35:24<04:25, 2.74s/it, loss=0.302, lr=1.43e-5] Steps: 90%|█████████ | 904/1000 [35:26<03:57, 2.48s/it, loss=0.302, lr=1.43e-5] Steps: 90%|█████████ | 904/1000 [35:26<03:57, 2.48s/it, loss=0.296, lr=1.4e-5] Steps: 90%|█████████ | 905/1000 [35:28<03:37, 2.29s/it, loss=0.296, lr=1.4e-5] Steps: 90%|█████████ | 905/1000 [35:28<03:37, 2.29s/it, loss=0.609, lr=1.38e-5] Steps: 91%|█████████ | 906/1000 [35:29<03:23, 2.17s/it, loss=0.609, lr=1.38e-5] Steps: 91%|█████████ | 906/1000 [35:29<03:23, 2.17s/it, loss=0.326, lr=1.35e-5] Steps: 91%|█████████ | 907/1000 [35:31<03:13, 2.08s/it, loss=0.326, lr=1.35e-5] Steps: 91%|█████████ | 907/1000 [35:31<03:13, 2.08s/it, loss=0.318, lr=1.32e-5] Steps: 91%|█████████ | 908/1000 [35:33<03:05, 2.02s/it, loss=0.318, lr=1.32e-5] Steps: 91%|█████████ | 908/1000 [35:33<03:05, 2.02s/it, loss=0.327, lr=1.29e-5] Steps: 91%|█████████ | 909/1000 [35:35<02:59, 1.97s/it, loss=0.327, lr=1.29e-5] Steps: 91%|█████████ | 909/1000 [35:35<02:59, 1.97s/it, loss=0.337, lr=1.26e-5] Steps: 91%|█████████ | 910/1000 [35:37<02:54, 1.94s/it, loss=0.337, lr=1.26e-5] Steps: 91%|█████████ | 910/1000 [35:37<02:54, 1.94s/it, loss=0.396, lr=1.24e-5] Steps: 91%|█████████ | 911/1000 [35:39<02:50, 1.92s/it, loss=0.396, lr=1.24e-5] Steps: 91%|█████████ | 911/1000 [35:39<02:50, 1.92s/it, loss=0.422, lr=1.21e-5] Steps: 91%|█████████ | 912/1000 [35:41<02:47, 1.91s/it, loss=0.422, lr=1.21e-5] Steps: 91%|█████████ | 912/1000 [35:41<02:47, 1.91s/it, loss=0.291, lr=1.18e-5] Steps: 91%|█████████▏| 913/1000 [35:43<02:44, 1.89s/it, loss=0.291, lr=1.18e-5] Steps: 91%|█████████▏| 913/1000 [35:43<02:44, 1.89s/it, loss=0.289, lr=1.16e-5] Steps: 91%|█████████▏| 914/1000 [35:44<02:42, 1.89s/it, loss=0.289, lr=1.16e-5] Steps: 91%|█████████▏| 914/1000 [35:44<02:42, 1.89s/it, loss=0.414, lr=1.13e-5] Steps: 92%|█████████▏| 915/1000 [35:46<02:40, 1.88s/it, loss=0.414, lr=1.13e-5] Steps: 92%|█████████▏| 915/1000 [35:46<02:40, 1.88s/it, loss=0.315, lr=1.1e-5] Steps: 92%|█████████▏| 916/1000 [35:48<02:37, 1.88s/it, loss=0.315, lr=1.1e-5] Steps: 92%|█████████▏| 916/1000 [35:48<02:37, 1.88s/it, loss=0.331, lr=1.08e-5] Steps: 92%|█████████▏| 917/1000 [35:50<02:35, 1.88s/it, loss=0.331, lr=1.08e-5] Steps: 92%|█████████▏| 917/1000 [35:50<02:35, 1.88s/it, loss=0.368, lr=1.05e-5] Steps: 92%|█████████▏| 918/1000 [35:52<02:33, 1.88s/it, loss=0.368, lr=1.05e-5] Steps: 92%|█████████▏| 918/1000 [35:52<02:33, 1.88s/it, loss=0.57, lr=1.03e-5] Steps: 92%|█████████▏| 919/1000 [35:54<02:31, 1.88s/it, loss=0.57, lr=1.03e-5] Steps: 92%|█████████▏| 919/1000 [35:54<02:31, 1.88s/it, loss=0.748, lr=1e-5] Steps: 92%|█████████▏| 920/1000 [35:56<02:30, 1.88s/it, loss=0.748, lr=1e-5] Steps: 92%|█████████▏| 920/1000 [35:56<02:30, 1.88s/it, loss=0.379, lr=9.79e-6] Steps: 92%|█████████▏| 921/1000 [35:58<02:27, 1.87s/it, loss=0.379, lr=9.79e-6] Steps: 92%|█████████▏| 921/1000 [35:58<02:27, 1.87s/it, loss=0.331, lr=9.55e-6] Steps: 92%|█████████▏| 922/1000 [35:59<02:26, 1.87s/it, loss=0.331, lr=9.55e-6] Steps: 92%|█████████▏| 922/1000 [35:59<02:26, 1.87s/it, loss=0.358, lr=9.31e-6] Steps: 92%|█████████▏| 923/1000 [36:01<02:24, 1.88s/it, loss=0.358, lr=9.31e-6] Steps: 92%|█████████▏| 923/1000 [36:01<02:24, 1.88s/it, loss=0.816, lr=9.07e-6] Steps: 92%|█████████▏| 924/1000 [36:03<02:23, 1.88s/it, loss=0.816, lr=9.07e-6] Steps: 92%|█████████▏| 924/1000 [36:03<02:23, 1.88s/it, loss=1.03, lr=8.84e-6] Steps: 92%|█████████▎| 925/1000 [36:05<02:21, 1.88s/it, loss=1.03, lr=8.84e-6] Steps: 92%|█████████▎| 925/1000 [36:05<02:21, 1.88s/it, loss=0.466, lr=8.61e-6] Steps: 93%|█████████▎| 926/1000 [36:07<02:19, 1.88s/it, loss=0.466, lr=8.61e-6] Steps: 93%|█████████▎| 926/1000 [36:07<02:19, 1.88s/it, loss=0.505, lr=8.39e-6] Steps: 93%|█████████▎| 927/1000 [36:09<02:16, 1.88s/it, loss=0.505, lr=8.39e-6] Steps: 93%|█████████▎| 927/1000 [36:09<02:16, 1.88s/it, loss=0.736, lr=8.16e-6] Steps: 93%|█████████▎| 928/1000 [36:11<02:14, 1.87s/it, loss=0.736, lr=8.16e-6] Steps: 93%|█████████▎| 928/1000 [36:11<02:14, 1.87s/it, loss=0.274, lr=7.94e-6] Steps: 93%|█████████▎| 929/1000 [36:13<02:12, 1.87s/it, loss=0.274, lr=7.94e-6] Steps: 93%|█████████▎| 929/1000 [36:13<02:12, 1.87s/it, loss=0.434, lr=7.72e-6] Steps: 93%|█████████▎| 930/1000 [36:14<02:10, 1.87s/it, loss=0.434, lr=7.72e-6] Steps: 93%|█████████▎| 930/1000 [36:14<02:10, 1.87s/it, loss=0.471, lr=7.51e-6] Steps: 93%|█████████▎| 931/1000 [36:22<04:14, 3.69s/it, loss=0.471, lr=7.51e-6] Steps: 93%|█████████▎| 931/1000 [36:22<04:14, 3.69s/it, loss=0.424, lr=7.3e-6] Steps: 93%|█████████▎| 932/1000 [36:24<03:33, 3.14s/it, loss=0.424, lr=7.3e-6] Steps: 93%|█████████▎| 932/1000 [36:24<03:33, 3.14s/it, loss=0.324, lr=7.09e-6] Steps: 93%|█████████▎| 933/1000 [36:26<03:04, 2.76s/it, loss=0.324, lr=7.09e-6] Steps: 93%|█████████▎| 933/1000 [36:26<03:04, 2.76s/it, loss=0.827, lr=6.88e-6] Steps: 93%|█████████▎| 934/1000 [36:28<02:44, 2.49s/it, loss=0.827, lr=6.88e-6] Steps: 93%|█████████▎| 934/1000 [36:28<02:44, 2.49s/it, loss=0.567, lr=6.68e-6] Steps: 94%|█████████▎| 935/1000 [36:30<02:29, 2.31s/it, loss=0.567, lr=6.68e-6] Steps: 94%|█████████▎| 935/1000 [36:30<02:29, 2.31s/it, loss=0.363, lr=6.48e-6] Steps: 94%|█████████▎| 936/1000 [36:32<02:19, 2.18s/it, loss=0.363, lr=6.48e-6] Steps: 94%|█████████▎| 936/1000 [36:32<02:19, 2.18s/it, loss=0.556, lr=6.28e-6] Steps: 94%|█████████▎| 937/1000 [36:34<02:11, 2.08s/it, loss=0.556, lr=6.28e-6] Steps: 94%|█████████▎| 937/1000 [36:34<02:11, 2.08s/it, loss=0.445, lr=6.09e-6] Steps: 94%|█████████▍| 938/1000 [36:35<02:05, 2.02s/it, loss=0.445, lr=6.09e-6] Steps: 94%|█████████▍| 938/1000 [36:35<02:05, 2.02s/it, loss=0.685, lr=5.9e-6] Steps: 94%|█████████▍| 939/1000 [36:37<02:00, 1.97s/it, loss=0.685, lr=5.9e-6] Steps: 94%|█████████▍| 939/1000 [36:37<02:00, 1.97s/it, loss=0.334, lr=5.71e-6] Steps: 94%|█████████▍| 940/1000 [36:39<01:56, 1.94s/it, loss=0.334, lr=5.71e-6] Steps: 94%|█████████▍| 940/1000 [36:39<01:56, 1.94s/it, loss=0.332, lr=5.53e-6] Steps: 94%|█████████▍| 941/1000 [36:41<01:53, 1.92s/it, loss=0.332, lr=5.53e-6] Steps: 94%|█████████▍| 941/1000 [36:41<01:53, 1.92s/it, loss=1.02, lr=5.34e-6] Steps: 94%|█████████▍| 942/1000 [36:43<01:50, 1.91s/it, loss=1.02, lr=5.34e-6] Steps: 94%|█████████▍| 942/1000 [36:43<01:50, 1.91s/it, loss=0.346, lr=5.17e-6] Steps: 94%|█████████▍| 943/1000 [36:45<01:48, 1.90s/it, loss=0.346, lr=5.17e-6] Steps: 94%|█████████▍| 943/1000 [36:45<01:48, 1.90s/it, loss=0.341, lr=4.99e-6] Steps: 94%|█████████▍| 944/1000 [36:47<01:45, 1.89s/it, loss=0.341, lr=4.99e-6] Steps: 94%|█████████▍| 944/1000 [36:47<01:45, 1.89s/it, loss=0.681, lr=4.82e-6] Steps: 94%|█████████▍| 945/1000 [36:49<01:43, 1.88s/it, loss=0.681, lr=4.82e-6] Steps: 94%|█████████▍| 945/1000 [36:49<01:43, 1.88s/it, loss=0.317, lr=4.65e-6] Steps: 95%|█████████▍| 946/1000 [36:50<01:41, 1.88s/it, loss=0.317, lr=4.65e-6] Steps: 95%|█████████▍| 946/1000 [36:50<01:41, 1.88s/it, loss=1.03, lr=4.48e-6] Steps: 95%|█████████▍| 947/1000 [36:52<01:39, 1.88s/it, loss=1.03, lr=4.48e-6] Steps: 95%|█████████▍| 947/1000 [36:52<01:39, 1.88s/it, loss=0.624, lr=4.32e-6] Steps: 95%|█████████▍| 948/1000 [36:54<01:37, 1.87s/it, loss=0.624, lr=4.32e-6] Steps: 95%|█████████▍| 948/1000 [36:54<01:37, 1.87s/it, loss=0.504, lr=4.16e-6] Steps: 95%|█████████▍| 949/1000 [36:56<01:35, 1.87s/it, loss=0.504, lr=4.16e-6] Steps: 95%|█████████▍| 949/1000 [36:56<01:35, 1.87s/it, loss=0.628, lr=4e-6] Steps: 95%|█████████▌| 950/1000 [36:58<01:33, 1.87s/it, loss=0.628, lr=4e-6] Steps: 95%|█████████▌| 950/1000 [36:58<01:33, 1.87s/it, loss=0.607, lr=3.84e-6] Steps: 95%|█████████▌| 951/1000 [37:00<01:31, 1.87s/it, loss=0.607, lr=3.84e-6] Steps: 95%|█████████▌| 951/1000 [37:00<01:31, 1.87s/it, loss=0.364, lr=3.69e-6] Steps: 95%|█████████▌| 952/1000 [37:02<01:29, 1.87s/it, loss=0.364, lr=3.69e-6] Steps: 95%|█████████▌| 952/1000 [37:02<01:29, 1.87s/it, loss=0.557, lr=3.54e-6] Steps: 95%|█████████▌| 953/1000 [37:03<01:27, 1.87s/it, loss=0.557, lr=3.54e-6] Steps: 95%|█████████▌| 953/1000 [37:03<01:27, 1.87s/it, loss=0.282, lr=3.4e-6] Steps: 95%|█████████▌| 954/1000 [37:05<01:25, 1.87s/it, loss=0.282, lr=3.4e-6] Steps: 95%|█████████▌| 954/1000 [37:05<01:25, 1.87s/it, loss=0.285, lr=3.25e-6] Steps: 96%|█████████▌| 955/1000 [37:07<01:23, 1.87s/it, loss=0.285, lr=3.25e-6] Steps: 96%|█████████▌| 955/1000 [37:07<01:23, 1.87s/it, loss=0.333, lr=3.11e-6] Steps: 96%|█████████▌| 956/1000 [37:09<01:22, 1.87s/it, loss=0.333, lr=3.11e-6] Steps: 96%|█████████▌| 956/1000 [37:09<01:22, 1.87s/it, loss=0.295, lr=2.98e-6] Steps: 96%|█████████▌| 957/1000 [37:11<01:20, 1.87s/it, loss=0.295, lr=2.98e-6] Steps: 96%|█████████▌| 957/1000 [37:11<01:20, 1.87s/it, loss=0.399, lr=2.84e-6] Steps: 96%|█████████▌| 958/1000 [37:13<01:18, 1.86s/it, loss=0.399, lr=2.84e-6] Steps: 96%|█████████▌| 958/1000 [37:13<01:18, 1.86s/it, loss=0.416, lr=2.71e-6] Steps: 96%|█████████▌| 959/1000 [37:15<01:16, 1.86s/it, loss=0.416, lr=2.71e-6] Steps: 96%|█████████▌| 959/1000 [37:15<01:16, 1.86s/it, loss=0.496, lr=2.59e-6] Steps: 96%|█████████▌| 960/1000 [37:17<01:14, 1.87s/it, loss=0.496, lr=2.59e-6] Steps: 96%|█████████▌| 960/1000 [37:17<01:14, 1.87s/it, loss=0.52, lr=2.46e-6] Steps: 96%|█████████▌| 961/1000 [37:24<02:24, 3.70s/it, loss=0.52, lr=2.46e-6] Steps: 96%|█████████▌| 961/1000 [37:24<02:24, 3.70s/it, loss=0.607, lr=2.34e-6] Steps: 96%|█████████▌| 962/1000 [37:26<01:59, 3.15s/it, loss=0.607, lr=2.34e-6] Steps: 96%|█████████▌| 962/1000 [37:26<01:59, 3.15s/it, loss=0.305, lr=2.22e-6] Steps: 96%|█████████▋| 963/1000 [37:28<01:42, 2.76s/it, loss=0.305, lr=2.22e-6] Steps: 96%|█████████▋| 963/1000 [37:28<01:42, 2.76s/it, loss=0.302, lr=2.11e-6] Steps: 96%|█████████▋| 964/1000 [37:30<01:29, 2.50s/it, loss=0.302, lr=2.11e-6] Steps: 96%|█████████▋| 964/1000 [37:30<01:29, 2.50s/it, loss=0.363, lr=2e-6] Steps: 96%|█████████▋| 965/1000 [37:32<01:20, 2.31s/it, loss=0.363, lr=2e-6] Steps: 96%|█████████▋| 965/1000 [37:32<01:20, 2.31s/it, loss=0.786, lr=1.89e-6] Steps: 97%|█████████▋| 966/1000 [37:34<01:14, 2.18s/it, loss=0.786, lr=1.89e-6] Steps: 97%|█████████▋| 966/1000 [37:34<01:14, 2.18s/it, loss=0.582, lr=1.78e-6] Steps: 97%|█████████▋| 967/1000 [37:36<01:08, 2.09s/it, loss=0.582, lr=1.78e-6] Steps: 97%|█████████▋| 967/1000 [37:36<01:08, 2.09s/it, loss=0.393, lr=1.68e-6] Steps: 97%|█████████▋| 968/1000 [37:38<01:04, 2.02s/it, loss=0.393, lr=1.68e-6] Steps: 97%|█████████▋| 968/1000 [37:38<01:04, 2.02s/it, loss=0.404, lr=1.58e-6] Steps: 97%|█████████▋| 969/1000 [37:39<01:01, 1.98s/it, loss=0.404, lr=1.58e-6] Steps: 97%|█████████▋| 969/1000 [37:39<01:01, 1.98s/it, loss=0.289, lr=1.48e-6] Steps: 97%|█████████▋| 970/1000 [37:41<00:58, 1.95s/it, loss=0.289, lr=1.48e-6] Steps: 97%|█████████▋| 970/1000 [37:41<00:58, 1.95s/it, loss=0.325, lr=1.39e-6] Steps: 97%|█████████▋| 971/1000 [37:43<00:55, 1.92s/it, loss=0.325, lr=1.39e-6] Steps: 97%|█████████▋| 971/1000 [37:43<00:55, 1.92s/it, loss=1.08, lr=1.3e-6] Steps: 97%|█████████▋| 972/1000 [37:45<00:53, 1.91s/it, loss=1.08, lr=1.3e-6] Steps: 97%|█████████▋| 972/1000 [37:45<00:53, 1.91s/it, loss=0.492, lr=1.21e-6] Steps: 97%|█████████▋| 973/1000 [37:47<00:51, 1.90s/it, loss=0.492, lr=1.21e-6] Steps: 97%|█████████▋| 973/1000 [37:47<00:51, 1.90s/it, loss=0.571, lr=1.12e-6] Steps: 97%|█████████▋| 974/1000 [37:49<00:49, 1.89s/it, loss=0.571, lr=1.12e-6] Steps: 97%|█████████▋| 974/1000 [37:49<00:49, 1.89s/it, loss=0.343, lr=1.04e-6] Steps: 98%|█████████▊| 975/1000 [37:51<00:47, 1.89s/it, loss=0.343, lr=1.04e-6] Steps: 98%|█████████▊| 975/1000 [37:51<00:47, 1.89s/it, loss=0.655, lr=9.63e-7] Steps: 98%|█████████▊| 976/1000 [37:53<00:45, 1.88s/it, loss=0.655, lr=9.63e-7] Steps: 98%|█████████▊| 976/1000 [37:53<00:45, 1.88s/it, loss=0.474, lr=8.88e-7] Steps: 98%|█████████▊| 977/1000 [37:54<00:43, 1.88s/it, loss=0.474, lr=8.88e-7] Steps: 98%|█████████▊| 977/1000 [37:54<00:43, 1.88s/it, loss=0.344, lr=8.15e-7] Steps: 98%|█████████▊| 978/1000 [37:56<00:41, 1.88s/it, loss=0.344, lr=8.15e-7] Steps: 98%|█████████▊| 978/1000 [37:56<00:41, 1.88s/it, loss=0.565, lr=7.46e-7] Steps: 98%|█████████▊| 979/1000 [37:58<00:39, 1.88s/it, loss=0.565, lr=7.46e-7] Steps: 98%|█████████▊| 979/1000 [37:58<00:39, 1.88s/it, loss=0.311, lr=6.8e-7] Steps: 98%|█████████▊| 980/1000 [38:00<00:37, 1.87s/it, loss=0.311, lr=6.8e-7] Steps: 98%|█████████▊| 980/1000 [38:00<00:37, 1.87s/it, loss=0.762, lr=6.17e-7] Steps: 98%|█████████▊| 981/1000 [38:02<00:35, 1.87s/it, loss=0.762, lr=6.17e-7] Steps: 98%|█████████▊| 981/1000 [38:02<00:35, 1.87s/it, loss=0.832, lr=5.56e-7] Steps: 98%|█████████▊| 982/1000 [38:04<00:33, 1.87s/it, loss=0.832, lr=5.56e-7] Steps: 98%|█████████▊| 982/1000 [38:04<00:33, 1.87s/it, loss=0.289, lr=4.99e-7] Steps: 98%|█████████▊| 983/1000 [38:06<00:31, 1.87s/it, loss=0.289, lr=4.99e-7] Steps: 98%|█████████▊| 983/1000 [38:06<00:31, 1.87s/it, loss=0.513, lr=4.46e-7] Steps: 98%|█████████▊| 984/1000 [38:08<00:29, 1.87s/it, loss=0.513, lr=4.46e-7] Steps: 98%|█████████▊| 984/1000 [38:08<00:29, 1.87s/it, loss=0.227, lr=3.95e-7] Steps: 98%|█████████▊| 985/1000 [38:09<00:28, 1.87s/it, loss=0.227, lr=3.95e-7] Steps: 98%|█████████▊| 985/1000 [38:09<00:28, 1.87s/it, loss=0.385, lr=3.47e-7] Steps: 99%|█████████▊| 986/1000 [38:11<00:26, 1.87s/it, loss=0.385, lr=3.47e-7] Steps: 99%|█████████▊| 986/1000 [38:11<00:26, 1.87s/it, loss=0.451, lr=3.02e-7] Steps: 99%|█████████▊| 987/1000 [38:13<00:24, 1.87s/it, loss=0.451, lr=3.02e-7] Steps: 99%|█████████▊| 987/1000 [38:13<00:24, 1.87s/it, loss=0.391, lr=2.61e-7] Steps: 99%|█████████▉| 988/1000 [38:15<00:22, 1.88s/it, loss=0.391, lr=2.61e-7] Steps: 99%|█████████▉| 988/1000 [38:15<00:22, 1.88s/it, loss=0.337, lr=2.22e-7] Steps: 99%|█████████▉| 989/1000 [38:17<00:20, 1.88s/it, loss=0.337, lr=2.22e-7] Steps: 99%|█████████▉| 989/1000 [38:17<00:20, 1.88s/it, loss=0.342, lr=1.87e-7] Steps: 99%|█████████▉| 990/1000 [38:19<00:18, 1.87s/it, loss=0.342, lr=1.87e-7] Steps: 99%|█████████▉| 990/1000 [38:19<00:18, 1.87s/it, loss=0.278, lr=1.54e-7] Steps: 99%|█████████▉| 991/1000 [38:26<00:32, 3.62s/it, loss=0.278, lr=1.54e-7] Steps: 99%|█████████▉| 991/1000 [38:26<00:32, 3.62s/it, loss=0.339, lr=1.25e-7] Steps: 99%|█████████▉| 992/1000 [38:28<00:24, 3.09s/it, loss=0.339, lr=1.25e-7] Steps: 99%|█████████▉| 992/1000 [38:28<00:24, 3.09s/it, loss=0.54, lr=9.87e-8] Steps: 99%|█████████▉| 993/1000 [38:30<00:19, 2.73s/it, loss=0.54, lr=9.87e-8] Steps: 99%|█████████▉| 993/1000 [38:30<00:19, 2.73s/it, loss=0.88, lr=7.56e-8] Steps: 99%|█████████▉| 994/1000 [38:32<00:14, 2.47s/it, loss=0.88, lr=7.56e-8] Steps: 99%|█████████▉| 994/1000 [38:32<00:14, 2.47s/it, loss=0.269, lr=5.55e-8] Steps: 100%|█████████▉| 995/1000 [38:34<00:11, 2.29s/it, loss=0.269, lr=5.55e-8] Steps: 100%|█████████▉| 995/1000 [38:34<00:11, 2.29s/it, loss=0.283, lr=3.86e-8] Steps: 100%|█████████▉| 996/1000 [38:36<00:08, 2.16s/it, loss=0.283, lr=3.86e-8] Steps: 100%|█████████▉| 996/1000 [38:36<00:08, 2.16s/it, loss=0.801, lr=2.47e-8] Steps: 100%|█████████▉| 997/1000 [38:38<00:06, 2.08s/it, loss=0.801, lr=2.47e-8] Steps: 100%|█████████▉| 997/1000 [38:38<00:06, 2.08s/it, loss=1, lr=1.39e-8] Steps: 100%|█████████▉| 998/1000 [38:40<00:04, 2.02s/it, loss=1, lr=1.39e-8] Steps: 100%|█████████▉| 998/1000 [38:40<00:04, 2.02s/it, loss=0.874, lr=6.17e-9] Steps: 100%|█████████▉| 999/1000 [38:41<00:01, 1.97s/it, loss=0.874, lr=6.17e-9] Steps: 100%|█████████▉| 999/1000 [38:41<00:01, 1.97s/it, loss=0.505, lr=1.54e-9] Steps: 100%|██████████| 1000/1000 [38:43<00:00, 1.94s/it, loss=0.505, lr=1.54e-9] Steps: 100%|██████████| 1000/1000 [38:43<00:00, 1.94s/it, loss=0.424, lr=0] Steps: 100%|██████████| 1000/1000 [38:47<00:00, 2.33s/it, loss=0.424, lr=0] ---Tar up output directory--- mochi-lora/ mochi-lora/pytorch_lora_weights.safetensors Uploading to Hugging Face: lucataco/mochi-lora-disney HF Repo URL: https://huggingface.co/lucataco/mochi-lora-disney pytorch_lora_weights.safetensors: 0%| | 0.00/76.1M [00:00<?, ?B/s] pytorch_lora_weights.safetensors: 2%|▏ | 1.69M/76.1M [00:00<00:04, 16.8MB/s] pytorch_lora_weights.safetensors: 21%|██ | 16.0M/76.1M [00:00<00:01, 43.1MB/s] pytorch_lora_weights.safetensors: 42%|████▏ | 32.0M/76.1M [00:00<00:00, 54.4MB/s] pytorch_lora_weights.safetensors: 63%|██████▎ | 48.0M/76.1M [00:00<00:00, 61.1MB/s] pytorch_lora_weights.safetensors: 84%|████████▍ | 64.0M/76.1M [00:01<00:00, 61.1MB/s] pytorch_lora_weights.safetensors: 100%|██████████| 76.1M/76.1M [00:01<00:00, 56.8MB/s] Successfully uploaded model to https://huggingface.co/lucataco/mochi-lora-disney
Prediction
genmoai/mochi-1-lora-trainer:170ea99fb48a30fef98cb1c9fb403a2882ab9d60c2ba15ad9383ace33c3fa385Input
- seed
- 42
- steps
- 500
- hf_token
- ████████████████████
This value was redacted after being sent to the model.
- optimizer
- adamw
- batch_size
- 1
- hf_repo_id
- lucataco/mochi-lora-vhs
- compile_dit
- input_videos
- vhs-4.zip
- learning_rate
- 0.0004
- trim_and_crop
- caption_dropout
- 0.1
{ "seed": 42, "steps": 500, "hf_token": "[REDACTED]", "optimizer": "adamw", "batch_size": 1, "hf_repo_id": "lucataco/mochi-lora-vhs", "compile_dit": true, "input_videos": "https://replicate.delivery/pbxt/M7v7tzysZ9DiC0Pdla5rGUDW9tafeTOKzx9iS5fiqhehavkX/vhs-4.zip", "learning_rate": 0.0004, "trim_and_crop": true, "caption_dropout": 0.1 }
Install Replicate’s Node.js client library:npm install replicate
Import and set up the client:import Replicate from "replicate"; const replicate = new Replicate({ auth: process.env.REPLICATE_API_TOKEN, });
Run genmoai/mochi-1-lora-trainer using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "genmoai/mochi-1-lora-trainer:170ea99fb48a30fef98cb1c9fb403a2882ab9d60c2ba15ad9383ace33c3fa385", { input: { seed: 42, steps: 500, hf_token: "[REDACTED]", optimizer: "adamw", batch_size: 1, hf_repo_id: "lucataco/mochi-lora-vhs", compile_dit: true, input_videos: "https://replicate.delivery/pbxt/M7v7tzysZ9DiC0Pdla5rGUDW9tafeTOKzx9iS5fiqhehavkX/vhs-4.zip", learning_rate: 0.0004, trim_and_crop: true, caption_dropout: 0.1 } } ); console.log(output);
To learn more, take a look at the guide on getting started with Node.js.
Install Replicate’s Python client library:pip install replicate
Import the client:import replicate
Run genmoai/mochi-1-lora-trainer using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "genmoai/mochi-1-lora-trainer:170ea99fb48a30fef98cb1c9fb403a2882ab9d60c2ba15ad9383ace33c3fa385", input={ "seed": 42, "steps": 500, "hf_token": "[REDACTED]", "optimizer": "adamw", "batch_size": 1, "hf_repo_id": "lucataco/mochi-lora-vhs", "compile_dit": True, "input_videos": "https://replicate.delivery/pbxt/M7v7tzysZ9DiC0Pdla5rGUDW9tafeTOKzx9iS5fiqhehavkX/vhs-4.zip", "learning_rate": 0.0004, "trim_and_crop": True, "caption_dropout": 0.1 } ) print(output)
To learn more, take a look at the guide on getting started with Python.
Run genmoai/mochi-1-lora-trainer using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
curl -s -X POST \ -H "Authorization: Bearer $REPLICATE_API_TOKEN" \ -H "Content-Type: application/json" \ -H "Prefer: wait" \ -d $'{ "version": "170ea99fb48a30fef98cb1c9fb403a2882ab9d60c2ba15ad9383ace33c3fa385", "input": { "seed": 42, "steps": 500, "hf_token": "[REDACTED]", "optimizer": "adamw", "batch_size": 1, "hf_repo_id": "lucataco/mochi-lora-vhs", "compile_dit": true, "input_videos": "https://replicate.delivery/pbxt/M7v7tzysZ9DiC0Pdla5rGUDW9tafeTOKzx9iS5fiqhehavkX/vhs-4.zip", "learning_rate": 0.0004, "trim_and_crop": true, "caption_dropout": 0.1 } }' \ https://api.replicate.com/v1/predictions
To learn more, take a look at Replicate’s HTTP API reference docs.
Output
{ "completed_at": "2024-12-11T18:27:26.196131Z", "created_at": "2024-12-11T17:55:06.813000Z", "data_removed": false, "error": null, "id": "mxey6b9vqnrm80ckpveaqjqb9w", "input": { "seed": 42, "steps": 500, "hf_token": "[REDACTED]", "optimizer": "adamw", "batch_size": 1, "hf_repo_id": "lucataco/mochi-lora-vhs", "compile_dit": true, "input_videos": "https://replicate.delivery/pbxt/M7v7tzysZ9DiC0Pdla5rGUDW9tafeTOKzx9iS5fiqhehavkX/vhs-4.zip", "learning_rate": 0.0004, "trim_and_crop": true, "caption_dropout": 0.1 }, "logs": "Cleaning up previous runs\nExtracted 8 files from zip to videos_input\n---Starting to Trim input videos---\nProcessing: videos_input/vhs1.mp4\nCopied videos_input/vhs1.txt to videos_prepared/vhs1.txt\nMoviepy - Building video videos_prepared/vhs1.mp4.\nMoviepy - Writing video videos_prepared/vhs1.mp4\n 0%| | 0/4 [00:00<?, ?it/s]\n0%| | 0/4 [00:00<?, ?it/s]\n 0%| | 0/4 [00:00<?, ?it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\n \u001b[A\n0%| | 0/4 [00:00<?, ?it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/vhs1.mp4\n 0%| | 0/4 [00:00<?, ?it/s]\nProcessing: videos_input/vhs2.mp4\nCopied videos_input/vhs2.txt to videos_prepared/vhs2.txt\nMoviepy - Building video videos_prepared/vhs2.mp4.\nMoviepy - Writing video videos_prepared/vhs2.mp4\n 25%|██▌ | 1/4 [00:00<00:00, 3.16it/s]\n25%|██▌ | 1/4 [00:00<00:00, 3.16it/s]\n 25%|██▌ | 1/4 [00:00<00:00, 3.16it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\n \u001b[A\n25%|██▌ | 1/4 [00:00<00:00, 3.16it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/vhs2.mp4\n 25%|██▌ | 1/4 [00:00<00:00, 3.16it/s]\nProcessing: videos_input/vhs3.mp4\nCopied videos_input/vhs3.txt to videos_prepared/vhs3.txt\nMoviepy - Building video videos_prepared/vhs3.mp4.\n 50%|█████ | 2/4 [00:00<00:00, 3.05it/s]\n50%|█████ | 2/4 [00:00<00:00, 3.05it/s]\nMoviepy - Writing video videos_prepared/vhs3.mp4\n 50%|█████ | 2/4 [00:00<00:00, 3.05it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\n \u001b[A\n50%|█████ | 2/4 [00:00<00:00, 3.05it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/vhs3.mp4\n 50%|█████ | 2/4 [00:00<00:00, 3.05it/s]\nProcessing: videos_input/vhs4.mp4\nCopied videos_input/vhs4.txt to videos_prepared/vhs4.txt\nMoviepy - Building video videos_prepared/vhs4.mp4.\nMoviepy - Writing video videos_prepared/vhs4.mp4\n 75%|███████▌ | 3/4 [00:00<00:00, 3.05it/s]\n75%|███████▌ | 3/4 [00:01<00:00, 3.05it/s]\n 75%|███████▌ | 3/4 [00:01<00:00, 3.05it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\n \u001b[A\n75%|███████▌ | 3/4 [00:01<00:00, 3.05it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/vhs4.mp4\n 75%|███████▌ | 3/4 [00:01<00:00, 3.05it/s]\n100%|██████████| 4/4 [00:01<00:00, 3.07it/s]\n100%|██████████| 4/4 [00:01<00:00, 3.07it/s]\n---Starting to Embed videos---\nLoading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s]\nLoading checkpoint shards: 50%|█████ | 1/2 [00:00<00:00, 1.67it/s]\nLoading checkpoint shards: 100%|██████████| 2/2 [00:01<00:00, 1.78it/s]\nLoading checkpoint shards: 100%|██████████| 2/2 [00:01<00:00, 1.76it/s]\nLoading pipeline components...: 0%| | 0/3 [00:00<?, ?it/s]\nLoading pipeline components...: 100%|██████████| 3/3 [00:00<00:00, 681.59it/s]\nProcessing videos_prepared/vhs1.mp4\nTrimmed video from 40 to first 37 frames\n0it [00:00, ?it/s]\nProcessing videos_prepared/vhs2.mp4\nTrimmed video from 40 to first 37 frames\n1it [00:01, 1.38s/it]\nProcessing videos_prepared/vhs3.mp4\nTrimmed video from 40 to first 37 frames\n2it [00:02, 1.15s/it]\nProcessing videos_prepared/vhs4.mp4\nTrimmed video from 40 to first 37 frames\n3it [00:03, 1.07s/it]\n4it [00:04, 1.03s/it]\n4it [00:04, 1.08s/it]\n---Starting training---\nFound 4 training videos in videos_prepared\nLoaded 4/4 valid file pairs.\n===== Memory before training =====\nmemory_allocated=18.780 GB\nmax_memory_allocated=18.780 GB\nmax_memory_reserved=19.250 GB\n***** Running training *****\nNum trainable parameters = 19005440\nNum examples = 4\nNum batches each epoch = 4\nNum epochs = 125\nInstantaneous batch size per device = 1\nTotal train batch size (w. parallel, distributed & accumulation) = 1\nTotal optimization steps = 500\nSteps: 0%| | 0/500 [00:00<?, ?it/s]W1211 17:57:31.075000 135012609543680 torch/fx/experimental/symbolic_shapes.py:4449] [0/0] xindex is not in var_ranges, defaulting to unknown range.\nW1211 17:57:31.089000 135012609543680 torch/fx/experimental/symbolic_shapes.py:4449] [0/0] xindex is not in var_ranges, defaulting to unknown range.\nW1211 17:57:31.224000 135012609543680 torch/fx/experimental/symbolic_shapes.py:4449] [0/0] xindex is not in var_ranges, defaulting to unknown range.\nSteps: 0%| | 1/500 [04:16<35:33:30, 256.53s/it]\nSteps: 0%| | 1/500 [04:16<35:33:30, 256.53s/it, loss=0.933, lr=2e-6]\nSteps: 0%| | 2/500 [04:18<14:45:44, 106.72s/it, loss=0.933, lr=2e-6]\nSteps: 0%| | 2/500 [04:18<14:45:44, 106.72s/it, loss=1.05, lr=4e-6] \nSteps: 1%| | 3/500 [04:20<8:07:22, 58.84s/it, loss=1.05, lr=4e-6] \nSteps: 1%| | 3/500 [04:20<8:07:22, 58.84s/it, loss=0.864, lr=6e-6]\nSteps: 1%| | 4/500 [04:22<5:00:27, 36.34s/it, loss=0.864, lr=6e-6]\nSteps: 1%| | 4/500 [04:22<5:00:27, 36.34s/it, loss=1.06, lr=8e-6] \nSteps: 1%| | 5/500 [04:29<3:34:15, 25.97s/it, loss=1.06, lr=8e-6]\nSteps: 1%| | 5/500 [04:29<3:34:15, 25.97s/it, loss=0.874, lr=1e-5]\nSteps: 1%| | 6/500 [04:31<2:26:21, 17.78s/it, loss=0.874, lr=1e-5]\nSteps: 1%| | 6/500 [04:31<2:26:21, 17.78s/it, loss=1.02, lr=1.2e-5]\nSteps: 1%|▏ | 7/500 [04:33<1:43:19, 12.58s/it, loss=1.02, lr=1.2e-5]\nSteps: 1%|▏ | 7/500 [04:33<1:43:19, 12.58s/it, loss=0.902, lr=1.4e-5]\nSteps: 2%|▏ | 8/500 [04:35<1:15:09, 9.17s/it, loss=0.902, lr=1.4e-5]\nSteps: 2%|▏ | 8/500 [04:35<1:15:09, 9.17s/it, loss=1.08, lr=1.6e-5] \nSteps: 2%|▏ | 9/500 [04:42<1:11:01, 8.68s/it, loss=1.08, lr=1.6e-5]\nSteps: 2%|▏ | 9/500 [04:42<1:11:01, 8.68s/it, loss=1.04, lr=1.8e-5]\nSteps: 2%|▏ | 10/500 [04:44<53:42, 6.58s/it, loss=1.04, lr=1.8e-5] \nSteps: 2%|▏ | 10/500 [04:44<53:42, 6.58s/it, loss=1.09, lr=2e-5] \nSteps: 2%|▏ | 11/500 [04:46<41:50, 5.13s/it, loss=1.09, lr=2e-5]\nSteps: 2%|▏ | 11/500 [04:46<41:50, 5.13s/it, loss=0.886, lr=2.2e-5]\nSteps: 2%|▏ | 12/500 [04:48<33:40, 4.14s/it, loss=0.886, lr=2.2e-5]\nSteps: 2%|▏ | 12/500 [04:48<33:40, 4.14s/it, loss=1.1, lr=2.4e-5] \nSteps: 3%|▎ | 13/500 [04:56<42:08, 5.19s/it, loss=1.1, lr=2.4e-5]\nSteps: 3%|▎ | 13/500 [04:56<42:08, 5.19s/it, loss=0.881, lr=2.6e-5]\nSteps: 3%|▎ | 14/500 [04:57<33:55, 4.19s/it, loss=0.881, lr=2.6e-5]\nSteps: 3%|▎ | 14/500 [04:57<33:55, 4.19s/it, loss=1.07, lr=2.8e-5] \nSteps: 3%|▎ | 15/500 [04:59<28:12, 3.49s/it, loss=1.07, lr=2.8e-5]\nSteps: 3%|▎ | 15/500 [04:59<28:12, 3.49s/it, loss=0.79, lr=3e-5] \nSteps: 3%|▎ | 16/500 [05:01<24:13, 3.00s/it, loss=0.79, lr=3e-5]\nSteps: 3%|▎ | 16/500 [05:01<24:13, 3.00s/it, loss=1.07, lr=3.2e-5]\nSteps: 3%|▎ | 17/500 [05:09<35:16, 4.38s/it, loss=1.07, lr=3.2e-5]\nSteps: 3%|▎ | 17/500 [05:09<35:16, 4.38s/it, loss=0.873, lr=3.4e-5]\nSteps: 4%|▎ | 18/500 [05:11<29:08, 3.63s/it, loss=0.873, lr=3.4e-5]\nSteps: 4%|▎ | 18/500 [05:11<29:08, 3.63s/it, loss=0.968, lr=3.6e-5]\nSteps: 4%|▍ | 19/500 [05:13<24:50, 3.10s/it, loss=0.968, lr=3.6e-5]\nSteps: 4%|▍ | 19/500 [05:13<24:50, 3.10s/it, loss=0.979, lr=3.8e-5]\nSteps: 4%|▍ | 20/500 [05:14<21:50, 2.73s/it, loss=0.979, lr=3.8e-5]\nSteps: 4%|▍ | 20/500 [05:14<21:50, 2.73s/it, loss=1.08, lr=4e-5] \nSteps: 4%|▍ | 21/500 [05:22<33:40, 4.22s/it, loss=1.08, lr=4e-5]\nSteps: 4%|▍ | 21/500 [05:22<33:40, 4.22s/it, loss=0.866, lr=4.2e-5]\nSteps: 4%|▍ | 22/500 [05:24<27:59, 3.51s/it, loss=0.866, lr=4.2e-5]\nSteps: 4%|▍ | 22/500 [05:24<27:59, 3.51s/it, loss=0.966, lr=4.4e-5]\nSteps: 5%|▍ | 23/500 [05:26<24:01, 3.02s/it, loss=0.966, lr=4.4e-5]\nSteps: 5%|▍ | 23/500 [05:26<24:01, 3.02s/it, loss=0.849, lr=4.6e-5]\nSteps: 5%|▍ | 24/500 [05:28<21:13, 2.68s/it, loss=0.849, lr=4.6e-5]\nSteps: 5%|▍ | 24/500 [05:28<21:13, 2.68s/it, loss=1.07, lr=4.8e-5] \nSteps: 5%|▌ | 25/500 [05:35<33:08, 4.19s/it, loss=1.07, lr=4.8e-5]\nSteps: 5%|▌ | 25/500 [05:35<33:08, 4.19s/it, loss=0.853, lr=5e-5] \nSteps: 5%|▌ | 26/500 [05:37<27:34, 3.49s/it, loss=0.853, lr=5e-5]\nSteps: 5%|▌ | 26/500 [05:37<27:34, 3.49s/it, loss=0.996, lr=5.2e-5]\nSteps: 5%|▌ | 27/500 [05:39<23:40, 3.00s/it, loss=0.996, lr=5.2e-5]\nSteps: 5%|▌ | 27/500 [05:39<23:40, 3.00s/it, loss=0.879, lr=5.4e-5]\nSteps: 6%|▌ | 28/500 [05:41<20:56, 2.66s/it, loss=0.879, lr=5.4e-5]\nSteps: 6%|▌ | 28/500 [05:41<20:56, 2.66s/it, loss=0.977, lr=5.6e-5]\nSteps: 6%|▌ | 29/500 [05:49<32:28, 4.14s/it, loss=0.977, lr=5.6e-5]\nSteps: 6%|▌ | 29/500 [05:49<32:28, 4.14s/it, loss=0.881, lr=5.8e-5]\nSteps: 6%|▌ | 30/500 [05:50<27:04, 3.46s/it, loss=0.881, lr=5.8e-5]\nSteps: 6%|▌ | 30/500 [05:50<27:04, 3.46s/it, loss=1.06, lr=6e-5] \nSteps: 6%|▌ | 31/500 [05:52<23:18, 2.98s/it, loss=1.06, lr=6e-5]\nSteps: 6%|▌ | 31/500 [05:52<23:18, 2.98s/it, loss=1.05, lr=6.2e-5]\nSteps: 6%|▋ | 32/500 [05:54<20:39, 2.65s/it, loss=1.05, lr=6.2e-5]\nSteps: 6%|▋ | 32/500 [05:54<20:39, 2.65s/it, loss=0.985, lr=6.4e-5]\nSteps: 7%|▋ | 33/500 [06:02<32:21, 4.16s/it, loss=0.985, lr=6.4e-5]\nSteps: 7%|▋ | 33/500 [06:02<32:21, 4.16s/it, loss=0.871, lr=6.6e-5]\nSteps: 7%|▋ | 34/500 [06:04<26:57, 3.47s/it, loss=0.871, lr=6.6e-5]\nSteps: 7%|▋ | 34/500 [06:04<26:57, 3.47s/it, loss=1.04, lr=6.8e-5] \nSteps: 7%|▋ | 35/500 [06:06<23:10, 2.99s/it, loss=1.04, lr=6.8e-5]\nSteps: 7%|▋ | 35/500 [06:06<23:10, 2.99s/it, loss=0.829, lr=7e-5] \nSteps: 7%|▋ | 36/500 [06:08<20:32, 2.66s/it, loss=0.829, lr=7e-5]\nSteps: 7%|▋ | 36/500 [06:08<20:32, 2.66s/it, loss=0.963, lr=7.2e-5]\nSteps: 7%|▋ | 37/500 [06:15<32:12, 4.17s/it, loss=0.963, lr=7.2e-5]\nSteps: 7%|▋ | 37/500 [06:15<32:12, 4.17s/it, loss=0.878, lr=7.4e-5]\nSteps: 8%|▊ | 38/500 [06:17<26:49, 3.48s/it, loss=0.878, lr=7.4e-5]\nSteps: 8%|▊ | 38/500 [06:17<26:49, 3.48s/it, loss=1.03, lr=7.6e-5] \nSteps: 8%|▊ | 39/500 [06:19<23:02, 3.00s/it, loss=1.03, lr=7.6e-5]\nSteps: 8%|▊ | 39/500 [06:19<23:02, 3.00s/it, loss=0.886, lr=7.8e-5]\nSteps: 8%|▊ | 40/500 [06:21<20:24, 2.66s/it, loss=0.886, lr=7.8e-5]\nSteps: 8%|▊ | 40/500 [06:21<20:24, 2.66s/it, loss=1.06, lr=8e-5] \nSteps: 8%|▊ | 41/500 [06:28<31:38, 4.14s/it, loss=1.06, lr=8e-5]\nSteps: 8%|▊ | 41/500 [06:28<31:38, 4.14s/it, loss=0.874, lr=8.2e-5]\nSteps: 8%|▊ | 42/500 [06:30<26:23, 3.46s/it, loss=0.874, lr=8.2e-5]\nSteps: 8%|▊ | 42/500 [06:30<26:23, 3.46s/it, loss=1.07, lr=8.4e-5] \nSteps: 9%|▊ | 43/500 [06:32<22:43, 2.98s/it, loss=1.07, lr=8.4e-5]\nSteps: 9%|▊ | 43/500 [06:32<22:43, 2.98s/it, loss=0.911, lr=8.6e-5]\nSteps: 9%|▉ | 44/500 [06:34<20:07, 2.65s/it, loss=0.911, lr=8.6e-5]\nSteps: 9%|▉ | 44/500 [06:34<20:07, 2.65s/it, loss=1.05, lr=8.8e-5] \nSteps: 9%|▉ | 45/500 [06:42<31:30, 4.15s/it, loss=1.05, lr=8.8e-5]\nSteps: 9%|▉ | 45/500 [06:42<31:30, 4.15s/it, loss=0.874, lr=9e-5] \nSteps: 9%|▉ | 46/500 [06:44<26:15, 3.47s/it, loss=0.874, lr=9e-5]\nSteps: 9%|▉ | 46/500 [06:44<26:15, 3.47s/it, loss=1.06, lr=9.2e-5]\nSteps: 9%|▉ | 47/500 [06:45<22:34, 2.99s/it, loss=1.06, lr=9.2e-5]\nSteps: 9%|▉ | 47/500 [06:45<22:34, 2.99s/it, loss=0.833, lr=9.4e-5]\nSteps: 10%|▉ | 48/500 [06:47<20:00, 2.66s/it, loss=0.833, lr=9.4e-5]\nSteps: 10%|▉ | 48/500 [06:47<20:00, 2.66s/it, loss=0.973, lr=9.6e-5]\nSteps: 10%|▉ | 49/500 [06:55<31:31, 4.19s/it, loss=0.973, lr=9.6e-5]\nSteps: 10%|▉ | 49/500 [06:55<31:31, 4.19s/it, loss=0.883, lr=9.8e-5]\nSteps: 10%|█ | 50/500 [06:57<26:13, 3.50s/it, loss=0.883, lr=9.8e-5]\nSteps: 10%|█ | 50/500 [06:57<26:13, 3.50s/it, loss=1.08, lr=0.0001] \nSteps: 10%|█ | 51/500 [06:59<22:31, 3.01s/it, loss=1.08, lr=0.0001]\nSteps: 10%|█ | 51/500 [06:59<22:31, 3.01s/it, loss=0.826, lr=0.000102]\nSteps: 10%|█ | 52/500 [07:01<19:56, 2.67s/it, loss=0.826, lr=0.000102]\nSteps: 10%|█ | 52/500 [07:01<19:56, 2.67s/it, loss=0.939, lr=0.000104]\nSteps: 11%|█ | 53/500 [07:11<37:37, 5.05s/it, loss=0.939, lr=0.000104]\nSteps: 11%|█ | 53/500 [07:11<37:37, 5.05s/it, loss=0.789, lr=0.000106]\nSteps: 11%|█ | 54/500 [07:13<30:27, 4.10s/it, loss=0.789, lr=0.000106]\nSteps: 11%|█ | 54/500 [07:13<30:27, 4.10s/it, loss=1.05, lr=0.000108] \nSteps: 11%|█ | 55/500 [07:15<25:25, 3.43s/it, loss=1.05, lr=0.000108]\nSteps: 11%|█ | 55/500 [07:15<25:25, 3.43s/it, loss=1.05, lr=0.00011] \nSteps: 11%|█ | 56/500 [07:17<21:55, 2.96s/it, loss=1.05, lr=0.00011]\nSteps: 11%|█ | 56/500 [07:17<21:55, 2.96s/it, loss=0.958, lr=0.000112]\nSteps: 11%|█▏ | 57/500 [07:24<31:59, 4.33s/it, loss=0.958, lr=0.000112]\nSteps: 11%|█▏ | 57/500 [07:24<31:59, 4.33s/it, loss=0.842, lr=0.000114]\nSteps: 12%|█▏ | 58/500 [07:26<26:28, 3.59s/it, loss=0.842, lr=0.000114]\nSteps: 12%|█▏ | 58/500 [07:26<26:28, 3.59s/it, loss=0.939, lr=0.000116]\nSteps: 12%|█▏ | 59/500 [07:28<22:36, 3.08s/it, loss=0.939, lr=0.000116]\nSteps: 12%|█▏ | 59/500 [07:28<22:36, 3.08s/it, loss=0.882, lr=0.000118]\nSteps: 12%|█▏ | 60/500 [07:30<19:54, 2.71s/it, loss=0.882, lr=0.000118]\nSteps: 12%|█▏ | 60/500 [07:30<19:54, 2.71s/it, loss=0.952, lr=0.00012] \nSteps: 12%|█▏ | 61/500 [07:38<30:27, 4.16s/it, loss=0.952, lr=0.00012]\nSteps: 12%|█▏ | 61/500 [07:38<30:27, 4.16s/it, loss=1.05, lr=0.000122]\nSteps: 12%|█▏ | 62/500 [07:40<25:22, 3.48s/it, loss=1.05, lr=0.000122]\nSteps: 12%|█▏ | 62/500 [07:40<25:22, 3.48s/it, loss=0.985, lr=0.000124]\nSteps: 13%|█▎ | 63/500 [07:41<21:48, 3.00s/it, loss=0.985, lr=0.000124]\nSteps: 13%|█▎ | 63/500 [07:41<21:48, 3.00s/it, loss=0.816, lr=0.000126]\nSteps: 13%|█▎ | 64/500 [07:43<19:18, 2.66s/it, loss=0.816, lr=0.000126]\nSteps: 13%|█▎ | 64/500 [07:43<19:18, 2.66s/it, loss=1.02, lr=0.000128] \nSteps: 13%|█▎ | 65/500 [07:51<30:04, 4.15s/it, loss=1.02, lr=0.000128]\nSteps: 13%|█▎ | 65/500 [07:51<30:04, 4.15s/it, loss=0.855, lr=0.00013]\nSteps: 13%|█▎ | 66/500 [07:53<25:03, 3.47s/it, loss=0.855, lr=0.00013]\nSteps: 13%|█▎ | 66/500 [07:53<25:03, 3.47s/it, loss=0.947, lr=0.000132]\nSteps: 13%|█▎ | 67/500 [07:55<21:33, 2.99s/it, loss=0.947, lr=0.000132]\nSteps: 13%|█▎ | 67/500 [07:55<21:33, 2.99s/it, loss=0.879, lr=0.000134]\nSteps: 14%|█▎ | 68/500 [07:56<19:05, 2.65s/it, loss=0.879, lr=0.000134]\nSteps: 14%|█▎ | 68/500 [07:57<19:05, 2.65s/it, loss=1.06, lr=0.000136] \nSteps: 14%|█▍ | 69/500 [08:04<30:06, 4.19s/it, loss=1.06, lr=0.000136]\nSteps: 14%|█▍ | 69/500 [08:04<30:06, 4.19s/it, loss=0.825, lr=0.000138]\nSteps: 14%|█▍ | 70/500 [08:06<25:02, 3.50s/it, loss=0.825, lr=0.000138]\nSteps: 14%|█▍ | 70/500 [08:06<25:02, 3.50s/it, loss=0.924, lr=0.00014] \nSteps: 14%|█▍ | 71/500 [08:08<21:30, 3.01s/it, loss=0.924, lr=0.00014]\nSteps: 14%|█▍ | 71/500 [08:08<21:30, 3.01s/it, loss=0.794, lr=0.000142]\nSteps: 14%|█▍ | 72/500 [08:10<19:01, 2.67s/it, loss=0.794, lr=0.000142]\nSteps: 14%|█▍ | 72/500 [08:10<19:01, 2.67s/it, loss=0.978, lr=0.000144]\nSteps: 15%|█▍ | 73/500 [08:18<29:45, 4.18s/it, loss=0.978, lr=0.000144]\nSteps: 15%|█▍ | 73/500 [08:18<29:45, 4.18s/it, loss=0.996, lr=0.000146]\nSteps: 15%|█▍ | 74/500 [08:19<24:46, 3.49s/it, loss=0.996, lr=0.000146]\nSteps: 15%|█▍ | 74/500 [08:19<24:46, 3.49s/it, loss=1.07, lr=0.000148] \nSteps: 15%|█▌ | 75/500 [08:21<21:16, 3.00s/it, loss=1.07, lr=0.000148]\nSteps: 15%|█▌ | 75/500 [08:21<21:16, 3.00s/it, loss=0.842, lr=0.00015]\nSteps: 15%|█▌ | 76/500 [08:23<18:50, 2.67s/it, loss=0.842, lr=0.00015]\nSteps: 15%|█▌ | 76/500 [08:23<18:50, 2.67s/it, loss=0.946, lr=0.000152]\nSteps: 15%|█▌ | 77/500 [08:31<29:24, 4.17s/it, loss=0.946, lr=0.000152]\nSteps: 15%|█▌ | 77/500 [08:31<29:24, 4.17s/it, loss=0.838, lr=0.000154]\nSteps: 16%|█▌ | 78/500 [08:33<24:29, 3.48s/it, loss=0.838, lr=0.000154]\nSteps: 16%|█▌ | 78/500 [08:33<24:29, 3.48s/it, loss=1.06, lr=0.000156] \nSteps: 16%|█▌ | 79/500 [08:35<21:02, 3.00s/it, loss=1.06, lr=0.000156]\nSteps: 16%|█▌ | 79/500 [08:35<21:02, 3.00s/it, loss=0.85, lr=0.000158]\nSteps: 16%|█▌ | 80/500 [08:37<18:37, 2.66s/it, loss=0.85, lr=0.000158]\nSteps: 16%|█▌ | 80/500 [08:37<18:37, 2.66s/it, loss=0.923, lr=0.00016]\nSteps: 16%|█▌ | 81/500 [08:44<29:10, 4.18s/it, loss=0.923, lr=0.00016]\nSteps: 16%|█▌ | 81/500 [08:44<29:10, 4.18s/it, loss=0.764, lr=0.000162]\nSteps: 16%|█▋ | 82/500 [08:46<24:17, 3.49s/it, loss=0.764, lr=0.000162]\nSteps: 16%|█▋ | 82/500 [08:46<24:17, 3.49s/it, loss=0.94, lr=0.000164] \nSteps: 17%|█▋ | 83/500 [08:48<20:51, 3.00s/it, loss=0.94, lr=0.000164]\nSteps: 17%|█▋ | 83/500 [08:48<20:51, 3.00s/it, loss=0.828, lr=0.000166]\nSteps: 17%|█▋ | 84/500 [08:50<18:27, 2.66s/it, loss=0.828, lr=0.000166]\nSteps: 17%|█▋ | 84/500 [08:50<18:27, 2.66s/it, loss=1.02, lr=0.000168] \nSteps: 17%|█▋ | 85/500 [08:58<28:59, 4.19s/it, loss=1.02, lr=0.000168]\nSteps: 17%|█▋ | 85/500 [08:58<28:59, 4.19s/it, loss=0.991, lr=0.00017]\nSteps: 17%|█▋ | 86/500 [08:59<24:07, 3.50s/it, loss=0.991, lr=0.00017]\nSteps: 17%|█▋ | 86/500 [09:00<24:07, 3.50s/it, loss=0.975, lr=0.000172]\nSteps: 17%|█▋ | 87/500 [09:01<20:42, 3.01s/it, loss=0.975, lr=0.000172]\nSteps: 17%|█▋ | 87/500 [09:01<20:42, 3.01s/it, loss=0.814, lr=0.000174]\nSteps: 18%|█▊ | 88/500 [09:03<18:19, 2.67s/it, loss=0.814, lr=0.000174]\nSteps: 18%|█▊ | 88/500 [09:03<18:19, 2.67s/it, loss=1.07, lr=0.000176] \nSteps: 18%|█▊ | 89/500 [09:11<28:25, 4.15s/it, loss=1.07, lr=0.000176]\nSteps: 18%|█▊ | 89/500 [09:11<28:25, 4.15s/it, loss=0.859, lr=0.000178]\nSteps: 18%|█▊ | 90/500 [09:13<23:41, 3.47s/it, loss=0.859, lr=0.000178]\nSteps: 18%|█▊ | 90/500 [09:13<23:41, 3.47s/it, loss=1.06, lr=0.00018] \nSteps: 18%|█▊ | 91/500 [09:15<20:21, 2.99s/it, loss=1.06, lr=0.00018]\nSteps: 18%|█▊ | 91/500 [09:15<20:21, 2.99s/it, loss=0.825, lr=0.000182]\nSteps: 18%|█▊ | 92/500 [09:16<18:02, 2.65s/it, loss=0.825, lr=0.000182]\nSteps: 18%|█▊ | 92/500 [09:16<18:02, 2.65s/it, loss=0.954, lr=0.000184]\nSteps: 19%|█▊ | 93/500 [09:24<28:00, 4.13s/it, loss=0.954, lr=0.000184]\nSteps: 19%|█▊ | 93/500 [09:24<28:00, 4.13s/it, loss=0.852, lr=0.000186]\nSteps: 19%|█▉ | 94/500 [09:26<23:21, 3.45s/it, loss=0.852, lr=0.000186]\nSteps: 19%|█▉ | 94/500 [09:26<23:21, 3.45s/it, loss=1.04, lr=0.000188] \nSteps: 19%|█▉ | 95/500 [09:28<20:06, 2.98s/it, loss=1.04, lr=0.000188]\nSteps: 19%|█▉ | 95/500 [09:28<20:06, 2.98s/it, loss=0.847, lr=0.00019]\nSteps: 19%|█▉ | 96/500 [09:30<17:49, 2.65s/it, loss=0.847, lr=0.00019]\nSteps: 19%|█▉ | 96/500 [09:30<17:49, 2.65s/it, loss=0.921, lr=0.000192]\nSteps: 19%|█▉ | 97/500 [09:37<27:56, 4.16s/it, loss=0.921, lr=0.000192]\nSteps: 19%|█▉ | 97/500 [09:37<27:56, 4.16s/it, loss=0.873, lr=0.000194]\nSteps: 20%|█▉ | 98/500 [09:39<23:16, 3.47s/it, loss=0.873, lr=0.000194]\nSteps: 20%|█▉ | 98/500 [09:39<23:16, 3.47s/it, loss=0.977, lr=0.000196]\nSteps: 20%|█▉ | 99/500 [09:41<20:00, 2.99s/it, loss=0.977, lr=0.000196]\nSteps: 20%|█▉ | 99/500 [09:41<20:00, 2.99s/it, loss=0.851, lr=0.000198]\nSteps: 20%|██ | 100/500 [09:43<17:44, 2.66s/it, loss=0.851, lr=0.000198]\nSteps: 20%|██ | 100/500 [09:43<17:44, 2.66s/it, loss=0.918, lr=0.0002] \nSteps: 20%|██ | 101/500 [09:51<28:24, 4.27s/it, loss=0.918, lr=0.0002]\nSteps: 20%|██ | 101/500 [09:51<28:24, 4.27s/it, loss=0.809, lr=0.000202]\nSteps: 20%|██ | 102/500 [09:53<23:33, 3.55s/it, loss=0.809, lr=0.000202]\nSteps: 20%|██ | 102/500 [09:53<23:33, 3.55s/it, loss=0.916, lr=0.000204]\nSteps: 21%|██ | 103/500 [09:55<20:10, 3.05s/it, loss=0.916, lr=0.000204]\nSteps: 21%|██ | 103/500 [09:55<20:10, 3.05s/it, loss=1.01, lr=0.000206] \nSteps: 21%|██ | 104/500 [09:57<17:48, 2.70s/it, loss=1.01, lr=0.000206]\nSteps: 21%|██ | 104/500 [09:57<17:48, 2.70s/it, loss=0.958, lr=0.000208]\nSteps: 21%|██ | 105/500 [10:05<28:03, 4.26s/it, loss=0.958, lr=0.000208]\nSteps: 21%|██ | 105/500 [10:05<28:03, 4.26s/it, loss=0.807, lr=0.00021] \nSteps: 21%|██ | 106/500 [10:06<23:16, 3.55s/it, loss=0.807, lr=0.00021]\nSteps: 21%|██ | 106/500 [10:06<23:16, 3.55s/it, loss=0.953, lr=0.000212]\nSteps: 21%|██▏ | 107/500 [10:08<19:56, 3.04s/it, loss=0.953, lr=0.000212]\nSteps: 21%|██▏ | 107/500 [10:08<19:56, 3.04s/it, loss=0.826, lr=0.000214]\nSteps: 22%|██▏ | 108/500 [10:10<17:35, 2.69s/it, loss=0.826, lr=0.000214]\nSteps: 22%|██▏ | 108/500 [10:10<17:35, 2.69s/it, loss=1.08, lr=0.000216] \nSteps: 22%|██▏ | 109/500 [10:18<27:22, 4.20s/it, loss=1.08, lr=0.000216]\nSteps: 22%|██▏ | 109/500 [10:18<27:22, 4.20s/it, loss=0.836, lr=0.000218]\nSteps: 22%|██▏ | 110/500 [10:20<22:46, 3.50s/it, loss=0.836, lr=0.000218]\nSteps: 22%|██▏ | 110/500 [10:20<22:46, 3.50s/it, loss=1.07, lr=0.00022] \nSteps: 22%|██▏ | 111/500 [10:22<19:32, 3.01s/it, loss=1.07, lr=0.00022]\nSteps: 22%|██▏ | 111/500 [10:22<19:32, 3.01s/it, loss=0.824, lr=0.000222]\nSteps: 22%|██▏ | 112/500 [10:24<17:16, 2.67s/it, loss=0.824, lr=0.000222]\nSteps: 22%|██▏ | 112/500 [10:24<17:16, 2.67s/it, loss=0.916, lr=0.000224]\nSteps: 23%|██▎ | 113/500 [10:31<26:58, 4.18s/it, loss=0.916, lr=0.000224]\nSteps: 23%|██▎ | 113/500 [10:31<26:58, 4.18s/it, loss=0.793, lr=0.000226]\nSteps: 23%|██▎ | 114/500 [10:33<22:27, 3.49s/it, loss=0.793, lr=0.000226]\nSteps: 23%|██▎ | 114/500 [10:33<22:27, 3.49s/it, loss=0.927, lr=0.000228]\nSteps: 23%|██▎ | 115/500 [10:35<19:17, 3.01s/it, loss=0.927, lr=0.000228]\nSteps: 23%|██▎ | 115/500 [10:35<19:17, 3.01s/it, loss=0.924, lr=0.00023] \nSteps: 23%|██▎ | 116/500 [10:37<17:03, 2.67s/it, loss=0.924, lr=0.00023]\nSteps: 23%|██▎ | 116/500 [10:37<17:03, 2.67s/it, loss=1.04, lr=0.000232]\nSteps: 23%|██▎ | 117/500 [10:44<26:32, 4.16s/it, loss=1.04, lr=0.000232]\nSteps: 23%|██▎ | 117/500 [10:44<26:32, 4.16s/it, loss=0.857, lr=0.000234]\nSteps: 24%|██▎ | 118/500 [10:46<22:06, 3.47s/it, loss=0.857, lr=0.000234]\nSteps: 24%|██▎ | 118/500 [10:46<22:06, 3.47s/it, loss=0.91, lr=0.000236] \nSteps: 24%|██▍ | 119/500 [10:48<19:00, 2.99s/it, loss=0.91, lr=0.000236]\nSteps: 24%|██▍ | 119/500 [10:48<19:00, 2.99s/it, loss=0.781, lr=0.000238]\nSteps: 24%|██▍ | 120/500 [10:50<16:49, 2.66s/it, loss=0.781, lr=0.000238]\nSteps: 24%|██▍ | 120/500 [10:50<16:49, 2.66s/it, loss=0.937, lr=0.00024] \nSteps: 24%|██▍ | 121/500 [10:58<26:42, 4.23s/it, loss=0.937, lr=0.00024]\nSteps: 24%|██▍ | 121/500 [10:58<26:42, 4.23s/it, loss=0.876, lr=0.000242]\nSteps: 24%|██▍ | 122/500 [11:00<22:10, 3.52s/it, loss=0.876, lr=0.000242]\nSteps: 24%|██▍ | 122/500 [11:00<22:10, 3.52s/it, loss=0.971, lr=0.000244]\nSteps: 25%|██▍ | 123/500 [11:02<19:00, 3.03s/it, loss=0.971, lr=0.000244]\nSteps: 25%|██▍ | 123/500 [11:02<19:00, 3.03s/it, loss=0.812, lr=0.000246]\nSteps: 25%|██▍ | 124/500 [11:04<16:47, 2.68s/it, loss=0.812, lr=0.000246]\nSteps: 25%|██▍ | 124/500 [11:04<16:47, 2.68s/it, loss=1, lr=0.000248] \nSteps: 25%|██▌ | 125/500 [11:11<26:07, 4.18s/it, loss=1, lr=0.000248]\nSteps: 25%|██▌ | 125/500 [11:11<26:07, 4.18s/it, loss=0.97, lr=0.00025]\nSteps: 25%|██▌ | 126/500 [11:13<21:44, 3.49s/it, loss=0.97, lr=0.00025]\nSteps: 25%|██▌ | 126/500 [11:13<21:44, 3.49s/it, loss=1.07, lr=0.000252]\nSteps: 25%|██▌ | 127/500 [11:15<18:40, 3.00s/it, loss=1.07, lr=0.000252]\nSteps: 25%|██▌ | 127/500 [11:15<18:40, 3.00s/it, loss=0.814, lr=0.000254]\nSteps: 26%|██▌ | 128/500 [11:17<16:31, 2.67s/it, loss=0.814, lr=0.000254]\nSteps: 26%|██▌ | 128/500 [11:17<16:31, 2.67s/it, loss=0.904, lr=0.000256]\nSteps: 26%|██▌ | 129/500 [11:25<25:55, 4.19s/it, loss=0.904, lr=0.000256]\nSteps: 26%|██▌ | 129/500 [11:25<25:55, 4.19s/it, loss=0.885, lr=0.000258]\nSteps: 26%|██▌ | 130/500 [11:27<21:33, 3.50s/it, loss=0.885, lr=0.000258]\nSteps: 26%|██▌ | 130/500 [11:27<21:33, 3.50s/it, loss=0.923, lr=0.00026] \nSteps: 26%|██▌ | 131/500 [11:28<18:30, 3.01s/it, loss=0.923, lr=0.00026]\nSteps: 26%|██▌ | 131/500 [11:28<18:30, 3.01s/it, loss=0.812, lr=0.000262]\nSteps: 26%|██▋ | 132/500 [11:30<16:21, 2.67s/it, loss=0.812, lr=0.000262]\nSteps: 26%|██▋ | 132/500 [11:30<16:21, 2.67s/it, loss=0.986, lr=0.000264]\nSteps: 27%|██▋ | 133/500 [11:38<25:53, 4.23s/it, loss=0.986, lr=0.000264]\nSteps: 27%|██▋ | 133/500 [11:38<25:53, 4.23s/it, loss=0.823, lr=0.000266]\nSteps: 27%|██▋ | 134/500 [11:40<21:31, 3.53s/it, loss=0.823, lr=0.000266]\nSteps: 27%|██▋ | 134/500 [11:40<21:31, 3.53s/it, loss=1.06, lr=0.000268] \nSteps: 27%|██▋ | 135/500 [11:42<18:26, 3.03s/it, loss=1.06, lr=0.000268]\nSteps: 27%|██▋ | 135/500 [11:42<18:26, 3.03s/it, loss=1.07, lr=0.00027] \nSteps: 27%|██▋ | 136/500 [11:44<16:17, 2.69s/it, loss=1.07, lr=0.00027]\nSteps: 27%|██▋ | 136/500 [11:44<16:17, 2.69s/it, loss=0.961, lr=0.000272]\nSteps: 27%|██▋ | 137/500 [11:51<25:19, 4.19s/it, loss=0.961, lr=0.000272]\nSteps: 27%|██▋ | 137/500 [11:52<25:19, 4.19s/it, loss=0.842, lr=0.000274]\nSteps: 28%|██▊ | 138/500 [11:53<21:04, 3.49s/it, loss=0.842, lr=0.000274]\nSteps: 28%|██▊ | 138/500 [11:53<21:04, 3.49s/it, loss=0.952, lr=0.000276]\nSteps: 28%|██▊ | 139/500 [11:55<18:05, 3.01s/it, loss=0.952, lr=0.000276]\nSteps: 28%|██▊ | 139/500 [11:55<18:05, 3.01s/it, loss=0.901, lr=0.000278]\nSteps: 28%|██▊ | 140/500 [11:57<16:00, 2.67s/it, loss=0.901, lr=0.000278]\nSteps: 28%|██▊ | 140/500 [11:57<16:00, 2.67s/it, loss=0.926, lr=0.00028] \nSteps: 28%|██▊ | 141/500 [12:05<25:06, 4.20s/it, loss=0.926, lr=0.00028]\nSteps: 28%|██▊ | 141/500 [12:05<25:06, 4.20s/it, loss=0.808, lr=0.000282]\nSteps: 28%|██▊ | 142/500 [12:07<20:52, 3.50s/it, loss=0.808, lr=0.000282]\nSteps: 28%|██▊ | 142/500 [12:07<20:52, 3.50s/it, loss=0.926, lr=0.000284]\nSteps: 29%|██▊ | 143/500 [12:09<17:55, 3.01s/it, loss=0.926, lr=0.000284]\nSteps: 29%|██▊ | 143/500 [12:09<17:55, 3.01s/it, loss=0.951, lr=0.000286]\nSteps: 29%|██▉ | 144/500 [12:11<15:51, 2.67s/it, loss=0.951, lr=0.000286]\nSteps: 29%|██▉ | 144/500 [12:11<15:51, 2.67s/it, loss=0.911, lr=0.000288]\nSteps: 29%|██▉ | 145/500 [12:18<24:43, 4.18s/it, loss=0.911, lr=0.000288]\nSteps: 29%|██▉ | 145/500 [12:18<24:43, 4.18s/it, loss=0.806, lr=0.00029] \nSteps: 29%|██▉ | 146/500 [12:20<20:34, 3.49s/it, loss=0.806, lr=0.00029]\nSteps: 29%|██▉ | 146/500 [12:20<20:34, 3.49s/it, loss=0.901, lr=0.000292]\nSteps: 29%|██▉ | 147/500 [12:22<17:40, 3.00s/it, loss=0.901, lr=0.000292]\nSteps: 29%|██▉ | 147/500 [12:22<17:40, 3.00s/it, loss=0.847, lr=0.000294]\nSteps: 30%|██▉ | 148/500 [12:24<15:38, 2.67s/it, loss=0.847, lr=0.000294]\nSteps: 30%|██▉ | 148/500 [12:24<15:38, 2.67s/it, loss=0.963, lr=0.000296]\nSteps: 30%|██▉ | 149/500 [12:32<24:33, 4.20s/it, loss=0.963, lr=0.000296]\nSteps: 30%|██▉ | 149/500 [12:32<24:33, 4.20s/it, loss=1, lr=0.000298] \nSteps: 30%|███ | 150/500 [12:33<20:24, 3.50s/it, loss=1, lr=0.000298]\nSteps: 30%|███ | 150/500 [12:33<20:24, 3.50s/it, loss=0.897, lr=0.0003]\nSteps: 30%|███ | 151/500 [12:35<17:30, 3.01s/it, loss=0.897, lr=0.0003]\nSteps: 30%|███ | 151/500 [12:35<17:30, 3.01s/it, loss=0.842, lr=0.000302]\nSteps: 30%|███ | 152/500 [12:37<15:28, 2.67s/it, loss=0.842, lr=0.000302]\nSteps: 30%|███ | 152/500 [12:37<15:28, 2.67s/it, loss=1.07, lr=0.000304] \nSteps: 31%|███ | 153/500 [12:45<24:13, 4.19s/it, loss=1.07, lr=0.000304]\nSteps: 31%|███ | 153/500 [12:45<24:13, 4.19s/it, loss=0.861, lr=0.000306]\nSteps: 31%|███ | 154/500 [12:47<20:09, 3.49s/it, loss=0.861, lr=0.000306]\nSteps: 31%|███ | 154/500 [12:47<20:09, 3.49s/it, loss=0.903, lr=0.000308]\nSteps: 31%|███ | 155/500 [12:49<17:17, 3.01s/it, loss=0.903, lr=0.000308]\nSteps: 31%|███ | 155/500 [12:49<17:17, 3.01s/it, loss=0.86, lr=0.00031] \nSteps: 31%|███ | 156/500 [12:51<15:18, 2.67s/it, loss=0.86, lr=0.00031]\nSteps: 31%|███ | 156/500 [12:51<15:18, 2.67s/it, loss=0.904, lr=0.000312]\nSteps: 31%|███▏ | 157/500 [12:58<24:02, 4.21s/it, loss=0.904, lr=0.000312]\nSteps: 31%|███▏ | 157/500 [12:58<24:02, 4.21s/it, loss=1.05, lr=0.000314] \nSteps: 32%|███▏ | 158/500 [13:00<19:58, 3.51s/it, loss=1.05, lr=0.000314]\nSteps: 32%|███▏ | 158/500 [13:00<19:58, 3.51s/it, loss=1.02, lr=0.000316]\nSteps: 32%|███▏ | 159/500 [13:02<17:08, 3.02s/it, loss=1.02, lr=0.000316]\nSteps: 32%|███▏ | 159/500 [13:02<17:08, 3.02s/it, loss=0.964, lr=0.000318]\nSteps: 32%|███▏ | 160/500 [13:04<15:08, 2.67s/it, loss=0.964, lr=0.000318]\nSteps: 32%|███▏ | 160/500 [13:04<15:08, 2.67s/it, loss=0.909, lr=0.00032] \nSteps: 32%|███▏ | 161/500 [13:12<23:30, 4.16s/it, loss=0.909, lr=0.00032]\nSteps: 32%|███▏ | 161/500 [13:12<23:30, 4.16s/it, loss=0.874, lr=0.000322]\nSteps: 32%|███▏ | 162/500 [13:13<19:34, 3.47s/it, loss=0.874, lr=0.000322]\nSteps: 32%|███▏ | 162/500 [13:14<19:34, 3.47s/it, loss=0.932, lr=0.000324]\nSteps: 33%|███▎ | 163/500 [13:15<16:49, 2.99s/it, loss=0.932, lr=0.000324]\nSteps: 33%|███▎ | 163/500 [13:15<16:49, 2.99s/it, loss=0.917, lr=0.000326]\nSteps: 33%|███▎ | 164/500 [13:17<14:53, 2.66s/it, loss=0.917, lr=0.000326]\nSteps: 33%|███▎ | 164/500 [13:17<14:53, 2.66s/it, loss=1.07, lr=0.000328] \nSteps: 33%|███▎ | 165/500 [13:25<23:19, 4.18s/it, loss=1.07, lr=0.000328]\nSteps: 33%|███▎ | 165/500 [13:25<23:19, 4.18s/it, loss=0.855, lr=0.00033]\nSteps: 33%|███▎ | 166/500 [13:27<19:24, 3.49s/it, loss=0.855, lr=0.00033]\nSteps: 33%|███▎ | 166/500 [13:27<19:24, 3.49s/it, loss=0.986, lr=0.000332]\nSteps: 33%|███▎ | 167/500 [13:29<16:39, 3.00s/it, loss=0.986, lr=0.000332]\nSteps: 33%|███▎ | 167/500 [13:29<16:39, 3.00s/it, loss=0.814, lr=0.000334]\nSteps: 34%|███▎ | 168/500 [13:31<14:44, 2.66s/it, loss=0.814, lr=0.000334]\nSteps: 34%|███▎ | 168/500 [13:31<14:44, 2.66s/it, loss=0.92, lr=0.000336] \nSteps: 34%|███▍ | 169/500 [13:38<23:13, 4.21s/it, loss=0.92, lr=0.000336]\nSteps: 34%|███▍ | 169/500 [13:38<23:13, 4.21s/it, loss=0.835, lr=0.000338]\nSteps: 34%|███▍ | 170/500 [13:40<19:17, 3.51s/it, loss=0.835, lr=0.000338]\nSteps: 34%|███▍ | 170/500 [13:40<19:17, 3.51s/it, loss=1.08, lr=0.00034] \nSteps: 34%|███▍ | 171/500 [13:42<16:33, 3.02s/it, loss=1.08, lr=0.00034]\nSteps: 34%|███▍ | 171/500 [13:42<16:33, 3.02s/it, loss=0.988, lr=0.000342]\nSteps: 34%|███▍ | 172/500 [13:44<14:37, 2.68s/it, loss=0.988, lr=0.000342]\nSteps: 34%|███▍ | 172/500 [13:44<14:37, 2.68s/it, loss=1, lr=0.000344] \nSteps: 35%|███▍ | 173/500 [13:52<22:40, 4.16s/it, loss=1, lr=0.000344]\nSteps: 35%|███▍ | 173/500 [13:52<22:40, 4.16s/it, loss=1.04, lr=0.000346]\nSteps: 35%|███▍ | 174/500 [13:54<18:52, 3.47s/it, loss=1.04, lr=0.000346]\nSteps: 35%|███▍ | 174/500 [13:54<18:52, 3.47s/it, loss=1.05, lr=0.000348]\nSteps: 35%|███▌ | 175/500 [13:55<16:13, 2.99s/it, loss=1.05, lr=0.000348]\nSteps: 35%|███▌ | 175/500 [13:55<16:13, 2.99s/it, loss=0.996, lr=0.00035]\nSteps: 35%|███▌ | 176/500 [13:57<14:20, 2.66s/it, loss=0.996, lr=0.00035]\nSteps: 35%|███▌ | 176/500 [13:57<14:20, 2.66s/it, loss=1.06, lr=0.000352]\nSteps: 35%|███▌ | 177/500 [14:05<22:27, 4.17s/it, loss=1.06, lr=0.000352]\nSteps: 35%|███▌ | 177/500 [14:05<22:27, 4.17s/it, loss=0.994, lr=0.000354]\nSteps: 36%|███▌ | 178/500 [14:07<18:41, 3.48s/it, loss=0.994, lr=0.000354]\nSteps: 36%|███▌ | 178/500 [14:07<18:41, 3.48s/it, loss=0.987, lr=0.000356]\nSteps: 36%|███▌ | 179/500 [14:09<16:02, 3.00s/it, loss=0.987, lr=0.000356]\nSteps: 36%|███▌ | 179/500 [14:09<16:02, 3.00s/it, loss=0.81, lr=0.000358] \nSteps: 36%|███▌ | 180/500 [14:11<14:11, 2.66s/it, loss=0.81, lr=0.000358]\nSteps: 36%|███▌ | 180/500 [14:11<14:11, 2.66s/it, loss=0.944, lr=0.00036]\nSteps: 36%|███▌ | 181/500 [14:18<22:05, 4.16s/it, loss=0.944, lr=0.00036]\nSteps: 36%|███▌ | 181/500 [14:18<22:05, 4.16s/it, loss=0.856, lr=0.000362]\nSteps: 36%|███▋ | 182/500 [14:20<18:23, 3.47s/it, loss=0.856, lr=0.000362]\nSteps: 36%|███▋ | 182/500 [14:20<18:23, 3.47s/it, loss=0.956, lr=0.000364]\nSteps: 37%|███▋ | 183/500 [14:22<15:48, 2.99s/it, loss=0.956, lr=0.000364]\nSteps: 37%|███▋ | 183/500 [14:22<15:48, 2.99s/it, loss=0.823, lr=0.000366]\nSteps: 37%|███▋ | 184/500 [14:24<13:59, 2.66s/it, loss=0.823, lr=0.000366]\nSteps: 37%|███▋ | 184/500 [14:24<13:59, 2.66s/it, loss=0.963, lr=0.000368]\nSteps: 37%|███▋ | 185/500 [14:31<21:45, 4.15s/it, loss=0.963, lr=0.000368]\nSteps: 37%|███▋ | 185/500 [14:31<21:45, 4.15s/it, loss=0.971, lr=0.00037] \nSteps: 37%|███▋ | 186/500 [14:33<18:07, 3.46s/it, loss=0.971, lr=0.00037]\nSteps: 37%|███▋ | 186/500 [14:33<18:07, 3.46s/it, loss=1.01, lr=0.000372]\nSteps: 37%|███▋ | 187/500 [14:35<15:34, 2.99s/it, loss=1.01, lr=0.000372]\nSteps: 37%|███▋ | 187/500 [14:35<15:34, 2.99s/it, loss=0.855, lr=0.000374]\nSteps: 38%|███▊ | 188/500 [14:37<13:47, 2.65s/it, loss=0.855, lr=0.000374]\nSteps: 38%|███▊ | 188/500 [14:37<13:47, 2.65s/it, loss=1.06, lr=0.000376] \nSteps: 38%|███▊ | 189/500 [14:45<21:33, 4.16s/it, loss=1.06, lr=0.000376]\nSteps: 38%|███▊ | 189/500 [14:45<21:33, 4.16s/it, loss=0.906, lr=0.000378]\nSteps: 38%|███▊ | 190/500 [14:47<17:56, 3.47s/it, loss=0.906, lr=0.000378]\nSteps: 38%|███▊ | 190/500 [14:47<17:56, 3.47s/it, loss=0.957, lr=0.00038] \nSteps: 38%|███▊ | 191/500 [14:49<15:24, 2.99s/it, loss=0.957, lr=0.00038]\nSteps: 38%|███▊ | 191/500 [14:49<15:24, 2.99s/it, loss=0.874, lr=0.000382]\nSteps: 38%|███▊ | 192/500 [14:50<13:38, 2.66s/it, loss=0.874, lr=0.000382]\nSteps: 38%|███▊ | 192/500 [14:50<13:38, 2.66s/it, loss=0.902, lr=0.000384]\nSteps: 39%|███▊ | 193/500 [14:58<21:24, 4.19s/it, loss=0.902, lr=0.000384]\nSteps: 39%|███▊ | 193/500 [14:58<21:24, 4.19s/it, loss=1.06, lr=0.000386] \nSteps: 39%|███▉ | 194/500 [15:00<17:48, 3.49s/it, loss=1.06, lr=0.000386]\nSteps: 39%|███▉ | 194/500 [15:00<17:48, 3.49s/it, loss=0.955, lr=0.000388]\nSteps: 39%|███▉ | 195/500 [15:02<15:16, 3.01s/it, loss=0.955, lr=0.000388]\nSteps: 39%|███▉ | 195/500 [15:02<15:16, 3.01s/it, loss=0.808, lr=0.00039] \nSteps: 39%|███▉ | 196/500 [15:04<13:30, 2.66s/it, loss=0.808, lr=0.00039]\nSteps: 39%|███▉ | 196/500 [15:04<13:30, 2.66s/it, loss=0.925, lr=0.000392]\nSteps: 39%|███▉ | 197/500 [15:12<21:19, 4.22s/it, loss=0.925, lr=0.000392]\nSteps: 39%|███▉ | 197/500 [15:12<21:19, 4.22s/it, loss=0.869, lr=0.000394]\nSteps: 40%|███▉ | 198/500 [15:13<17:42, 3.52s/it, loss=0.869, lr=0.000394]\nSteps: 40%|███▉ | 198/500 [15:13<17:42, 3.52s/it, loss=1.08, lr=0.000396] \nSteps: 40%|███▉ | 199/500 [15:15<15:10, 3.02s/it, loss=1.08, lr=0.000396]\nSteps: 40%|███▉ | 199/500 [15:15<15:10, 3.02s/it, loss=0.829, lr=0.000398]\nSteps: 40%|████ | 200/500 [15:17<13:23, 2.68s/it, loss=0.829, lr=0.000398]\nSteps: 40%|████ | 200/500 [15:17<13:23, 2.68s/it, loss=1.05, lr=0.0004] \nSteps: 40%|████ | 201/500 [15:25<20:55, 4.20s/it, loss=1.05, lr=0.0004]\nSteps: 40%|████ | 201/500 [15:25<20:55, 4.20s/it, loss=1.03, lr=0.0004]\nSteps: 40%|████ | 202/500 [15:27<17:23, 3.50s/it, loss=1.03, lr=0.0004]\nSteps: 40%|████ | 202/500 [15:27<17:23, 3.50s/it, loss=1.06, lr=0.0004]\nSteps: 41%|████ | 203/500 [15:29<14:54, 3.01s/it, loss=1.06, lr=0.0004]\nSteps: 41%|████ | 203/500 [15:29<14:54, 3.01s/it, loss=0.846, lr=0.0004]\nSteps: 41%|████ | 204/500 [15:31<13:10, 2.67s/it, loss=0.846, lr=0.0004]\nSteps: 41%|████ | 204/500 [15:31<13:10, 2.67s/it, loss=0.921, lr=0.0004]\nSteps: 41%|████ | 205/500 [15:38<20:37, 4.19s/it, loss=0.921, lr=0.0004]\nSteps: 41%|████ | 205/500 [15:38<20:37, 4.19s/it, loss=0.856, lr=0.0004]\nSteps: 41%|████ | 206/500 [15:40<17:08, 3.50s/it, loss=0.856, lr=0.0004]\nSteps: 41%|████ | 206/500 [15:40<17:08, 3.50s/it, loss=1.06, lr=0.0004] \nSteps: 41%|████▏ | 207/500 [15:42<14:41, 3.01s/it, loss=1.06, lr=0.0004]\nSteps: 41%|████▏ | 207/500 [15:42<14:41, 3.01s/it, loss=0.81, lr=0.000399]\nSteps: 42%|████▏ | 208/500 [15:44<12:59, 2.67s/it, loss=0.81, lr=0.000399]\nSteps: 42%|████▏ | 208/500 [15:44<12:59, 2.67s/it, loss=0.961, lr=0.000399]\nSteps: 42%|████▏ | 209/500 [15:52<20:15, 4.18s/it, loss=0.961, lr=0.000399]\nSteps: 42%|████▏ | 209/500 [15:52<20:15, 4.18s/it, loss=0.809, lr=0.000399]\nSteps: 42%|████▏ | 210/500 [15:54<16:50, 3.48s/it, loss=0.809, lr=0.000399]\nSteps: 42%|████▏ | 210/500 [15:54<16:50, 3.48s/it, loss=0.983, lr=0.000399]\nSteps: 42%|████▏ | 211/500 [15:55<14:27, 3.00s/it, loss=0.983, lr=0.000399]\nSteps: 42%|████▏ | 211/500 [15:55<14:27, 3.00s/it, loss=0.865, lr=0.000399]\nSteps: 42%|████▏ | 212/500 [15:57<12:46, 2.66s/it, loss=0.865, lr=0.000399]\nSteps: 42%|████▏ | 212/500 [15:57<12:46, 2.66s/it, loss=0.927, lr=0.000398]\nSteps: 43%|████▎ | 213/500 [16:05<19:55, 4.17s/it, loss=0.927, lr=0.000398]\nSteps: 43%|████▎ | 213/500 [16:05<19:55, 4.17s/it, loss=0.799, lr=0.000398]\nSteps: 43%|████▎ | 214/500 [16:07<16:34, 3.48s/it, loss=0.799, lr=0.000398]\nSteps: 43%|████▎ | 214/500 [16:07<16:34, 3.48s/it, loss=1.02, lr=0.000398] \nSteps: 43%|████▎ | 215/500 [16:09<14:13, 3.00s/it, loss=1.02, lr=0.000398]\nSteps: 43%|████▎ | 215/500 [16:09<14:13, 3.00s/it, loss=0.864, lr=0.000398]\nSteps: 43%|████▎ | 216/500 [16:11<12:34, 2.66s/it, loss=0.864, lr=0.000398]\nSteps: 43%|████▎ | 216/500 [16:11<12:34, 2.66s/it, loss=0.976, lr=0.000397]\nSteps: 43%|████▎ | 217/500 [16:18<19:35, 4.15s/it, loss=0.976, lr=0.000397]\nSteps: 43%|████▎ | 217/500 [16:18<19:35, 4.15s/it, loss=0.859, lr=0.000397]\nSteps: 44%|████▎ | 218/500 [16:20<16:18, 3.47s/it, loss=0.859, lr=0.000397]\nSteps: 44%|████▎ | 218/500 [16:20<16:18, 3.47s/it, loss=0.9, lr=0.000396] \nSteps: 44%|████▍ | 219/500 [16:22<14:00, 2.99s/it, loss=0.9, lr=0.000396]\nSteps: 44%|████▍ | 219/500 [16:22<14:00, 2.99s/it, loss=0.935, lr=0.000396]\nSteps: 44%|████▍ | 220/500 [16:24<12:23, 2.66s/it, loss=0.935, lr=0.000396]\nSteps: 44%|████▍ | 220/500 [16:24<12:23, 2.66s/it, loss=0.919, lr=0.000396]\nSteps: 44%|████▍ | 221/500 [16:31<19:15, 4.14s/it, loss=0.919, lr=0.000396]\nSteps: 44%|████▍ | 221/500 [16:31<19:15, 4.14s/it, loss=0.849, lr=0.000395]\nSteps: 44%|████▍ | 222/500 [16:33<16:01, 3.46s/it, loss=0.849, lr=0.000395]\nSteps: 44%|████▍ | 222/500 [16:33<16:01, 3.46s/it, loss=0.985, lr=0.000395]\nSteps: 45%|████▍ | 223/500 [16:35<13:46, 2.98s/it, loss=0.985, lr=0.000395]\nSteps: 45%|████▍ | 223/500 [16:35<13:46, 2.98s/it, loss=0.798, lr=0.000394]\nSteps: 45%|████▍ | 224/500 [16:37<12:11, 2.65s/it, loss=0.798, lr=0.000394]\nSteps: 45%|████▍ | 224/500 [16:37<12:11, 2.65s/it, loss=0.896, lr=0.000394]\nSteps: 45%|████▌ | 225/500 [16:45<19:01, 4.15s/it, loss=0.896, lr=0.000394]\nSteps: 45%|████▌ | 225/500 [16:45<19:01, 4.15s/it, loss=0.772, lr=0.000393]\nSteps: 45%|████▌ | 226/500 [16:47<15:50, 3.47s/it, loss=0.772, lr=0.000393]\nSteps: 45%|████▌ | 226/500 [16:47<15:50, 3.47s/it, loss=0.968, lr=0.000393]\nSteps: 45%|████▌ | 227/500 [16:48<13:36, 2.99s/it, loss=0.968, lr=0.000393]\nSteps: 45%|████▌ | 227/500 [16:48<13:36, 2.99s/it, loss=0.943, lr=0.000392]\nSteps: 46%|████▌ | 228/500 [16:50<12:01, 2.65s/it, loss=0.943, lr=0.000392]\nSteps: 46%|████▌ | 228/500 [16:50<12:01, 2.65s/it, loss=0.951, lr=0.000391]\nSteps: 46%|████▌ | 229/500 [16:58<18:52, 4.18s/it, loss=0.951, lr=0.000391]\nSteps: 46%|████▌ | 229/500 [16:58<18:52, 4.18s/it, loss=0.839, lr=0.000391]\nSteps: 46%|████▌ | 230/500 [17:00<15:41, 3.49s/it, loss=0.839, lr=0.000391]\nSteps: 46%|████▌ | 230/500 [17:00<15:41, 3.49s/it, loss=1.02, lr=0.00039] \nSteps: 46%|████▌ | 231/500 [17:02<13:27, 3.00s/it, loss=1.02, lr=0.00039]\nSteps: 46%|████▌ | 231/500 [17:02<13:27, 3.00s/it, loss=0.854, lr=0.00039]\nSteps: 46%|████▋ | 232/500 [17:04<11:53, 2.66s/it, loss=0.854, lr=0.00039]\nSteps: 46%|████▋ | 232/500 [17:04<11:53, 2.66s/it, loss=0.958, lr=0.000389]\nSteps: 47%|████▋ | 233/500 [17:11<18:44, 4.21s/it, loss=0.958, lr=0.000389]\nSteps: 47%|████▋ | 233/500 [17:11<18:44, 4.21s/it, loss=1.06, lr=0.000388] \nSteps: 47%|████▋ | 234/500 [17:13<15:33, 3.51s/it, loss=1.06, lr=0.000388]\nSteps: 47%|████▋ | 234/500 [17:13<15:33, 3.51s/it, loss=1.07, lr=0.000387]\nSteps: 47%|████▋ | 235/500 [17:15<13:20, 3.02s/it, loss=1.07, lr=0.000387]\nSteps: 47%|████▋ | 235/500 [17:15<13:20, 3.02s/it, loss=0.996, lr=0.000387]\nSteps: 47%|████▋ | 236/500 [17:17<11:46, 2.68s/it, loss=0.996, lr=0.000387]\nSteps: 47%|████▋ | 236/500 [17:17<11:46, 2.68s/it, loss=0.889, lr=0.000386]\nSteps: 47%|████▋ | 237/500 [17:25<18:25, 4.20s/it, loss=0.889, lr=0.000386]\nSteps: 47%|████▋ | 237/500 [17:25<18:25, 4.20s/it, loss=0.789, lr=0.000385]\nSteps: 48%|████▊ | 238/500 [17:27<15:18, 3.51s/it, loss=0.789, lr=0.000385]\nSteps: 48%|████▊ | 238/500 [17:27<15:18, 3.51s/it, loss=1.04, lr=0.000384] \nSteps: 48%|████▊ | 239/500 [17:29<13:07, 3.02s/it, loss=1.04, lr=0.000384]\nSteps: 48%|████▊ | 239/500 [17:29<13:07, 3.02s/it, loss=0.85, lr=0.000384]\nSteps: 48%|████▊ | 240/500 [17:31<11:35, 2.67s/it, loss=0.85, lr=0.000384]\nSteps: 48%|████▊ | 240/500 [17:31<11:35, 2.67s/it, loss=0.976, lr=0.000383]\nSteps: 48%|████▊ | 241/500 [17:38<18:00, 4.17s/it, loss=0.976, lr=0.000383]\nSteps: 48%|████▊ | 241/500 [17:38<18:00, 4.17s/it, loss=0.842, lr=0.000382]\nSteps: 48%|████▊ | 242/500 [17:40<14:57, 3.48s/it, loss=0.842, lr=0.000382]\nSteps: 48%|████▊ | 242/500 [17:40<14:57, 3.48s/it, loss=1.01, lr=0.000381] \nSteps: 49%|████▊ | 243/500 [17:42<12:50, 3.00s/it, loss=1.01, lr=0.000381]\nSteps: 49%|████▊ | 243/500 [17:42<12:50, 3.00s/it, loss=0.848, lr=0.00038]\nSteps: 49%|████▉ | 244/500 [17:44<11:21, 2.66s/it, loss=0.848, lr=0.00038]\nSteps: 49%|████▉ | 244/500 [17:44<11:21, 2.66s/it, loss=1.07, lr=0.000379]\nSteps: 49%|████▉ | 245/500 [17:51<17:42, 4.17s/it, loss=1.07, lr=0.000379]\nSteps: 49%|████▉ | 245/500 [17:51<17:42, 4.17s/it, loss=1.05, lr=0.000378]\nSteps: 49%|████▉ | 246/500 [17:53<14:43, 3.48s/it, loss=1.05, lr=0.000378]\nSteps: 49%|████▉ | 246/500 [17:53<14:43, 3.48s/it, loss=0.908, lr=0.000377]\nSteps: 49%|████▉ | 247/500 [17:55<12:38, 3.00s/it, loss=0.908, lr=0.000377]\nSteps: 49%|████▉ | 247/500 [17:55<12:38, 3.00s/it, loss=0.8, lr=0.000376] \nSteps: 50%|████▉ | 248/500 [17:57<11:10, 2.66s/it, loss=0.8, lr=0.000376]\nSteps: 50%|████▉ | 248/500 [17:57<11:10, 2.66s/it, loss=1.07, lr=0.000375]\nSteps: 50%|████▉ | 249/500 [18:05<17:29, 4.18s/it, loss=1.07, lr=0.000375]\nSteps: 50%|████▉ | 249/500 [18:05<17:29, 4.18s/it, loss=0.966, lr=0.000374]\nSteps: 50%|█████ | 250/500 [18:07<14:31, 3.49s/it, loss=0.966, lr=0.000374]\nSteps: 50%|█████ | 250/500 [18:07<14:31, 3.49s/it, loss=1.07, lr=0.000373] \nSteps: 50%|█████ | 251/500 [18:09<12:27, 3.00s/it, loss=1.07, lr=0.000373]\nSteps: 50%|█████ | 251/500 [18:09<12:27, 3.00s/it, loss=0.789, lr=0.000372]\nSteps: 50%|█████ | 252/500 [18:10<11:00, 2.66s/it, loss=0.789, lr=0.000372]\nSteps: 50%|█████ | 252/500 [18:10<11:00, 2.66s/it, loss=0.945, lr=0.000371]\nSteps: 51%|█████ | 253/500 [18:18<17:08, 4.16s/it, loss=0.945, lr=0.000371]\nSteps: 51%|█████ | 253/500 [18:18<17:08, 4.16s/it, loss=0.83, lr=0.00037] \nSteps: 51%|█████ | 254/500 [18:20<14:15, 3.48s/it, loss=0.83, lr=0.00037]\nSteps: 51%|█████ | 254/500 [18:20<14:15, 3.48s/it, loss=0.999, lr=0.000369]\nSteps: 51%|█████ | 255/500 [18:22<12:13, 3.00s/it, loss=0.999, lr=0.000369]\nSteps: 51%|█████ | 255/500 [18:22<12:13, 3.00s/it, loss=0.883, lr=0.000368]\nSteps: 51%|█████ | 256/500 [18:24<10:48, 2.66s/it, loss=0.883, lr=0.000368]\nSteps: 51%|█████ | 256/500 [18:24<10:48, 2.66s/it, loss=1.07, lr=0.000367] \nSteps: 51%|█████▏ | 257/500 [18:32<17:04, 4.21s/it, loss=1.07, lr=0.000367]\nSteps: 51%|█████▏ | 257/500 [18:32<17:04, 4.21s/it, loss=0.81, lr=0.000365]\nSteps: 52%|█████▏ | 258/500 [18:33<14:10, 3.51s/it, loss=0.81, lr=0.000365]\nSteps: 52%|█████▏ | 258/500 [18:33<14:10, 3.51s/it, loss=0.94, lr=0.000364]\nSteps: 52%|█████▏ | 259/500 [18:35<12:08, 3.02s/it, loss=0.94, lr=0.000364]\nSteps: 52%|█████▏ | 259/500 [18:35<12:08, 3.02s/it, loss=0.963, lr=0.000363]\nSteps: 52%|█████▏ | 260/500 [18:37<10:42, 2.68s/it, loss=0.963, lr=0.000363]\nSteps: 52%|█████▏ | 260/500 [18:37<10:42, 2.68s/it, loss=0.942, lr=0.000362]\nSteps: 52%|█████▏ | 261/500 [18:45<16:53, 4.24s/it, loss=0.942, lr=0.000362]\nSteps: 52%|█████▏ | 261/500 [18:45<16:53, 4.24s/it, loss=0.962, lr=0.000361]\nSteps: 52%|█████▏ | 262/500 [18:47<14:00, 3.53s/it, loss=0.962, lr=0.000361]\nSteps: 52%|█████▏ | 262/500 [18:47<14:00, 3.53s/it, loss=0.922, lr=0.000359]\nSteps: 53%|█████▎ | 263/500 [18:49<11:58, 3.03s/it, loss=0.922, lr=0.000359]\nSteps: 53%|█████▎ | 263/500 [18:49<11:58, 3.03s/it, loss=0.8, lr=0.000358] \nSteps: 53%|█████▎ | 264/500 [18:51<10:33, 2.69s/it, loss=0.8, lr=0.000358]\nSteps: 53%|█████▎ | 264/500 [18:51<10:33, 2.69s/it, loss=0.954, lr=0.000357]\nSteps: 53%|█████▎ | 265/500 [18:58<16:26, 4.20s/it, loss=0.954, lr=0.000357]\nSteps: 53%|█████▎ | 265/500 [18:58<16:26, 4.20s/it, loss=0.852, lr=0.000355]\nSteps: 53%|█████▎ | 266/500 [19:00<13:39, 3.50s/it, loss=0.852, lr=0.000355]\nSteps: 53%|█████▎ | 266/500 [19:00<13:39, 3.50s/it, loss=0.9, lr=0.000354] \nSteps: 53%|█████▎ | 267/500 [19:02<11:42, 3.02s/it, loss=0.9, lr=0.000354]\nSteps: 53%|█████▎ | 267/500 [19:02<11:42, 3.02s/it, loss=0.838, lr=0.000353]\nSteps: 54%|█████▎ | 268/500 [19:04<10:20, 2.67s/it, loss=0.838, lr=0.000353]\nSteps: 54%|█████▎ | 268/500 [19:04<10:20, 2.67s/it, loss=1.07, lr=0.000351] \nSteps: 54%|█████▍ | 269/500 [19:12<16:02, 4.17s/it, loss=1.07, lr=0.000351]\nSteps: 54%|█████▍ | 269/500 [19:12<16:02, 4.17s/it, loss=0.983, lr=0.00035]\nSteps: 54%|█████▍ | 270/500 [19:14<13:20, 3.48s/it, loss=0.983, lr=0.00035]\nSteps: 54%|█████▍ | 270/500 [19:14<13:20, 3.48s/it, loss=0.957, lr=0.000349]\nSteps: 54%|█████▍ | 271/500 [19:15<11:26, 3.00s/it, loss=0.957, lr=0.000349]\nSteps: 54%|█████▍ | 271/500 [19:15<11:26, 3.00s/it, loss=0.828, lr=0.000347]\nSteps: 54%|█████▍ | 272/500 [19:17<10:06, 2.66s/it, loss=0.828, lr=0.000347]\nSteps: 54%|█████▍ | 272/500 [19:17<10:06, 2.66s/it, loss=0.946, lr=0.000346]\nSteps: 55%|█████▍ | 273/500 [19:25<15:43, 4.16s/it, loss=0.946, lr=0.000346]\nSteps: 55%|█████▍ | 273/500 [19:25<15:43, 4.16s/it, loss=1.01, lr=0.000344] \nSteps: 55%|█████▍ | 274/500 [19:27<13:04, 3.47s/it, loss=1.01, lr=0.000344]\nSteps: 55%|█████▍ | 274/500 [19:27<13:04, 3.47s/it, loss=0.915, lr=0.000343]\nSteps: 55%|█████▌ | 275/500 [19:29<11:13, 2.99s/it, loss=0.915, lr=0.000343]\nSteps: 55%|█████▌ | 275/500 [19:29<11:13, 2.99s/it, loss=0.881, lr=0.000341]\nSteps: 55%|█████▌ | 276/500 [19:31<09:55, 2.66s/it, loss=0.881, lr=0.000341]\nSteps: 55%|█████▌ | 276/500 [19:31<09:55, 2.66s/it, loss=0.896, lr=0.00034] \nSteps: 55%|█████▌ | 277/500 [19:38<15:23, 4.14s/it, loss=0.896, lr=0.00034]\nSteps: 55%|█████▌ | 277/500 [19:38<15:23, 4.14s/it, loss=0.863, lr=0.000338]\nSteps: 56%|█████▌ | 278/500 [19:40<12:48, 3.46s/it, loss=0.863, lr=0.000338]\nSteps: 56%|█████▌ | 278/500 [19:40<12:48, 3.46s/it, loss=0.968, lr=0.000337]\nSteps: 56%|█████▌ | 279/500 [19:42<10:59, 2.99s/it, loss=0.968, lr=0.000337]\nSteps: 56%|█████▌ | 279/500 [19:42<10:59, 2.99s/it, loss=0.817, lr=0.000335]\nSteps: 56%|█████▌ | 280/500 [19:44<09:43, 2.65s/it, loss=0.817, lr=0.000335]\nSteps: 56%|█████▌ | 280/500 [19:44<09:43, 2.65s/it, loss=1.07, lr=0.000334] \nSteps: 56%|█████▌ | 281/500 [19:51<15:08, 4.15s/it, loss=1.07, lr=0.000334]\nSteps: 56%|█████▌ | 281/500 [19:51<15:08, 4.15s/it, loss=0.795, lr=0.000332]\nSteps: 56%|█████▋ | 282/500 [19:53<12:35, 3.46s/it, loss=0.795, lr=0.000332]\nSteps: 56%|█████▋ | 282/500 [19:53<12:35, 3.46s/it, loss=0.99, lr=0.000331] \nSteps: 57%|█████▋ | 283/500 [19:55<10:48, 2.99s/it, loss=0.99, lr=0.000331]\nSteps: 57%|█████▋ | 283/500 [19:55<10:48, 2.99s/it, loss=0.844, lr=0.000329]\nSteps: 57%|█████▋ | 284/500 [19:57<09:32, 2.65s/it, loss=0.844, lr=0.000329]\nSteps: 57%|█████▋ | 284/500 [19:57<09:32, 2.65s/it, loss=0.94, lr=0.000327] \nSteps: 57%|█████▋ | 285/500 [20:05<14:54, 4.16s/it, loss=0.94, lr=0.000327]\nSteps: 57%|█████▋ | 285/500 [20:05<14:54, 4.16s/it, loss=0.9, lr=0.000326] \nSteps: 57%|█████▋ | 286/500 [20:07<12:24, 3.48s/it, loss=0.9, lr=0.000326]\nSteps: 57%|█████▋ | 286/500 [20:07<12:24, 3.48s/it, loss=1.06, lr=0.000324]\nSteps: 57%|█████▋ | 287/500 [20:09<10:38, 3.00s/it, loss=1.06, lr=0.000324]\nSteps: 57%|█████▋ | 287/500 [20:09<10:38, 3.00s/it, loss=1.02, lr=0.000323]\nSteps: 58%|█████▊ | 288/500 [20:10<09:23, 2.66s/it, loss=1.02, lr=0.000323]\nSteps: 58%|█████▊ | 288/500 [20:10<09:23, 2.66s/it, loss=1.03, lr=0.000321]\nSteps: 58%|█████▊ | 289/500 [20:18<14:35, 4.15s/it, loss=1.03, lr=0.000321]\nSteps: 58%|█████▊ | 289/500 [20:18<14:35, 4.15s/it, loss=1.05, lr=0.000319]\nSteps: 58%|█████▊ | 290/500 [20:20<12:07, 3.47s/it, loss=1.05, lr=0.000319]\nSteps: 58%|█████▊ | 290/500 [20:20<12:07, 3.47s/it, loss=0.899, lr=0.000318]\nSteps: 58%|█████▊ | 291/500 [20:22<10:24, 2.99s/it, loss=0.899, lr=0.000318]\nSteps: 58%|█████▊ | 291/500 [20:22<10:24, 2.99s/it, loss=1.03, lr=0.000316] \nSteps: 58%|█████▊ | 292/500 [20:24<09:11, 2.65s/it, loss=1.03, lr=0.000316]\nSteps: 58%|█████▊ | 292/500 [20:24<09:11, 2.65s/it, loss=1.03, lr=0.000314]\nSteps: 59%|█████▊ | 293/500 [20:31<14:28, 4.19s/it, loss=1.03, lr=0.000314]\nSteps: 59%|█████▊ | 293/500 [20:31<14:28, 4.19s/it, loss=0.821, lr=0.000312]\nSteps: 59%|█████▉ | 294/500 [20:33<12:00, 3.50s/it, loss=0.821, lr=0.000312]\nSteps: 59%|█████▉ | 294/500 [20:33<12:00, 3.50s/it, loss=0.884, lr=0.000311]\nSteps: 59%|█████▉ | 295/500 [20:35<10:17, 3.01s/it, loss=0.884, lr=0.000311]\nSteps: 59%|█████▉ | 295/500 [20:35<10:17, 3.01s/it, loss=0.792, lr=0.000309]\nSteps: 59%|█████▉ | 296/500 [20:37<09:04, 2.67s/it, loss=0.792, lr=0.000309]\nSteps: 59%|█████▉ | 296/500 [20:37<09:04, 2.67s/it, loss=1.01, lr=0.000307] \nSteps: 59%|█████▉ | 297/500 [20:45<14:02, 4.15s/it, loss=1.01, lr=0.000307]\nSteps: 59%|█████▉ | 297/500 [20:45<14:02, 4.15s/it, loss=0.787, lr=0.000305]\nSteps: 60%|█████▉ | 298/500 [20:47<11:40, 3.47s/it, loss=0.787, lr=0.000305]\nSteps: 60%|█████▉ | 298/500 [20:47<11:40, 3.47s/it, loss=0.909, lr=0.000304]\nSteps: 60%|█████▉ | 299/500 [20:48<10:00, 2.99s/it, loss=0.909, lr=0.000304]\nSteps: 60%|█████▉ | 299/500 [20:48<10:00, 2.99s/it, loss=0.832, lr=0.000302]\nSteps: 60%|██████ | 300/500 [20:50<08:50, 2.65s/it, loss=0.832, lr=0.000302]\nSteps: 60%|██████ | 300/500 [20:50<08:50, 2.65s/it, loss=0.945, lr=0.0003] \nSteps: 60%|██████ | 301/500 [20:58<13:47, 4.16s/it, loss=0.945, lr=0.0003]\nSteps: 60%|██████ | 301/500 [20:58<13:47, 4.16s/it, loss=0.866, lr=0.000298]\nSteps: 60%|██████ | 302/500 [21:00<11:27, 3.47s/it, loss=0.866, lr=0.000298]\nSteps: 60%|██████ | 302/500 [21:00<11:27, 3.47s/it, loss=0.905, lr=0.000296]\nSteps: 61%|██████ | 303/500 [21:02<09:49, 2.99s/it, loss=0.905, lr=0.000296]\nSteps: 61%|██████ | 303/500 [21:02<09:49, 2.99s/it, loss=0.818, lr=0.000295]\nSteps: 61%|██████ | 304/500 [21:04<08:40, 2.66s/it, loss=0.818, lr=0.000295]\nSteps: 61%|██████ | 304/500 [21:04<08:40, 2.66s/it, loss=0.912, lr=0.000293]\nSteps: 61%|██████ | 305/500 [21:11<13:32, 4.16s/it, loss=0.912, lr=0.000293]\nSteps: 61%|██████ | 305/500 [21:11<13:32, 4.16s/it, loss=0.784, lr=0.000291]\nSteps: 61%|██████ | 306/500 [21:13<11:14, 3.48s/it, loss=0.784, lr=0.000291]\nSteps: 61%|██████ | 306/500 [21:13<11:14, 3.48s/it, loss=1.03, lr=0.000289] \nSteps: 61%|██████▏ | 307/500 [21:15<09:38, 3.00s/it, loss=1.03, lr=0.000289]\nSteps: 61%|██████▏ | 307/500 [21:15<09:38, 3.00s/it, loss=1.05, lr=0.000287]\nSteps: 62%|██████▏ | 308/500 [21:17<08:30, 2.66s/it, loss=1.05, lr=0.000287]\nSteps: 62%|██████▏ | 308/500 [21:17<08:30, 2.66s/it, loss=1.04, lr=0.000285]\nSteps: 62%|██████▏ | 309/500 [21:25<13:15, 4.16s/it, loss=1.04, lr=0.000285]\nSteps: 62%|██████▏ | 309/500 [21:25<13:15, 4.16s/it, loss=1.06, lr=0.000283]\nSteps: 62%|██████▏ | 310/500 [21:26<11:00, 3.47s/it, loss=1.06, lr=0.000283]\nSteps: 62%|██████▏ | 310/500 [21:26<11:00, 3.47s/it, loss=0.99, lr=0.000281]\nSteps: 62%|██████▏ | 311/500 [21:28<09:25, 2.99s/it, loss=0.99, lr=0.000281]\nSteps: 62%|██████▏ | 311/500 [21:28<09:25, 2.99s/it, loss=0.86, lr=0.000279]\nSteps: 62%|██████▏ | 312/500 [21:30<08:19, 2.66s/it, loss=0.86, lr=0.000279]\nSteps: 62%|██████▏ | 312/500 [21:30<08:19, 2.66s/it, loss=0.877, lr=0.000278]\nSteps: 63%|██████▎ | 313/500 [21:38<13:00, 4.17s/it, loss=0.877, lr=0.000278]\nSteps: 63%|██████▎ | 313/500 [21:38<13:00, 4.17s/it, loss=0.82, lr=0.000276] \nSteps: 63%|██████▎ | 314/500 [21:40<10:47, 3.48s/it, loss=0.82, lr=0.000276]\nSteps: 63%|██████▎ | 314/500 [21:40<10:47, 3.48s/it, loss=0.89, lr=0.000274]\nSteps: 63%|██████▎ | 315/500 [21:42<09:15, 3.00s/it, loss=0.89, lr=0.000274]\nSteps: 63%|██████▎ | 315/500 [21:42<09:15, 3.00s/it, loss=0.855, lr=0.000272]\nSteps: 63%|██████▎ | 316/500 [21:43<08:09, 2.66s/it, loss=0.855, lr=0.000272]\nSteps: 63%|██████▎ | 316/500 [21:43<08:09, 2.66s/it, loss=1.01, lr=0.00027] \nSteps: 63%|██████▎ | 317/500 [21:51<12:43, 4.17s/it, loss=1.01, lr=0.00027]\nSteps: 63%|██████▎ | 317/500 [21:51<12:43, 4.17s/it, loss=0.9, lr=0.000268]\nSteps: 64%|██████▎ | 318/500 [21:53<10:33, 3.48s/it, loss=0.9, lr=0.000268]\nSteps: 64%|██████▎ | 318/500 [21:53<10:33, 3.48s/it, loss=0.966, lr=0.000266]\nSteps: 64%|██████▍ | 319/500 [21:55<09:02, 3.00s/it, loss=0.966, lr=0.000266]\nSteps: 64%|██████▍ | 319/500 [21:55<09:02, 3.00s/it, loss=0.968, lr=0.000264]\nSteps: 64%|██████▍ | 320/500 [21:57<07:58, 2.66s/it, loss=0.968, lr=0.000264]\nSteps: 64%|██████▍ | 320/500 [21:57<07:58, 2.66s/it, loss=0.891, lr=0.000262]\nSteps: 64%|██████▍ | 321/500 [22:04<12:20, 4.14s/it, loss=0.891, lr=0.000262]\nSteps: 64%|██████▍ | 321/500 [22:04<12:20, 4.14s/it, loss=0.787, lr=0.00026] \nSteps: 64%|██████▍ | 322/500 [22:06<10:15, 3.46s/it, loss=0.787, lr=0.00026]\nSteps: 64%|██████▍ | 322/500 [22:06<10:15, 3.46s/it, loss=0.878, lr=0.000258]\nSteps: 65%|██████▍ | 323/500 [22:08<08:47, 2.98s/it, loss=0.878, lr=0.000258]\nSteps: 65%|██████▍ | 323/500 [22:08<08:47, 2.98s/it, loss=0.852, lr=0.000256]\nSteps: 65%|██████▍ | 324/500 [22:10<07:46, 2.65s/it, loss=0.852, lr=0.000256]\nSteps: 65%|██████▍ | 324/500 [22:10<07:46, 2.65s/it, loss=1.02, lr=0.000254] \nSteps: 65%|██████▌ | 325/500 [22:18<12:03, 4.14s/it, loss=1.02, lr=0.000254]\nSteps: 65%|██████▌ | 325/500 [22:18<12:03, 4.14s/it, loss=0.878, lr=0.000252]\nSteps: 65%|██████▌ | 326/500 [22:19<10:01, 3.46s/it, loss=0.878, lr=0.000252]\nSteps: 65%|██████▌ | 326/500 [22:19<10:01, 3.46s/it, loss=0.878, lr=0.00025] \nSteps: 65%|██████▌ | 327/500 [22:21<08:35, 2.98s/it, loss=0.878, lr=0.00025]\nSteps: 65%|██████▌ | 327/500 [22:21<08:35, 2.98s/it, loss=0.845, lr=0.000248]\nSteps: 66%|██████▌ | 328/500 [22:23<07:35, 2.65s/it, loss=0.845, lr=0.000248]\nSteps: 66%|██████▌ | 328/500 [22:23<07:35, 2.65s/it, loss=0.905, lr=0.000246]\nSteps: 66%|██████▌ | 329/500 [22:31<11:55, 4.18s/it, loss=0.905, lr=0.000246]\nSteps: 66%|██████▌ | 329/500 [22:31<11:55, 4.18s/it, loss=1.05, lr=0.000244] \nSteps: 66%|██████▌ | 330/500 [22:33<09:53, 3.49s/it, loss=1.05, lr=0.000244]\nSteps: 66%|██████▌ | 330/500 [22:33<09:53, 3.49s/it, loss=0.936, lr=0.000242]\nSteps: 66%|██████▌ | 331/500 [22:35<08:27, 3.00s/it, loss=0.936, lr=0.000242]\nSteps: 66%|██████▌ | 331/500 [22:35<08:27, 3.00s/it, loss=0.834, lr=0.00024] \nSteps: 66%|██████▋ | 332/500 [22:37<07:27, 2.67s/it, loss=0.834, lr=0.00024]\nSteps: 66%|██████▋ | 332/500 [22:37<07:27, 2.67s/it, loss=1, lr=0.000237] \nSteps: 67%|██████▋ | 333/500 [22:44<11:31, 4.14s/it, loss=1, lr=0.000237]\nSteps: 67%|██████▋ | 333/500 [22:44<11:31, 4.14s/it, loss=0.791, lr=0.000235]\nSteps: 67%|██████▋ | 334/500 [22:46<09:34, 3.46s/it, loss=0.791, lr=0.000235]\nSteps: 67%|██████▋ | 334/500 [22:46<09:34, 3.46s/it, loss=0.893, lr=0.000233]\nSteps: 67%|██████▋ | 335/500 [22:48<08:12, 2.98s/it, loss=0.893, lr=0.000233]\nSteps: 67%|██████▋ | 335/500 [22:48<08:12, 2.98s/it, loss=1.03, lr=0.000231] \nSteps: 67%|██████▋ | 336/500 [22:50<07:14, 2.65s/it, loss=1.03, lr=0.000231]\nSteps: 67%|██████▋ | 336/500 [22:50<07:14, 2.65s/it, loss=1.03, lr=0.000229]\nSteps: 67%|██████▋ | 337/500 [22:58<11:23, 4.20s/it, loss=1.03, lr=0.000229]\nSteps: 67%|██████▋ | 337/500 [22:58<11:23, 4.20s/it, loss=1.03, lr=0.000227]\nSteps: 68%|██████▊ | 338/500 [22:59<09:26, 3.50s/it, loss=1.03, lr=0.000227]\nSteps: 68%|██████▊ | 338/500 [22:59<09:26, 3.50s/it, loss=0.882, lr=0.000225]\nSteps: 68%|██████▊ | 339/500 [23:01<08:04, 3.01s/it, loss=0.882, lr=0.000225]\nSteps: 68%|██████▊ | 339/500 [23:01<08:04, 3.01s/it, loss=0.792, lr=0.000223]\nSteps: 68%|██████▊ | 340/500 [23:03<07:07, 2.67s/it, loss=0.792, lr=0.000223]\nSteps: 68%|██████▊ | 340/500 [23:03<07:07, 2.67s/it, loss=0.974, lr=0.000221]\nSteps: 68%|██████▊ | 341/500 [23:11<10:57, 4.14s/it, loss=0.974, lr=0.000221]\nSteps: 68%|██████▊ | 341/500 [23:11<10:57, 4.14s/it, loss=0.83, lr=0.000219] \nSteps: 68%|██████▊ | 342/500 [23:13<09:06, 3.46s/it, loss=0.83, lr=0.000219]\nSteps: 68%|██████▊ | 342/500 [23:13<09:06, 3.46s/it, loss=0.874, lr=0.000217]\nSteps: 69%|██████▊ | 343/500 [23:14<07:48, 2.98s/it, loss=0.874, lr=0.000217]\nSteps: 69%|██████▊ | 343/500 [23:15<07:48, 2.98s/it, loss=0.789, lr=0.000215]\nSteps: 69%|██████▉ | 344/500 [23:16<06:53, 2.65s/it, loss=0.789, lr=0.000215]\nSteps: 69%|██████▉ | 344/500 [23:16<06:53, 2.65s/it, loss=0.975, lr=0.000213]\nSteps: 69%|██████▉ | 345/500 [23:24<10:41, 4.14s/it, loss=0.975, lr=0.000213]\nSteps: 69%|██████▉ | 345/500 [23:24<10:41, 4.14s/it, loss=0.838, lr=0.00021] \nSteps: 69%|██████▉ | 346/500 [23:26<08:52, 3.46s/it, loss=0.838, lr=0.00021]\nSteps: 69%|██████▉ | 346/500 [23:26<08:52, 3.46s/it, loss=1.02, lr=0.000208]\nSteps: 69%|██████▉ | 347/500 [23:28<07:36, 2.98s/it, loss=1.02, lr=0.000208]\nSteps: 69%|██████▉ | 347/500 [23:28<07:36, 2.98s/it, loss=0.815, lr=0.000206]\nSteps: 70%|██████▉ | 348/500 [23:30<06:43, 2.65s/it, loss=0.815, lr=0.000206]\nSteps: 70%|██████▉ | 348/500 [23:30<06:43, 2.65s/it, loss=0.865, lr=0.000204]\nSteps: 70%|██████▉ | 349/500 [23:37<10:27, 4.15s/it, loss=0.865, lr=0.000204]\nSteps: 70%|██████▉ | 349/500 [23:37<10:27, 4.15s/it, loss=0.806, lr=0.000202]\nSteps: 70%|███████ | 350/500 [23:39<08:40, 3.47s/it, loss=0.806, lr=0.000202]\nSteps: 70%|███████ | 350/500 [23:39<08:40, 3.47s/it, loss=0.869, lr=0.0002] \nSteps: 70%|███████ | 351/500 [23:41<07:25, 2.99s/it, loss=0.869, lr=0.0002]\nSteps: 70%|███████ | 351/500 [23:41<07:25, 2.99s/it, loss=0.812, lr=0.000198]\nSteps: 70%|███████ | 352/500 [23:43<06:33, 2.66s/it, loss=0.812, lr=0.000198]\nSteps: 70%|███████ | 352/500 [23:43<06:33, 2.66s/it, loss=1.01, lr=0.000196] \nSteps: 71%|███████ | 353/500 [23:51<10:15, 4.19s/it, loss=1.01, lr=0.000196]\nSteps: 71%|███████ | 353/500 [23:51<10:15, 4.19s/it, loss=1.01, lr=0.000194]\nSteps: 71%|███████ | 354/500 [23:53<08:29, 3.49s/it, loss=1.01, lr=0.000194]\nSteps: 71%|███████ | 354/500 [23:53<08:29, 3.49s/it, loss=0.951, lr=0.000192]\nSteps: 71%|███████ | 355/500 [23:54<07:15, 3.01s/it, loss=0.951, lr=0.000192]\nSteps: 71%|███████ | 355/500 [23:54<07:15, 3.01s/it, loss=0.849, lr=0.00019] \nSteps: 71%|███████ | 356/500 [23:56<06:23, 2.67s/it, loss=0.849, lr=0.00019]\nSteps: 71%|███████ | 356/500 [23:56<06:23, 2.67s/it, loss=1.06, lr=0.000187]\nSteps: 71%|███████▏ | 357/500 [24:04<09:51, 4.14s/it, loss=1.06, lr=0.000187]\nSteps: 71%|███████▏ | 357/500 [24:04<09:51, 4.14s/it, loss=1.03, lr=0.000185]\nSteps: 72%|███████▏ | 358/500 [24:06<08:10, 3.46s/it, loss=1.03, lr=0.000185]\nSteps: 72%|███████▏ | 358/500 [24:06<08:10, 3.46s/it, loss=0.889, lr=0.000183]\nSteps: 72%|███████▏ | 359/500 [24:08<07:00, 2.98s/it, loss=0.889, lr=0.000183]\nSteps: 72%|███████▏ | 359/500 [24:08<07:00, 2.98s/it, loss=0.818, lr=0.000181]\nSteps: 72%|███████▏ | 360/500 [24:09<06:10, 2.65s/it, loss=0.818, lr=0.000181]\nSteps: 72%|███████▏ | 360/500 [24:09<06:10, 2.65s/it, loss=1, lr=0.000179] \nSteps: 72%|███████▏ | 361/500 [24:17<09:34, 4.13s/it, loss=1, lr=0.000179]\nSteps: 72%|███████▏ | 361/500 [24:17<09:34, 4.13s/it, loss=0.996, lr=0.000177]\nSteps: 72%|███████▏ | 362/500 [24:19<07:57, 3.46s/it, loss=0.996, lr=0.000177]\nSteps: 72%|███████▏ | 362/500 [24:19<07:57, 3.46s/it, loss=1.04, lr=0.000175] \nSteps: 73%|███████▎ | 363/500 [24:21<06:48, 2.98s/it, loss=1.04, lr=0.000175]\nSteps: 73%|███████▎ | 363/500 [24:21<06:48, 2.98s/it, loss=0.784, lr=0.000173]\nSteps: 73%|███████▎ | 364/500 [24:23<06:00, 2.65s/it, loss=0.784, lr=0.000173]\nSteps: 73%|███████▎ | 364/500 [24:23<06:00, 2.65s/it, loss=0.997, lr=0.000171]\nSteps: 73%|███████▎ | 365/500 [24:30<09:22, 4.17s/it, loss=0.997, lr=0.000171]\nSteps: 73%|███████▎ | 365/500 [24:30<09:22, 4.17s/it, loss=0.794, lr=0.000169]\nSteps: 73%|███████▎ | 366/500 [24:32<07:45, 3.48s/it, loss=0.794, lr=0.000169]\nSteps: 73%|███████▎ | 366/500 [24:32<07:45, 3.48s/it, loss=0.874, lr=0.000167]\nSteps: 73%|███████▎ | 367/500 [24:34<06:38, 3.00s/it, loss=0.874, lr=0.000167]\nSteps: 73%|███████▎ | 367/500 [24:34<06:38, 3.00s/it, loss=0.848, lr=0.000165]\nSteps: 74%|███████▎ | 368/500 [24:36<05:50, 2.66s/it, loss=0.848, lr=0.000165]\nSteps: 74%|███████▎ | 368/500 [24:36<05:50, 2.66s/it, loss=0.964, lr=0.000163]\nSteps: 74%|███████▍ | 369/500 [24:44<09:07, 4.18s/it, loss=0.964, lr=0.000163]\nSteps: 74%|███████▍ | 369/500 [24:44<09:07, 4.18s/it, loss=0.778, lr=0.00016] \nSteps: 74%|███████▍ | 370/500 [24:46<07:33, 3.49s/it, loss=0.778, lr=0.00016]\nSteps: 74%|███████▍ | 370/500 [24:46<07:33, 3.49s/it, loss=1.04, lr=0.000158]\nSteps: 74%|███████▍ | 371/500 [24:47<06:27, 3.00s/it, loss=1.04, lr=0.000158]\nSteps: 74%|███████▍ | 371/500 [24:47<06:27, 3.00s/it, loss=1, lr=0.000156] \nSteps: 74%|███████▍ | 372/500 [24:49<05:41, 2.67s/it, loss=1, lr=0.000156]\nSteps: 74%|███████▍ | 372/500 [24:49<05:41, 2.67s/it, loss=0.937, lr=0.000154]\nSteps: 75%|███████▍ | 373/500 [24:57<08:47, 4.15s/it, loss=0.937, lr=0.000154]\nSteps: 75%|███████▍ | 373/500 [24:57<08:47, 4.15s/it, loss=1.05, lr=0.000152] \nSteps: 75%|███████▍ | 374/500 [24:59<07:17, 3.47s/it, loss=1.05, lr=0.000152]\nSteps: 75%|███████▍ | 374/500 [24:59<07:17, 3.47s/it, loss=0.894, lr=0.00015]\nSteps: 75%|███████▌ | 375/500 [25:01<06:13, 2.99s/it, loss=0.894, lr=0.00015]\nSteps: 75%|███████▌ | 375/500 [25:01<06:13, 2.99s/it, loss=0.821, lr=0.000148]\nSteps: 75%|███████▌ | 376/500 [25:03<05:29, 2.66s/it, loss=0.821, lr=0.000148]\nSteps: 75%|███████▌ | 376/500 [25:03<05:29, 2.66s/it, loss=1.04, lr=0.000146] \nSteps: 75%|███████▌ | 377/500 [25:10<08:34, 4.19s/it, loss=1.04, lr=0.000146]\nSteps: 75%|███████▌ | 377/500 [25:10<08:34, 4.19s/it, loss=0.978, lr=0.000144]\nSteps: 76%|███████▌ | 378/500 [25:12<07:05, 3.49s/it, loss=0.978, lr=0.000144]\nSteps: 76%|███████▌ | 378/500 [25:12<07:05, 3.49s/it, loss=0.943, lr=0.000142]\nSteps: 76%|███████▌ | 379/500 [25:14<06:03, 3.01s/it, loss=0.943, lr=0.000142]\nSteps: 76%|███████▌ | 379/500 [25:14<06:03, 3.01s/it, loss=1.05, lr=0.00014] \nSteps: 76%|███████▌ | 380/500 [25:16<05:20, 2.67s/it, loss=1.05, lr=0.00014]\nSteps: 76%|███████▌ | 380/500 [25:16<05:20, 2.67s/it, loss=0.892, lr=0.000138]\nSteps: 76%|███████▌ | 381/500 [25:24<08:12, 4.14s/it, loss=0.892, lr=0.000138]\nSteps: 76%|███████▌ | 381/500 [25:24<08:12, 4.14s/it, loss=0.82, lr=0.000136] \nSteps: 76%|███████▋ | 382/500 [25:25<06:48, 3.46s/it, loss=0.82, lr=0.000136]\nSteps: 76%|███████▋ | 382/500 [25:25<06:48, 3.46s/it, loss=1.02, lr=0.000134]\nSteps: 77%|███████▋ | 383/500 [25:27<05:49, 2.98s/it, loss=1.02, lr=0.000134]\nSteps: 77%|███████▋ | 383/500 [25:27<05:49, 2.98s/it, loss=0.785, lr=0.000132]\nSteps: 77%|███████▋ | 384/500 [25:29<05:07, 2.65s/it, loss=0.785, lr=0.000132]\nSteps: 77%|███████▋ | 384/500 [25:29<05:07, 2.65s/it, loss=0.898, lr=0.00013] \nSteps: 77%|███████▋ | 385/500 [25:37<07:56, 4.14s/it, loss=0.898, lr=0.00013]\nSteps: 77%|███████▋ | 385/500 [25:37<07:56, 4.14s/it, loss=0.836, lr=0.000128]\nSteps: 77%|███████▋ | 386/500 [25:39<06:34, 3.46s/it, loss=0.836, lr=0.000128]\nSteps: 77%|███████▋ | 386/500 [25:39<06:34, 3.46s/it, loss=0.894, lr=0.000126]\nSteps: 77%|███████▋ | 387/500 [25:41<05:37, 2.98s/it, loss=0.894, lr=0.000126]\nSteps: 77%|███████▋ | 387/500 [25:41<05:37, 2.98s/it, loss=0.776, lr=0.000124]\nSteps: 78%|███████▊ | 388/500 [25:42<04:56, 2.65s/it, loss=0.776, lr=0.000124]\nSteps: 78%|███████▊ | 388/500 [25:42<04:56, 2.65s/it, loss=1.06, lr=0.000122] \nSteps: 78%|███████▊ | 389/500 [25:50<07:38, 4.14s/it, loss=1.06, lr=0.000122]\nSteps: 78%|███████▊ | 389/500 [25:50<07:38, 4.14s/it, loss=0.79, lr=0.000121]\nSteps: 78%|███████▊ | 390/500 [25:52<06:20, 3.46s/it, loss=0.79, lr=0.000121]\nSteps: 78%|███████▊ | 390/500 [25:52<06:20, 3.46s/it, loss=0.867, lr=0.000119]\nSteps: 78%|███████▊ | 391/500 [25:54<05:25, 2.98s/it, loss=0.867, lr=0.000119]\nSteps: 78%|███████▊ | 391/500 [25:54<05:25, 2.98s/it, loss=0.79, lr=0.000117] \nSteps: 78%|███████▊ | 392/500 [25:56<04:46, 2.65s/it, loss=0.79, lr=0.000117]\nSteps: 78%|███████▊ | 392/500 [25:56<04:46, 2.65s/it, loss=0.867, lr=0.000115]\nSteps: 79%|███████▊ | 393/500 [26:03<07:21, 4.12s/it, loss=0.867, lr=0.000115]\nSteps: 79%|███████▊ | 393/500 [26:03<07:21, 4.12s/it, loss=0.818, lr=0.000113]\nSteps: 79%|███████▉ | 394/500 [26:05<06:05, 3.45s/it, loss=0.818, lr=0.000113]\nSteps: 79%|███████▉ | 394/500 [26:05<06:05, 3.45s/it, loss=0.931, lr=0.000111]\nSteps: 79%|███████▉ | 395/500 [26:07<05:12, 2.97s/it, loss=0.931, lr=0.000111]\nSteps: 79%|███████▉ | 395/500 [26:07<05:12, 2.97s/it, loss=0.821, lr=0.000109]\nSteps: 79%|███████▉ | 396/500 [26:09<04:35, 2.65s/it, loss=0.821, lr=0.000109]\nSteps: 79%|███████▉ | 396/500 [26:09<04:35, 2.65s/it, loss=0.916, lr=0.000107]\nSteps: 79%|███████▉ | 397/500 [26:17<07:10, 4.18s/it, loss=0.916, lr=0.000107]\nSteps: 79%|███████▉ | 397/500 [26:17<07:10, 4.18s/it, loss=0.805, lr=0.000105]\nSteps: 80%|███████▉ | 398/500 [26:18<05:55, 3.49s/it, loss=0.805, lr=0.000105]\nSteps: 80%|███████▉ | 398/500 [26:18<05:55, 3.49s/it, loss=1.06, lr=0.000104] \nSteps: 80%|███████▉ | 399/500 [26:20<05:03, 3.00s/it, loss=1.06, lr=0.000104]\nSteps: 80%|███████▉ | 399/500 [26:20<05:03, 3.00s/it, loss=0.812, lr=0.000102]\nSteps: 80%|████████ | 400/500 [26:22<04:26, 2.66s/it, loss=0.812, lr=0.000102]\nSteps: 80%|████████ | 400/500 [26:22<04:26, 2.66s/it, loss=0.863, lr=0.0001] \nSteps: 80%|████████ | 401/500 [26:30<06:54, 4.19s/it, loss=0.863, lr=0.0001]\nSteps: 80%|████████ | 401/500 [26:30<06:54, 4.19s/it, loss=0.843, lr=9.82e-5]\nSteps: 80%|████████ | 402/500 [26:32<05:42, 3.49s/it, loss=0.843, lr=9.82e-5]\nSteps: 80%|████████ | 402/500 [26:32<05:42, 3.49s/it, loss=0.926, lr=9.64e-5]\nSteps: 81%|████████ | 403/500 [26:34<04:51, 3.01s/it, loss=0.926, lr=9.64e-5]\nSteps: 81%|████████ | 403/500 [26:34<04:51, 3.01s/it, loss=0.953, lr=9.46e-5]\nSteps: 81%|████████ | 404/500 [26:36<04:15, 2.67s/it, loss=0.953, lr=9.46e-5]\nSteps: 81%|████████ | 404/500 [26:36<04:15, 2.67s/it, loss=1.01, lr=9.28e-5] \nSteps: 81%|████████ | 405/500 [26:43<06:40, 4.22s/it, loss=1.01, lr=9.28e-5]\nSteps: 81%|████████ | 405/500 [26:43<06:40, 4.22s/it, loss=0.825, lr=9.11e-5]\nSteps: 81%|████████ | 406/500 [26:45<05:30, 3.52s/it, loss=0.825, lr=9.11e-5]\nSteps: 81%|████████ | 406/500 [26:45<05:30, 3.52s/it, loss=0.909, lr=8.93e-5]\nSteps: 81%|████████▏ | 407/500 [26:47<04:41, 3.02s/it, loss=0.909, lr=8.93e-5]\nSteps: 81%|████████▏ | 407/500 [26:47<04:41, 3.02s/it, loss=0.781, lr=8.76e-5]\nSteps: 82%|████████▏ | 408/500 [26:49<04:06, 2.68s/it, loss=0.781, lr=8.76e-5]\nSteps: 82%|████████▏ | 408/500 [26:49<04:06, 2.68s/it, loss=0.862, lr=8.59e-5]\nSteps: 82%|████████▏ | 409/500 [26:57<06:22, 4.20s/it, loss=0.862, lr=8.59e-5]\nSteps: 82%|████████▏ | 409/500 [26:57<06:22, 4.20s/it, loss=0.822, lr=8.41e-5]\nSteps: 82%|████████▏ | 410/500 [26:59<05:15, 3.50s/it, loss=0.822, lr=8.41e-5]\nSteps: 82%|████████▏ | 410/500 [26:59<05:15, 3.50s/it, loss=1.01, lr=8.24e-5] \nSteps: 82%|████████▏ | 411/500 [27:01<04:28, 3.01s/it, loss=1.01, lr=8.24e-5]\nSteps: 82%|████████▏ | 411/500 [27:01<04:28, 3.01s/it, loss=0.829, lr=8.08e-5]\nSteps: 82%|████████▏ | 412/500 [27:02<03:55, 2.67s/it, loss=0.829, lr=8.08e-5]\nSteps: 82%|████████▏ | 412/500 [27:02<03:55, 2.67s/it, loss=0.92, lr=7.91e-5] \nSteps: 83%|████████▎ | 413/500 [27:10<06:02, 4.16s/it, loss=0.92, lr=7.91e-5]\nSteps: 83%|████████▎ | 413/500 [27:10<06:02, 4.16s/it, loss=0.789, lr=7.74e-5]\nSteps: 83%|████████▎ | 414/500 [27:12<04:59, 3.48s/it, loss=0.789, lr=7.74e-5]\nSteps: 83%|████████▎ | 414/500 [27:12<04:59, 3.48s/it, loss=0.979, lr=7.58e-5]\nSteps: 83%|████████▎ | 415/500 [27:14<04:14, 3.00s/it, loss=0.979, lr=7.58e-5]\nSteps: 83%|████████▎ | 415/500 [27:14<04:14, 3.00s/it, loss=0.963, lr=7.41e-5]\nSteps: 83%|████████▎ | 416/500 [27:16<03:43, 2.66s/it, loss=0.963, lr=7.41e-5]\nSteps: 83%|████████▎ | 416/500 [27:16<03:43, 2.66s/it, loss=0.903, lr=7.25e-5]\nSteps: 83%|████████▎ | 417/500 [27:23<05:42, 4.13s/it, loss=0.903, lr=7.25e-5]\nSteps: 83%|████████▎ | 417/500 [27:23<05:42, 4.13s/it, loss=0.803, lr=7.09e-5]\nSteps: 84%|████████▎ | 418/500 [27:25<04:42, 3.45s/it, loss=0.803, lr=7.09e-5]\nSteps: 84%|████████▎ | 418/500 [27:25<04:42, 3.45s/it, loss=0.924, lr=6.93e-5]\nSteps: 84%|████████▍ | 419/500 [27:27<04:01, 2.98s/it, loss=0.924, lr=6.93e-5]\nSteps: 84%|████████▍ | 419/500 [27:27<04:01, 2.98s/it, loss=0.769, lr=6.77e-5]\nSteps: 84%|████████▍ | 420/500 [27:29<03:31, 2.64s/it, loss=0.769, lr=6.77e-5]\nSteps: 84%|████████▍ | 420/500 [27:29<03:31, 2.64s/it, loss=0.99, lr=6.62e-5] \nSteps: 84%|████████▍ | 421/500 [27:37<05:28, 4.16s/it, loss=0.99, lr=6.62e-5]\nSteps: 84%|████████▍ | 421/500 [27:37<05:28, 4.16s/it, loss=0.778, lr=6.46e-5]\nSteps: 84%|████████▍ | 422/500 [27:38<04:31, 3.48s/it, loss=0.778, lr=6.46e-5]\nSteps: 84%|████████▍ | 422/500 [27:38<04:31, 3.48s/it, loss=0.994, lr=6.31e-5]\nSteps: 85%|████████▍ | 423/500 [27:40<03:50, 3.00s/it, loss=0.994, lr=6.31e-5]\nSteps: 85%|████████▍ | 423/500 [27:40<03:50, 3.00s/it, loss=0.845, lr=6.16e-5]\nSteps: 85%|████████▍ | 424/500 [27:42<03:22, 2.66s/it, loss=0.845, lr=6.16e-5]\nSteps: 85%|████████▍ | 424/500 [27:42<03:22, 2.66s/it, loss=0.944, lr=6.01e-5]\nSteps: 85%|████████▌ | 425/500 [27:50<05:13, 4.18s/it, loss=0.944, lr=6.01e-5]\nSteps: 85%|████████▌ | 425/500 [27:50<05:13, 4.18s/it, loss=0.771, lr=5.86e-5]\nSteps: 85%|████████▌ | 426/500 [27:52<04:17, 3.49s/it, loss=0.771, lr=5.86e-5]\nSteps: 85%|████████▌ | 426/500 [27:52<04:17, 3.49s/it, loss=0.932, lr=5.71e-5]\nSteps: 85%|████████▌ | 427/500 [27:54<03:39, 3.00s/it, loss=0.932, lr=5.71e-5]\nSteps: 85%|████████▌ | 427/500 [27:54<03:39, 3.00s/it, loss=0.771, lr=5.56e-5]\nSteps: 86%|████████▌ | 428/500 [27:55<03:11, 2.66s/it, loss=0.771, lr=5.56e-5]\nSteps: 86%|████████▌ | 428/500 [27:55<03:11, 2.66s/it, loss=0.861, lr=5.42e-5]\nSteps: 86%|████████▌ | 429/500 [28:03<04:57, 4.19s/it, loss=0.861, lr=5.42e-5]\nSteps: 86%|████████▌ | 429/500 [28:03<04:57, 4.19s/it, loss=0.836, lr=5.28e-5]\nSteps: 86%|████████▌ | 430/500 [28:05<04:04, 3.50s/it, loss=0.836, lr=5.28e-5]\nSteps: 86%|████████▌ | 430/500 [28:05<04:04, 3.50s/it, loss=0.99, lr=5.14e-5] \nSteps: 86%|████████▌ | 431/500 [28:07<03:27, 3.01s/it, loss=0.99, lr=5.14e-5]\nSteps: 86%|████████▌ | 431/500 [28:07<03:27, 3.01s/it, loss=0.804, lr=5e-5] \nSteps: 86%|████████▋ | 432/500 [28:09<03:01, 2.67s/it, loss=0.804, lr=5e-5]\nSteps: 86%|████████▋ | 432/500 [28:09<03:01, 2.67s/it, loss=0.885, lr=4.86e-5]\nSteps: 87%|████████▋ | 433/500 [28:17<04:42, 4.22s/it, loss=0.885, lr=4.86e-5]\nSteps: 87%|████████▋ | 433/500 [28:17<04:42, 4.22s/it, loss=1, lr=4.72e-5] \nSteps: 87%|████████▋ | 434/500 [28:19<03:51, 3.51s/it, loss=1, lr=4.72e-5]\nSteps: 87%|████████▋ | 434/500 [28:19<03:51, 3.51s/it, loss=0.999, lr=4.59e-5]\nSteps: 87%|████████▋ | 435/500 [28:20<03:16, 3.02s/it, loss=0.999, lr=4.59e-5]\nSteps: 87%|████████▋ | 435/500 [28:20<03:16, 3.02s/it, loss=0.774, lr=4.46e-5]\nSteps: 87%|████████▋ | 436/500 [28:22<02:51, 2.68s/it, loss=0.774, lr=4.46e-5]\nSteps: 87%|████████▋ | 436/500 [28:22<02:51, 2.68s/it, loss=0.946, lr=4.33e-5]\nSteps: 87%|████████▋ | 437/500 [28:30<04:21, 4.15s/it, loss=0.946, lr=4.33e-5]\nSteps: 87%|████████▋ | 437/500 [28:30<04:21, 4.15s/it, loss=0.841, lr=4.2e-5] \nSteps: 88%|████████▊ | 438/500 [28:32<03:34, 3.46s/it, loss=0.841, lr=4.2e-5]\nSteps: 88%|████████▊ | 438/500 [28:32<03:34, 3.46s/it, loss=1.01, lr=4.07e-5]\nSteps: 88%|████████▊ | 439/500 [28:34<03:02, 2.99s/it, loss=1.01, lr=4.07e-5]\nSteps: 88%|████████▊ | 439/500 [28:34<03:02, 2.99s/it, loss=0.882, lr=3.94e-5]\nSteps: 88%|████████▊ | 440/500 [28:36<02:39, 2.65s/it, loss=0.882, lr=3.94e-5]\nSteps: 88%|████████▊ | 440/500 [28:36<02:39, 2.65s/it, loss=0.937, lr=3.82e-5]\nSteps: 88%|████████▊ | 441/500 [28:44<04:11, 4.26s/it, loss=0.937, lr=3.82e-5]\nSteps: 88%|████████▊ | 441/500 [28:44<04:11, 4.26s/it, loss=0.851, lr=3.7e-5] \nSteps: 88%|████████▊ | 442/500 [28:45<03:25, 3.54s/it, loss=0.851, lr=3.7e-5]\nSteps: 88%|████████▊ | 442/500 [28:45<03:25, 3.54s/it, loss=0.866, lr=3.58e-5]\nSteps: 89%|████████▊ | 443/500 [28:47<02:53, 3.04s/it, loss=0.866, lr=3.58e-5]\nSteps: 89%|████████▊ | 443/500 [28:47<02:53, 3.04s/it, loss=0.918, lr=3.46e-5]\nSteps: 89%|████████▉ | 444/500 [28:49<02:30, 2.69s/it, loss=0.918, lr=3.46e-5]\nSteps: 89%|████████▉ | 444/500 [28:49<02:30, 2.69s/it, loss=0.892, lr=3.34e-5]\nSteps: 89%|████████▉ | 445/500 [28:57<03:49, 4.18s/it, loss=0.892, lr=3.34e-5]\nSteps: 89%|████████▉ | 445/500 [28:57<03:49, 4.18s/it, loss=0.787, lr=3.23e-5]\nSteps: 89%|████████▉ | 446/500 [28:59<03:08, 3.49s/it, loss=0.787, lr=3.23e-5]\nSteps: 89%|████████▉ | 446/500 [28:59<03:08, 3.49s/it, loss=0.89, lr=3.11e-5] \nSteps: 89%|████████▉ | 447/500 [29:01<02:39, 3.00s/it, loss=0.89, lr=3.11e-5]\nSteps: 89%|████████▉ | 447/500 [29:01<02:39, 3.00s/it, loss=1.05, lr=3e-5] \nSteps: 90%|████████▉ | 448/500 [29:02<02:18, 2.66s/it, loss=1.05, lr=3e-5]\nSteps: 90%|████████▉ | 448/500 [29:02<02:18, 2.66s/it, loss=1.01, lr=2.89e-5]\nSteps: 90%|████████▉ | 449/500 [29:10<03:31, 4.15s/it, loss=1.01, lr=2.89e-5]\nSteps: 90%|████████▉ | 449/500 [29:10<03:31, 4.15s/it, loss=0.801, lr=2.79e-5]\nSteps: 90%|█████████ | 450/500 [29:12<02:53, 3.47s/it, loss=0.801, lr=2.79e-5]\nSteps: 90%|█████████ | 450/500 [29:12<02:53, 3.47s/it, loss=0.908, lr=2.68e-5]\nSteps: 90%|█████████ | 451/500 [29:14<02:26, 2.99s/it, loss=0.908, lr=2.68e-5]\nSteps: 90%|█████████ | 451/500 [29:14<02:26, 2.99s/it, loss=0.758, lr=2.58e-5]\nSteps: 90%|█████████ | 452/500 [29:16<02:07, 2.65s/it, loss=0.758, lr=2.58e-5]\nSteps: 90%|█████████ | 452/500 [29:16<02:07, 2.65s/it, loss=0.872, lr=2.47e-5]\nSteps: 91%|█████████ | 453/500 [29:23<03:15, 4.15s/it, loss=0.872, lr=2.47e-5]\nSteps: 91%|█████████ | 453/500 [29:23<03:15, 4.15s/it, loss=0.784, lr=2.37e-5]\nSteps: 91%|█████████ | 454/500 [29:25<02:39, 3.47s/it, loss=0.784, lr=2.37e-5]\nSteps: 91%|█████████ | 454/500 [29:25<02:39, 3.47s/it, loss=0.86, lr=2.28e-5] \nSteps: 91%|█████████ | 455/500 [29:27<02:14, 2.99s/it, loss=0.86, lr=2.28e-5]\nSteps: 91%|█████████ | 455/500 [29:27<02:14, 2.99s/it, loss=0.839, lr=2.18e-5]\nSteps: 91%|█████████ | 456/500 [29:29<01:56, 2.66s/it, loss=0.839, lr=2.18e-5]\nSteps: 91%|█████████ | 456/500 [29:29<01:56, 2.66s/it, loss=1, lr=2.09e-5] \nSteps: 91%|█████████▏| 457/500 [29:37<02:59, 4.18s/it, loss=1, lr=2.09e-5]\nSteps: 91%|█████████▏| 457/500 [29:37<02:59, 4.18s/it, loss=1.05, lr=1.99e-5]\nSteps: 92%|█████████▏| 458/500 [29:39<02:26, 3.49s/it, loss=1.05, lr=1.99e-5]\nSteps: 92%|█████████▏| 458/500 [29:39<02:26, 3.49s/it, loss=0.987, lr=1.9e-5]\nSteps: 92%|█████████▏| 459/500 [29:40<02:03, 3.00s/it, loss=0.987, lr=1.9e-5]\nSteps: 92%|█████████▏| 459/500 [29:40<02:03, 3.00s/it, loss=0.831, lr=1.82e-5]\nSteps: 92%|█████████▏| 460/500 [29:42<01:46, 2.66s/it, loss=0.831, lr=1.82e-5]\nSteps: 92%|█████████▏| 460/500 [29:42<01:46, 2.66s/it, loss=0.996, lr=1.73e-5]\nSteps: 92%|█████████▏| 461/500 [29:50<02:42, 4.17s/it, loss=0.996, lr=1.73e-5]\nSteps: 92%|█████████▏| 461/500 [29:50<02:42, 4.17s/it, loss=0.805, lr=1.64e-5]\nSteps: 92%|█████████▏| 462/500 [29:52<02:12, 3.48s/it, loss=0.805, lr=1.64e-5]\nSteps: 92%|█████████▏| 462/500 [29:52<02:12, 3.48s/it, loss=1.05, lr=1.56e-5] \nSteps: 93%|█████████▎| 463/500 [29:54<01:50, 3.00s/it, loss=1.05, lr=1.56e-5]\nSteps: 93%|█████████▎| 463/500 [29:54<01:50, 3.00s/it, loss=0.837, lr=1.48e-5]\nSteps: 93%|█████████▎| 464/500 [29:56<01:35, 2.66s/it, loss=0.837, lr=1.48e-5]\nSteps: 93%|█████████▎| 464/500 [29:56<01:35, 2.66s/it, loss=0.87, lr=1.4e-5] \nSteps: 93%|█████████▎| 465/500 [30:03<02:24, 4.13s/it, loss=0.87, lr=1.4e-5]\nSteps: 93%|█████████▎| 465/500 [30:03<02:24, 4.13s/it, loss=0.973, lr=1.33e-5]\nSteps: 93%|█████████▎| 466/500 [30:05<01:57, 3.45s/it, loss=0.973, lr=1.33e-5]\nSteps: 93%|█████████▎| 466/500 [30:05<01:57, 3.45s/it, loss=0.984, lr=1.25e-5]\nSteps: 93%|█████████▎| 467/500 [30:07<01:38, 2.98s/it, loss=0.984, lr=1.25e-5]\nSteps: 93%|█████████▎| 467/500 [30:07<01:38, 2.98s/it, loss=0.84, lr=1.18e-5] \nSteps: 94%|█████████▎| 468/500 [30:09<01:24, 2.65s/it, loss=0.84, lr=1.18e-5]\nSteps: 94%|█████████▎| 468/500 [30:09<01:24, 2.65s/it, loss=0.917, lr=1.11e-5]\nSteps: 94%|█████████▍| 469/500 [30:16<02:08, 4.15s/it, loss=0.917, lr=1.11e-5]\nSteps: 94%|█████████▍| 469/500 [30:16<02:08, 4.15s/it, loss=0.838, lr=1.04e-5]\nSteps: 94%|█████████▍| 470/500 [30:18<01:43, 3.46s/it, loss=0.838, lr=1.04e-5]\nSteps: 94%|█████████▍| 470/500 [30:18<01:43, 3.46s/it, loss=0.949, lr=9.79e-6]\nSteps: 94%|█████████▍| 471/500 [30:20<01:26, 2.99s/it, loss=0.949, lr=9.79e-6]\nSteps: 94%|█████████▍| 471/500 [30:20<01:26, 2.99s/it, loss=0.809, lr=9.15e-6]\nSteps: 94%|█████████▍| 472/500 [30:22<01:14, 2.65s/it, loss=0.809, lr=9.15e-6]\nSteps: 94%|█████████▍| 472/500 [30:22<01:14, 2.65s/it, loss=1.05, lr=8.54e-6] \nSteps: 95%|█████████▍| 473/500 [30:30<01:52, 4.17s/it, loss=1.05, lr=8.54e-6]\nSteps: 95%|█████████▍| 473/500 [30:30<01:52, 4.17s/it, loss=0.781, lr=7.94e-6]\nSteps: 95%|█████████▍| 474/500 [30:32<01:30, 3.48s/it, loss=0.781, lr=7.94e-6]\nSteps: 95%|█████████▍| 474/500 [30:32<01:30, 3.48s/it, loss=1.02, lr=7.37e-6] \nSteps: 95%|█████████▌| 475/500 [30:33<01:14, 3.00s/it, loss=1.02, lr=7.37e-6]\nSteps: 95%|█████████▌| 475/500 [30:33<01:14, 3.00s/it, loss=0.77, lr=6.81e-6]\nSteps: 95%|█████████▌| 476/500 [30:35<01:03, 2.66s/it, loss=0.77, lr=6.81e-6]\nSteps: 95%|█████████▌| 476/500 [30:35<01:03, 2.66s/it, loss=0.964, lr=6.28e-6]\nSteps: 95%|█████████▌| 477/500 [30:43<01:35, 4.16s/it, loss=0.964, lr=6.28e-6]\nSteps: 95%|█████████▌| 477/500 [30:43<01:35, 4.16s/it, loss=0.813, lr=5.77e-6]\nSteps: 96%|█████████▌| 478/500 [30:45<01:16, 3.47s/it, loss=0.813, lr=5.77e-6]\nSteps: 96%|█████████▌| 478/500 [30:45<01:16, 3.47s/it, loss=1.06, lr=5.28e-6] \nSteps: 96%|█████████▌| 479/500 [30:47<01:02, 3.00s/it, loss=1.06, lr=5.28e-6]\nSteps: 96%|█████████▌| 479/500 [30:47<01:02, 3.00s/it, loss=0.959, lr=4.82e-6]\nSteps: 96%|█████████▌| 480/500 [30:49<00:53, 2.66s/it, loss=0.959, lr=4.82e-6]\nSteps: 96%|█████████▌| 480/500 [30:49<00:53, 2.66s/it, loss=0.859, lr=4.37e-6]\nSteps: 96%|█████████▌| 481/500 [30:56<01:19, 4.17s/it, loss=0.859, lr=4.37e-6]\nSteps: 96%|█████████▌| 481/500 [30:56<01:19, 4.17s/it, loss=1.03, lr=3.95e-6] \nSteps: 96%|█████████▋| 482/500 [30:58<01:02, 3.48s/it, loss=1.03, lr=3.95e-6]\nSteps: 96%|█████████▋| 482/500 [30:58<01:02, 3.48s/it, loss=1.07, lr=3.54e-6]\nSteps: 97%|█████████▋| 483/500 [31:00<00:50, 3.00s/it, loss=1.07, lr=3.54e-6]\nSteps: 97%|█████████▋| 483/500 [31:00<00:50, 3.00s/it, loss=0.791, lr=3.16e-6]\nSteps: 97%|█████████▋| 484/500 [31:02<00:42, 2.66s/it, loss=0.791, lr=3.16e-6]\nSteps: 97%|█████████▋| 484/500 [31:02<00:42, 2.66s/it, loss=0.934, lr=2.8e-6] \nSteps: 97%|█████████▋| 485/500 [31:10<01:02, 4.15s/it, loss=0.934, lr=2.8e-6]\nSteps: 97%|█████████▋| 485/500 [31:10<01:02, 4.15s/it, loss=0.777, lr=2.46e-6]\nSteps: 97%|█████████▋| 486/500 [31:11<00:48, 3.47s/it, loss=0.777, lr=2.46e-6]\nSteps: 97%|█████████▋| 486/500 [31:11<00:48, 3.47s/it, loss=0.957, lr=2.15e-6]\nSteps: 97%|█████████▋| 487/500 [31:13<00:38, 2.99s/it, loss=0.957, lr=2.15e-6]\nSteps: 97%|█████████▋| 487/500 [31:13<00:38, 2.99s/it, loss=1.04, lr=1.85e-6] \nSteps: 98%|█████████▊| 488/500 [31:15<00:31, 2.65s/it, loss=1.04, lr=1.85e-6]\nSteps: 98%|█████████▊| 488/500 [31:15<00:31, 2.65s/it, loss=1.01, lr=1.58e-6]\nSteps: 98%|█████████▊| 489/500 [31:23<00:45, 4.12s/it, loss=1.01, lr=1.58e-6]\nSteps: 98%|█████████▊| 489/500 [31:23<00:45, 4.12s/it, loss=0.779, lr=1.33e-6]\nSteps: 98%|█████████▊| 490/500 [31:25<00:34, 3.45s/it, loss=0.779, lr=1.33e-6]\nSteps: 98%|█████████▊| 490/500 [31:25<00:34, 3.45s/it, loss=0.955, lr=1.1e-6] \nSteps: 98%|█████████▊| 491/500 [31:26<00:26, 2.97s/it, loss=0.955, lr=1.1e-6]\nSteps: 98%|█████████▊| 491/500 [31:26<00:26, 2.97s/it, loss=0.839, lr=8.88e-7]\nSteps: 98%|█████████▊| 492/500 [31:28<00:21, 2.64s/it, loss=0.839, lr=8.88e-7]\nSteps: 98%|█████████▊| 492/500 [31:28<00:21, 2.64s/it, loss=1.06, lr=7.01e-7] \nSteps: 99%|█████████▊| 493/500 [31:36<00:29, 4.19s/it, loss=1.06, lr=7.01e-7]\nSteps: 99%|█████████▊| 493/500 [31:36<00:29, 4.19s/it, loss=0.975, lr=5.37e-7]\nSteps: 99%|█████████▉| 494/500 [31:38<00:20, 3.49s/it, loss=0.975, lr=5.37e-7]\nSteps: 99%|█████████▉| 494/500 [31:38<00:20, 3.49s/it, loss=0.987, lr=3.95e-7]\nSteps: 99%|█████████▉| 495/500 [31:40<00:15, 3.01s/it, loss=0.987, lr=3.95e-7]\nSteps: 99%|█████████▉| 495/500 [31:40<00:15, 3.01s/it, loss=0.783, lr=2.74e-7]\nSteps: 99%|█████████▉| 496/500 [31:42<00:10, 2.67s/it, loss=0.783, lr=2.74e-7]\nSteps: 99%|█████████▉| 496/500 [31:42<00:10, 2.67s/it, loss=0.941, lr=1.75e-7]\nSteps: 99%|█████████▉| 497/500 [31:49<00:12, 4.18s/it, loss=0.941, lr=1.75e-7]\nSteps: 99%|█████████▉| 497/500 [31:49<00:12, 4.18s/it, loss=0.78, lr=9.87e-8] \nSteps: 100%|█████████▉| 498/500 [31:51<00:06, 3.49s/it, loss=0.78, lr=9.87e-8]\nSteps: 100%|█████████▉| 498/500 [31:51<00:06, 3.49s/it, loss=0.928, lr=4.39e-8]\nSteps: 100%|█████████▉| 499/500 [31:53<00:03, 3.00s/it, loss=0.928, lr=4.39e-8]\nSteps: 100%|█████████▉| 499/500 [31:53<00:03, 3.00s/it, loss=1.03, lr=1.1e-8] \nSteps: 100%|██████████| 500/500 [31:55<00:00, 2.66s/it, loss=1.03, lr=1.1e-8]\nSteps: 100%|██████████| 500/500 [31:55<00:00, 2.66s/it, loss=0.906, lr=0] \nSteps: 100%|██████████| 500/500 [31:59<00:00, 3.84s/it, loss=0.906, lr=0]\n---Tar up output directory---\nmochi-lora/\nmochi-lora/pytorch_lora_weights.safetensors\nUploading to Hugging Face: lucataco/mochi-lora-vhs\nHF Repo URL: https://huggingface.co/lucataco/mochi-lora-vhs\npytorch_lora_weights.safetensors: 0%| | 0.00/76.1M [00:00<?, ?B/s]\npytorch_lora_weights.safetensors: 10%|▉ | 7.34M/76.1M [00:00<00:00, 73.4MB/s]\npytorch_lora_weights.safetensors: 21%|██ | 16.0M/76.1M [00:00<00:01, 42.3MB/s]\npytorch_lora_weights.safetensors: 42%|████▏ | 32.0M/76.1M [00:00<00:00, 46.4MB/s]\npytorch_lora_weights.safetensors: 63%|██████▎ | 48.0M/76.1M [00:00<00:00, 54.6MB/s]\npytorch_lora_weights.safetensors: 84%|████████▍ | 64.0M/76.1M [00:01<00:00, 57.3MB/s]\npytorch_lora_weights.safetensors: 100%|██████████| 76.1M/76.1M [00:01<00:00, 54.6MB/s]\nSuccessfully uploaded model to https://huggingface.co/lucataco/mochi-lora-vhs", "metrics": { "predict_time": 1939.37619092, "total_time": 1939.383131 }, "output": { "weights": "https://replicate.delivery/xezq/eWTxCE13svWocSUepK0rWZdxJhthKmFtpE2SzDBfrSVdcxznA/trained_model.tar" }, "started_at": "2024-12-11T17:55:06.819940Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/mxey6b9vqnrm80ckpveaqjqb9w", "cancel": "https://api.replicate.com/v1/predictions/mxey6b9vqnrm80ckpveaqjqb9w/cancel" }, "version": "170ea99fb48a30fef98cb1c9fb403a2882ab9d60c2ba15ad9383ace33c3fa385" }
Generated inCleaning up previous runs Extracted 8 files from zip to videos_input ---Starting to Trim input videos--- Processing: videos_input/vhs1.mp4 Copied videos_input/vhs1.txt to videos_prepared/vhs1.txt Moviepy - Building video videos_prepared/vhs1.mp4. Moviepy - Writing video videos_prepared/vhs1.mp4 0%| | 0/4 [00:00<?, ?it/s] 0%| | 0/4 [00:00<?, ?it/s] 0%| | 0/4 [00:00<?, ?it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] 0%| | 0/4 [00:00<?, ?it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/vhs1.mp4 0%| | 0/4 [00:00<?, ?it/s] Processing: videos_input/vhs2.mp4 Copied videos_input/vhs2.txt to videos_prepared/vhs2.txt Moviepy - Building video videos_prepared/vhs2.mp4. Moviepy - Writing video videos_prepared/vhs2.mp4 25%|██▌ | 1/4 [00:00<00:00, 3.16it/s] 25%|██▌ | 1/4 [00:00<00:00, 3.16it/s] 25%|██▌ | 1/4 [00:00<00:00, 3.16it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] 25%|██▌ | 1/4 [00:00<00:00, 3.16it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/vhs2.mp4 25%|██▌ | 1/4 [00:00<00:00, 3.16it/s] Processing: videos_input/vhs3.mp4 Copied videos_input/vhs3.txt to videos_prepared/vhs3.txt Moviepy - Building video videos_prepared/vhs3.mp4. 50%|█████ | 2/4 [00:00<00:00, 3.05it/s] 50%|█████ | 2/4 [00:00<00:00, 3.05it/s] Moviepy - Writing video videos_prepared/vhs3.mp4 50%|█████ | 2/4 [00:00<00:00, 3.05it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] 50%|█████ | 2/4 [00:00<00:00, 3.05it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/vhs3.mp4 50%|█████ | 2/4 [00:00<00:00, 3.05it/s] Processing: videos_input/vhs4.mp4 Copied videos_input/vhs4.txt to videos_prepared/vhs4.txt Moviepy - Building video videos_prepared/vhs4.mp4. Moviepy - Writing video videos_prepared/vhs4.mp4 75%|███████▌ | 3/4 [00:00<00:00, 3.05it/s] 75%|███████▌ | 3/4 [00:01<00:00, 3.05it/s] 75%|███████▌ | 3/4 [00:01<00:00, 3.05it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] 75%|███████▌ | 3/4 [00:01<00:00, 3.05it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/vhs4.mp4 75%|███████▌ | 3/4 [00:01<00:00, 3.05it/s] 100%|██████████| 4/4 [00:01<00:00, 3.07it/s] 100%|██████████| 4/4 [00:01<00:00, 3.07it/s] ---Starting to Embed videos--- Loading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s] Loading checkpoint shards: 50%|█████ | 1/2 [00:00<00:00, 1.67it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:01<00:00, 1.78it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:01<00:00, 1.76it/s] Loading pipeline components...: 0%| | 0/3 [00:00<?, ?it/s] Loading pipeline components...: 100%|██████████| 3/3 [00:00<00:00, 681.59it/s] Processing videos_prepared/vhs1.mp4 Trimmed video from 40 to first 37 frames 0it [00:00, ?it/s] Processing videos_prepared/vhs2.mp4 Trimmed video from 40 to first 37 frames 1it [00:01, 1.38s/it] Processing videos_prepared/vhs3.mp4 Trimmed video from 40 to first 37 frames 2it [00:02, 1.15s/it] Processing videos_prepared/vhs4.mp4 Trimmed video from 40 to first 37 frames 3it [00:03, 1.07s/it] 4it [00:04, 1.03s/it] 4it [00:04, 1.08s/it] ---Starting training--- Found 4 training videos in videos_prepared Loaded 4/4 valid file pairs. ===== Memory before training ===== memory_allocated=18.780 GB max_memory_allocated=18.780 GB max_memory_reserved=19.250 GB ***** Running training ***** Num trainable parameters = 19005440 Num examples = 4 Num batches each epoch = 4 Num epochs = 125 Instantaneous batch size per device = 1 Total train batch size (w. parallel, distributed & accumulation) = 1 Total optimization steps = 500 Steps: 0%| | 0/500 [00:00<?, ?it/s]W1211 17:57:31.075000 135012609543680 torch/fx/experimental/symbolic_shapes.py:4449] [0/0] xindex is not in var_ranges, defaulting to unknown range. W1211 17:57:31.089000 135012609543680 torch/fx/experimental/symbolic_shapes.py:4449] [0/0] xindex is not in var_ranges, defaulting to unknown range. W1211 17:57:31.224000 135012609543680 torch/fx/experimental/symbolic_shapes.py:4449] [0/0] xindex is not in var_ranges, defaulting to unknown range. Steps: 0%| | 1/500 [04:16<35:33:30, 256.53s/it] Steps: 0%| | 1/500 [04:16<35:33:30, 256.53s/it, loss=0.933, lr=2e-6] Steps: 0%| | 2/500 [04:18<14:45:44, 106.72s/it, loss=0.933, lr=2e-6] Steps: 0%| | 2/500 [04:18<14:45:44, 106.72s/it, loss=1.05, lr=4e-6] Steps: 1%| | 3/500 [04:20<8:07:22, 58.84s/it, loss=1.05, lr=4e-6] Steps: 1%| | 3/500 [04:20<8:07:22, 58.84s/it, loss=0.864, lr=6e-6] Steps: 1%| | 4/500 [04:22<5:00:27, 36.34s/it, loss=0.864, lr=6e-6] Steps: 1%| | 4/500 [04:22<5:00:27, 36.34s/it, loss=1.06, lr=8e-6] Steps: 1%| | 5/500 [04:29<3:34:15, 25.97s/it, loss=1.06, lr=8e-6] Steps: 1%| | 5/500 [04:29<3:34:15, 25.97s/it, loss=0.874, lr=1e-5] Steps: 1%| | 6/500 [04:31<2:26:21, 17.78s/it, loss=0.874, lr=1e-5] Steps: 1%| | 6/500 [04:31<2:26:21, 17.78s/it, loss=1.02, lr=1.2e-5] Steps: 1%|▏ | 7/500 [04:33<1:43:19, 12.58s/it, loss=1.02, lr=1.2e-5] Steps: 1%|▏ | 7/500 [04:33<1:43:19, 12.58s/it, loss=0.902, lr=1.4e-5] Steps: 2%|▏ | 8/500 [04:35<1:15:09, 9.17s/it, loss=0.902, lr=1.4e-5] Steps: 2%|▏ | 8/500 [04:35<1:15:09, 9.17s/it, loss=1.08, lr=1.6e-5] Steps: 2%|▏ | 9/500 [04:42<1:11:01, 8.68s/it, loss=1.08, lr=1.6e-5] Steps: 2%|▏ | 9/500 [04:42<1:11:01, 8.68s/it, loss=1.04, lr=1.8e-5] Steps: 2%|▏ | 10/500 [04:44<53:42, 6.58s/it, loss=1.04, lr=1.8e-5] Steps: 2%|▏ | 10/500 [04:44<53:42, 6.58s/it, loss=1.09, lr=2e-5] Steps: 2%|▏ | 11/500 [04:46<41:50, 5.13s/it, loss=1.09, lr=2e-5] Steps: 2%|▏ | 11/500 [04:46<41:50, 5.13s/it, loss=0.886, lr=2.2e-5] Steps: 2%|▏ | 12/500 [04:48<33:40, 4.14s/it, loss=0.886, lr=2.2e-5] Steps: 2%|▏ | 12/500 [04:48<33:40, 4.14s/it, loss=1.1, lr=2.4e-5] Steps: 3%|▎ | 13/500 [04:56<42:08, 5.19s/it, loss=1.1, lr=2.4e-5] Steps: 3%|▎ | 13/500 [04:56<42:08, 5.19s/it, loss=0.881, lr=2.6e-5] Steps: 3%|▎ | 14/500 [04:57<33:55, 4.19s/it, loss=0.881, lr=2.6e-5] Steps: 3%|▎ | 14/500 [04:57<33:55, 4.19s/it, loss=1.07, lr=2.8e-5] Steps: 3%|▎ | 15/500 [04:59<28:12, 3.49s/it, loss=1.07, lr=2.8e-5] Steps: 3%|▎ | 15/500 [04:59<28:12, 3.49s/it, loss=0.79, lr=3e-5] Steps: 3%|▎ | 16/500 [05:01<24:13, 3.00s/it, loss=0.79, lr=3e-5] Steps: 3%|▎ | 16/500 [05:01<24:13, 3.00s/it, loss=1.07, lr=3.2e-5] Steps: 3%|▎ | 17/500 [05:09<35:16, 4.38s/it, loss=1.07, lr=3.2e-5] Steps: 3%|▎ | 17/500 [05:09<35:16, 4.38s/it, loss=0.873, lr=3.4e-5] Steps: 4%|▎ | 18/500 [05:11<29:08, 3.63s/it, loss=0.873, lr=3.4e-5] Steps: 4%|▎ | 18/500 [05:11<29:08, 3.63s/it, loss=0.968, lr=3.6e-5] Steps: 4%|▍ | 19/500 [05:13<24:50, 3.10s/it, loss=0.968, lr=3.6e-5] Steps: 4%|▍ | 19/500 [05:13<24:50, 3.10s/it, loss=0.979, lr=3.8e-5] Steps: 4%|▍ | 20/500 [05:14<21:50, 2.73s/it, loss=0.979, lr=3.8e-5] Steps: 4%|▍ | 20/500 [05:14<21:50, 2.73s/it, loss=1.08, lr=4e-5] Steps: 4%|▍ | 21/500 [05:22<33:40, 4.22s/it, loss=1.08, lr=4e-5] Steps: 4%|▍ | 21/500 [05:22<33:40, 4.22s/it, loss=0.866, lr=4.2e-5] Steps: 4%|▍ | 22/500 [05:24<27:59, 3.51s/it, loss=0.866, lr=4.2e-5] Steps: 4%|▍ | 22/500 [05:24<27:59, 3.51s/it, loss=0.966, lr=4.4e-5] Steps: 5%|▍ | 23/500 [05:26<24:01, 3.02s/it, loss=0.966, lr=4.4e-5] Steps: 5%|▍ | 23/500 [05:26<24:01, 3.02s/it, loss=0.849, lr=4.6e-5] Steps: 5%|▍ | 24/500 [05:28<21:13, 2.68s/it, loss=0.849, lr=4.6e-5] Steps: 5%|▍ | 24/500 [05:28<21:13, 2.68s/it, loss=1.07, lr=4.8e-5] Steps: 5%|▌ | 25/500 [05:35<33:08, 4.19s/it, loss=1.07, lr=4.8e-5] Steps: 5%|▌ | 25/500 [05:35<33:08, 4.19s/it, loss=0.853, lr=5e-5] Steps: 5%|▌ | 26/500 [05:37<27:34, 3.49s/it, loss=0.853, lr=5e-5] Steps: 5%|▌ | 26/500 [05:37<27:34, 3.49s/it, loss=0.996, lr=5.2e-5] Steps: 5%|▌ | 27/500 [05:39<23:40, 3.00s/it, loss=0.996, lr=5.2e-5] Steps: 5%|▌ | 27/500 [05:39<23:40, 3.00s/it, loss=0.879, lr=5.4e-5] Steps: 6%|▌ | 28/500 [05:41<20:56, 2.66s/it, loss=0.879, lr=5.4e-5] Steps: 6%|▌ | 28/500 [05:41<20:56, 2.66s/it, loss=0.977, lr=5.6e-5] Steps: 6%|▌ | 29/500 [05:49<32:28, 4.14s/it, loss=0.977, lr=5.6e-5] Steps: 6%|▌ | 29/500 [05:49<32:28, 4.14s/it, loss=0.881, lr=5.8e-5] Steps: 6%|▌ | 30/500 [05:50<27:04, 3.46s/it, loss=0.881, lr=5.8e-5] Steps: 6%|▌ | 30/500 [05:50<27:04, 3.46s/it, loss=1.06, lr=6e-5] Steps: 6%|▌ | 31/500 [05:52<23:18, 2.98s/it, loss=1.06, lr=6e-5] Steps: 6%|▌ | 31/500 [05:52<23:18, 2.98s/it, loss=1.05, lr=6.2e-5] Steps: 6%|▋ | 32/500 [05:54<20:39, 2.65s/it, loss=1.05, lr=6.2e-5] Steps: 6%|▋ | 32/500 [05:54<20:39, 2.65s/it, loss=0.985, lr=6.4e-5] Steps: 7%|▋ | 33/500 [06:02<32:21, 4.16s/it, loss=0.985, lr=6.4e-5] Steps: 7%|▋ | 33/500 [06:02<32:21, 4.16s/it, loss=0.871, lr=6.6e-5] Steps: 7%|▋ | 34/500 [06:04<26:57, 3.47s/it, loss=0.871, lr=6.6e-5] Steps: 7%|▋ | 34/500 [06:04<26:57, 3.47s/it, loss=1.04, lr=6.8e-5] Steps: 7%|▋ | 35/500 [06:06<23:10, 2.99s/it, loss=1.04, lr=6.8e-5] Steps: 7%|▋ | 35/500 [06:06<23:10, 2.99s/it, loss=0.829, lr=7e-5] Steps: 7%|▋ | 36/500 [06:08<20:32, 2.66s/it, loss=0.829, lr=7e-5] Steps: 7%|▋ | 36/500 [06:08<20:32, 2.66s/it, loss=0.963, lr=7.2e-5] Steps: 7%|▋ | 37/500 [06:15<32:12, 4.17s/it, loss=0.963, lr=7.2e-5] Steps: 7%|▋ | 37/500 [06:15<32:12, 4.17s/it, loss=0.878, lr=7.4e-5] Steps: 8%|▊ | 38/500 [06:17<26:49, 3.48s/it, loss=0.878, lr=7.4e-5] Steps: 8%|▊ | 38/500 [06:17<26:49, 3.48s/it, loss=1.03, lr=7.6e-5] Steps: 8%|▊ | 39/500 [06:19<23:02, 3.00s/it, loss=1.03, lr=7.6e-5] Steps: 8%|▊ | 39/500 [06:19<23:02, 3.00s/it, loss=0.886, lr=7.8e-5] Steps: 8%|▊ | 40/500 [06:21<20:24, 2.66s/it, loss=0.886, lr=7.8e-5] Steps: 8%|▊ | 40/500 [06:21<20:24, 2.66s/it, loss=1.06, lr=8e-5] Steps: 8%|▊ | 41/500 [06:28<31:38, 4.14s/it, loss=1.06, lr=8e-5] Steps: 8%|▊ | 41/500 [06:28<31:38, 4.14s/it, loss=0.874, lr=8.2e-5] Steps: 8%|▊ | 42/500 [06:30<26:23, 3.46s/it, loss=0.874, lr=8.2e-5] Steps: 8%|▊ | 42/500 [06:30<26:23, 3.46s/it, loss=1.07, lr=8.4e-5] Steps: 9%|▊ | 43/500 [06:32<22:43, 2.98s/it, loss=1.07, lr=8.4e-5] Steps: 9%|▊ | 43/500 [06:32<22:43, 2.98s/it, loss=0.911, lr=8.6e-5] Steps: 9%|▉ | 44/500 [06:34<20:07, 2.65s/it, loss=0.911, lr=8.6e-5] Steps: 9%|▉ | 44/500 [06:34<20:07, 2.65s/it, loss=1.05, lr=8.8e-5] Steps: 9%|▉ | 45/500 [06:42<31:30, 4.15s/it, loss=1.05, lr=8.8e-5] Steps: 9%|▉ | 45/500 [06:42<31:30, 4.15s/it, loss=0.874, lr=9e-5] Steps: 9%|▉ | 46/500 [06:44<26:15, 3.47s/it, loss=0.874, lr=9e-5] Steps: 9%|▉ | 46/500 [06:44<26:15, 3.47s/it, loss=1.06, lr=9.2e-5] Steps: 9%|▉ | 47/500 [06:45<22:34, 2.99s/it, loss=1.06, lr=9.2e-5] Steps: 9%|▉ | 47/500 [06:45<22:34, 2.99s/it, loss=0.833, lr=9.4e-5] Steps: 10%|▉ | 48/500 [06:47<20:00, 2.66s/it, loss=0.833, lr=9.4e-5] Steps: 10%|▉ | 48/500 [06:47<20:00, 2.66s/it, loss=0.973, lr=9.6e-5] Steps: 10%|▉ | 49/500 [06:55<31:31, 4.19s/it, loss=0.973, lr=9.6e-5] Steps: 10%|▉ | 49/500 [06:55<31:31, 4.19s/it, loss=0.883, lr=9.8e-5] Steps: 10%|█ | 50/500 [06:57<26:13, 3.50s/it, loss=0.883, lr=9.8e-5] Steps: 10%|█ | 50/500 [06:57<26:13, 3.50s/it, loss=1.08, lr=0.0001] Steps: 10%|█ | 51/500 [06:59<22:31, 3.01s/it, loss=1.08, lr=0.0001] Steps: 10%|█ | 51/500 [06:59<22:31, 3.01s/it, loss=0.826, lr=0.000102] Steps: 10%|█ | 52/500 [07:01<19:56, 2.67s/it, loss=0.826, lr=0.000102] Steps: 10%|█ | 52/500 [07:01<19:56, 2.67s/it, loss=0.939, lr=0.000104] Steps: 11%|█ | 53/500 [07:11<37:37, 5.05s/it, loss=0.939, lr=0.000104] Steps: 11%|█ | 53/500 [07:11<37:37, 5.05s/it, loss=0.789, lr=0.000106] Steps: 11%|█ | 54/500 [07:13<30:27, 4.10s/it, loss=0.789, lr=0.000106] Steps: 11%|█ | 54/500 [07:13<30:27, 4.10s/it, loss=1.05, lr=0.000108] Steps: 11%|█ | 55/500 [07:15<25:25, 3.43s/it, loss=1.05, lr=0.000108] Steps: 11%|█ | 55/500 [07:15<25:25, 3.43s/it, loss=1.05, lr=0.00011] Steps: 11%|█ | 56/500 [07:17<21:55, 2.96s/it, loss=1.05, lr=0.00011] Steps: 11%|█ | 56/500 [07:17<21:55, 2.96s/it, loss=0.958, lr=0.000112] Steps: 11%|█▏ | 57/500 [07:24<31:59, 4.33s/it, loss=0.958, lr=0.000112] Steps: 11%|█▏ | 57/500 [07:24<31:59, 4.33s/it, loss=0.842, lr=0.000114] Steps: 12%|█▏ | 58/500 [07:26<26:28, 3.59s/it, loss=0.842, lr=0.000114] Steps: 12%|█▏ | 58/500 [07:26<26:28, 3.59s/it, loss=0.939, lr=0.000116] Steps: 12%|█▏ | 59/500 [07:28<22:36, 3.08s/it, loss=0.939, lr=0.000116] Steps: 12%|█▏ | 59/500 [07:28<22:36, 3.08s/it, loss=0.882, lr=0.000118] Steps: 12%|█▏ | 60/500 [07:30<19:54, 2.71s/it, loss=0.882, lr=0.000118] Steps: 12%|█▏ | 60/500 [07:30<19:54, 2.71s/it, loss=0.952, lr=0.00012] Steps: 12%|█▏ | 61/500 [07:38<30:27, 4.16s/it, loss=0.952, lr=0.00012] Steps: 12%|█▏ | 61/500 [07:38<30:27, 4.16s/it, loss=1.05, lr=0.000122] Steps: 12%|█▏ | 62/500 [07:40<25:22, 3.48s/it, loss=1.05, lr=0.000122] Steps: 12%|█▏ | 62/500 [07:40<25:22, 3.48s/it, loss=0.985, lr=0.000124] Steps: 13%|█▎ | 63/500 [07:41<21:48, 3.00s/it, loss=0.985, lr=0.000124] Steps: 13%|█▎ | 63/500 [07:41<21:48, 3.00s/it, loss=0.816, lr=0.000126] Steps: 13%|█▎ | 64/500 [07:43<19:18, 2.66s/it, loss=0.816, lr=0.000126] Steps: 13%|█▎ | 64/500 [07:43<19:18, 2.66s/it, loss=1.02, lr=0.000128] Steps: 13%|█▎ | 65/500 [07:51<30:04, 4.15s/it, loss=1.02, lr=0.000128] Steps: 13%|█▎ | 65/500 [07:51<30:04, 4.15s/it, loss=0.855, lr=0.00013] Steps: 13%|█▎ | 66/500 [07:53<25:03, 3.47s/it, loss=0.855, lr=0.00013] Steps: 13%|█▎ | 66/500 [07:53<25:03, 3.47s/it, loss=0.947, lr=0.000132] Steps: 13%|█▎ | 67/500 [07:55<21:33, 2.99s/it, loss=0.947, lr=0.000132] Steps: 13%|█▎ | 67/500 [07:55<21:33, 2.99s/it, loss=0.879, lr=0.000134] Steps: 14%|█▎ | 68/500 [07:56<19:05, 2.65s/it, loss=0.879, lr=0.000134] Steps: 14%|█▎ | 68/500 [07:57<19:05, 2.65s/it, loss=1.06, lr=0.000136] Steps: 14%|█▍ | 69/500 [08:04<30:06, 4.19s/it, loss=1.06, lr=0.000136] Steps: 14%|█▍ | 69/500 [08:04<30:06, 4.19s/it, loss=0.825, lr=0.000138] Steps: 14%|█▍ | 70/500 [08:06<25:02, 3.50s/it, loss=0.825, lr=0.000138] Steps: 14%|█▍ | 70/500 [08:06<25:02, 3.50s/it, loss=0.924, lr=0.00014] Steps: 14%|█▍ | 71/500 [08:08<21:30, 3.01s/it, loss=0.924, lr=0.00014] Steps: 14%|█▍ | 71/500 [08:08<21:30, 3.01s/it, loss=0.794, lr=0.000142] Steps: 14%|█▍ | 72/500 [08:10<19:01, 2.67s/it, loss=0.794, lr=0.000142] Steps: 14%|█▍ | 72/500 [08:10<19:01, 2.67s/it, loss=0.978, lr=0.000144] Steps: 15%|█▍ | 73/500 [08:18<29:45, 4.18s/it, loss=0.978, lr=0.000144] Steps: 15%|█▍ | 73/500 [08:18<29:45, 4.18s/it, loss=0.996, lr=0.000146] Steps: 15%|█▍ | 74/500 [08:19<24:46, 3.49s/it, loss=0.996, lr=0.000146] Steps: 15%|█▍ | 74/500 [08:19<24:46, 3.49s/it, loss=1.07, lr=0.000148] Steps: 15%|█▌ | 75/500 [08:21<21:16, 3.00s/it, loss=1.07, lr=0.000148] Steps: 15%|█▌ | 75/500 [08:21<21:16, 3.00s/it, loss=0.842, lr=0.00015] Steps: 15%|█▌ | 76/500 [08:23<18:50, 2.67s/it, loss=0.842, lr=0.00015] Steps: 15%|█▌ | 76/500 [08:23<18:50, 2.67s/it, loss=0.946, lr=0.000152] Steps: 15%|█▌ | 77/500 [08:31<29:24, 4.17s/it, loss=0.946, lr=0.000152] Steps: 15%|█▌ | 77/500 [08:31<29:24, 4.17s/it, loss=0.838, lr=0.000154] Steps: 16%|█▌ | 78/500 [08:33<24:29, 3.48s/it, loss=0.838, lr=0.000154] Steps: 16%|█▌ | 78/500 [08:33<24:29, 3.48s/it, loss=1.06, lr=0.000156] Steps: 16%|█▌ | 79/500 [08:35<21:02, 3.00s/it, loss=1.06, lr=0.000156] Steps: 16%|█▌ | 79/500 [08:35<21:02, 3.00s/it, loss=0.85, lr=0.000158] Steps: 16%|█▌ | 80/500 [08:37<18:37, 2.66s/it, loss=0.85, lr=0.000158] Steps: 16%|█▌ | 80/500 [08:37<18:37, 2.66s/it, loss=0.923, lr=0.00016] Steps: 16%|█▌ | 81/500 [08:44<29:10, 4.18s/it, loss=0.923, lr=0.00016] Steps: 16%|█▌ | 81/500 [08:44<29:10, 4.18s/it, loss=0.764, lr=0.000162] Steps: 16%|█▋ | 82/500 [08:46<24:17, 3.49s/it, loss=0.764, lr=0.000162] Steps: 16%|█▋ | 82/500 [08:46<24:17, 3.49s/it, loss=0.94, lr=0.000164] Steps: 17%|█▋ | 83/500 [08:48<20:51, 3.00s/it, loss=0.94, lr=0.000164] Steps: 17%|█▋ | 83/500 [08:48<20:51, 3.00s/it, loss=0.828, lr=0.000166] Steps: 17%|█▋ | 84/500 [08:50<18:27, 2.66s/it, loss=0.828, lr=0.000166] Steps: 17%|█▋ | 84/500 [08:50<18:27, 2.66s/it, loss=1.02, lr=0.000168] Steps: 17%|█▋ | 85/500 [08:58<28:59, 4.19s/it, loss=1.02, lr=0.000168] Steps: 17%|█▋ | 85/500 [08:58<28:59, 4.19s/it, loss=0.991, lr=0.00017] Steps: 17%|█▋ | 86/500 [08:59<24:07, 3.50s/it, loss=0.991, lr=0.00017] Steps: 17%|█▋ | 86/500 [09:00<24:07, 3.50s/it, loss=0.975, lr=0.000172] Steps: 17%|█▋ | 87/500 [09:01<20:42, 3.01s/it, loss=0.975, lr=0.000172] Steps: 17%|█▋ | 87/500 [09:01<20:42, 3.01s/it, loss=0.814, lr=0.000174] Steps: 18%|█▊ | 88/500 [09:03<18:19, 2.67s/it, loss=0.814, lr=0.000174] Steps: 18%|█▊ | 88/500 [09:03<18:19, 2.67s/it, loss=1.07, lr=0.000176] Steps: 18%|█▊ | 89/500 [09:11<28:25, 4.15s/it, loss=1.07, lr=0.000176] Steps: 18%|█▊ | 89/500 [09:11<28:25, 4.15s/it, loss=0.859, lr=0.000178] Steps: 18%|█▊ | 90/500 [09:13<23:41, 3.47s/it, loss=0.859, lr=0.000178] Steps: 18%|█▊ | 90/500 [09:13<23:41, 3.47s/it, loss=1.06, lr=0.00018] Steps: 18%|█▊ | 91/500 [09:15<20:21, 2.99s/it, loss=1.06, lr=0.00018] Steps: 18%|█▊ | 91/500 [09:15<20:21, 2.99s/it, loss=0.825, lr=0.000182] Steps: 18%|█▊ | 92/500 [09:16<18:02, 2.65s/it, loss=0.825, lr=0.000182] Steps: 18%|█▊ | 92/500 [09:16<18:02, 2.65s/it, loss=0.954, lr=0.000184] Steps: 19%|█▊ | 93/500 [09:24<28:00, 4.13s/it, loss=0.954, lr=0.000184] Steps: 19%|█▊ | 93/500 [09:24<28:00, 4.13s/it, loss=0.852, lr=0.000186] Steps: 19%|█▉ | 94/500 [09:26<23:21, 3.45s/it, loss=0.852, lr=0.000186] Steps: 19%|█▉ | 94/500 [09:26<23:21, 3.45s/it, loss=1.04, lr=0.000188] Steps: 19%|█▉ | 95/500 [09:28<20:06, 2.98s/it, loss=1.04, lr=0.000188] Steps: 19%|█▉ | 95/500 [09:28<20:06, 2.98s/it, loss=0.847, lr=0.00019] Steps: 19%|█▉ | 96/500 [09:30<17:49, 2.65s/it, loss=0.847, lr=0.00019] Steps: 19%|█▉ | 96/500 [09:30<17:49, 2.65s/it, loss=0.921, lr=0.000192] Steps: 19%|█▉ | 97/500 [09:37<27:56, 4.16s/it, loss=0.921, lr=0.000192] Steps: 19%|█▉ | 97/500 [09:37<27:56, 4.16s/it, loss=0.873, lr=0.000194] Steps: 20%|█▉ | 98/500 [09:39<23:16, 3.47s/it, loss=0.873, lr=0.000194] Steps: 20%|█▉ | 98/500 [09:39<23:16, 3.47s/it, loss=0.977, lr=0.000196] Steps: 20%|█▉ | 99/500 [09:41<20:00, 2.99s/it, loss=0.977, lr=0.000196] Steps: 20%|█▉ | 99/500 [09:41<20:00, 2.99s/it, loss=0.851, lr=0.000198] Steps: 20%|██ | 100/500 [09:43<17:44, 2.66s/it, loss=0.851, lr=0.000198] Steps: 20%|██ | 100/500 [09:43<17:44, 2.66s/it, loss=0.918, lr=0.0002] Steps: 20%|██ | 101/500 [09:51<28:24, 4.27s/it, loss=0.918, lr=0.0002] Steps: 20%|██ | 101/500 [09:51<28:24, 4.27s/it, loss=0.809, lr=0.000202] Steps: 20%|██ | 102/500 [09:53<23:33, 3.55s/it, loss=0.809, lr=0.000202] Steps: 20%|██ | 102/500 [09:53<23:33, 3.55s/it, loss=0.916, lr=0.000204] Steps: 21%|██ | 103/500 [09:55<20:10, 3.05s/it, loss=0.916, lr=0.000204] Steps: 21%|██ | 103/500 [09:55<20:10, 3.05s/it, loss=1.01, lr=0.000206] Steps: 21%|██ | 104/500 [09:57<17:48, 2.70s/it, loss=1.01, lr=0.000206] Steps: 21%|██ | 104/500 [09:57<17:48, 2.70s/it, loss=0.958, lr=0.000208] Steps: 21%|██ | 105/500 [10:05<28:03, 4.26s/it, loss=0.958, lr=0.000208] Steps: 21%|██ | 105/500 [10:05<28:03, 4.26s/it, loss=0.807, lr=0.00021] Steps: 21%|██ | 106/500 [10:06<23:16, 3.55s/it, loss=0.807, lr=0.00021] Steps: 21%|██ | 106/500 [10:06<23:16, 3.55s/it, loss=0.953, lr=0.000212] Steps: 21%|██▏ | 107/500 [10:08<19:56, 3.04s/it, loss=0.953, lr=0.000212] Steps: 21%|██▏ | 107/500 [10:08<19:56, 3.04s/it, loss=0.826, lr=0.000214] Steps: 22%|██▏ | 108/500 [10:10<17:35, 2.69s/it, loss=0.826, lr=0.000214] Steps: 22%|██▏ | 108/500 [10:10<17:35, 2.69s/it, loss=1.08, lr=0.000216] Steps: 22%|██▏ | 109/500 [10:18<27:22, 4.20s/it, loss=1.08, lr=0.000216] Steps: 22%|██▏ | 109/500 [10:18<27:22, 4.20s/it, loss=0.836, lr=0.000218] Steps: 22%|██▏ | 110/500 [10:20<22:46, 3.50s/it, loss=0.836, lr=0.000218] Steps: 22%|██▏ | 110/500 [10:20<22:46, 3.50s/it, loss=1.07, lr=0.00022] Steps: 22%|██▏ | 111/500 [10:22<19:32, 3.01s/it, loss=1.07, lr=0.00022] Steps: 22%|██▏ | 111/500 [10:22<19:32, 3.01s/it, loss=0.824, lr=0.000222] Steps: 22%|██▏ | 112/500 [10:24<17:16, 2.67s/it, loss=0.824, lr=0.000222] Steps: 22%|██▏ | 112/500 [10:24<17:16, 2.67s/it, loss=0.916, lr=0.000224] Steps: 23%|██▎ | 113/500 [10:31<26:58, 4.18s/it, loss=0.916, lr=0.000224] Steps: 23%|██▎ | 113/500 [10:31<26:58, 4.18s/it, loss=0.793, lr=0.000226] Steps: 23%|██▎ | 114/500 [10:33<22:27, 3.49s/it, loss=0.793, lr=0.000226] Steps: 23%|██▎ | 114/500 [10:33<22:27, 3.49s/it, loss=0.927, lr=0.000228] Steps: 23%|██▎ | 115/500 [10:35<19:17, 3.01s/it, loss=0.927, lr=0.000228] Steps: 23%|██▎ | 115/500 [10:35<19:17, 3.01s/it, loss=0.924, lr=0.00023] Steps: 23%|██▎ | 116/500 [10:37<17:03, 2.67s/it, loss=0.924, lr=0.00023] Steps: 23%|██▎ | 116/500 [10:37<17:03, 2.67s/it, loss=1.04, lr=0.000232] Steps: 23%|██▎ | 117/500 [10:44<26:32, 4.16s/it, loss=1.04, lr=0.000232] Steps: 23%|██▎ | 117/500 [10:44<26:32, 4.16s/it, loss=0.857, lr=0.000234] Steps: 24%|██▎ | 118/500 [10:46<22:06, 3.47s/it, loss=0.857, lr=0.000234] Steps: 24%|██▎ | 118/500 [10:46<22:06, 3.47s/it, loss=0.91, lr=0.000236] Steps: 24%|██▍ | 119/500 [10:48<19:00, 2.99s/it, loss=0.91, lr=0.000236] Steps: 24%|██▍ | 119/500 [10:48<19:00, 2.99s/it, loss=0.781, lr=0.000238] Steps: 24%|██▍ | 120/500 [10:50<16:49, 2.66s/it, loss=0.781, lr=0.000238] Steps: 24%|██▍ | 120/500 [10:50<16:49, 2.66s/it, loss=0.937, lr=0.00024] Steps: 24%|██▍ | 121/500 [10:58<26:42, 4.23s/it, loss=0.937, lr=0.00024] Steps: 24%|██▍ | 121/500 [10:58<26:42, 4.23s/it, loss=0.876, lr=0.000242] Steps: 24%|██▍ | 122/500 [11:00<22:10, 3.52s/it, loss=0.876, lr=0.000242] Steps: 24%|██▍ | 122/500 [11:00<22:10, 3.52s/it, loss=0.971, lr=0.000244] Steps: 25%|██▍ | 123/500 [11:02<19:00, 3.03s/it, loss=0.971, lr=0.000244] Steps: 25%|██▍ | 123/500 [11:02<19:00, 3.03s/it, loss=0.812, lr=0.000246] Steps: 25%|██▍ | 124/500 [11:04<16:47, 2.68s/it, loss=0.812, lr=0.000246] Steps: 25%|██▍ | 124/500 [11:04<16:47, 2.68s/it, loss=1, lr=0.000248] Steps: 25%|██▌ | 125/500 [11:11<26:07, 4.18s/it, loss=1, lr=0.000248] Steps: 25%|██▌ | 125/500 [11:11<26:07, 4.18s/it, loss=0.97, lr=0.00025] Steps: 25%|██▌ | 126/500 [11:13<21:44, 3.49s/it, loss=0.97, lr=0.00025] Steps: 25%|██▌ | 126/500 [11:13<21:44, 3.49s/it, loss=1.07, lr=0.000252] Steps: 25%|██▌ | 127/500 [11:15<18:40, 3.00s/it, loss=1.07, lr=0.000252] Steps: 25%|██▌ | 127/500 [11:15<18:40, 3.00s/it, loss=0.814, lr=0.000254] Steps: 26%|██▌ | 128/500 [11:17<16:31, 2.67s/it, loss=0.814, lr=0.000254] Steps: 26%|██▌ | 128/500 [11:17<16:31, 2.67s/it, loss=0.904, lr=0.000256] Steps: 26%|██▌ | 129/500 [11:25<25:55, 4.19s/it, loss=0.904, lr=0.000256] Steps: 26%|██▌ | 129/500 [11:25<25:55, 4.19s/it, loss=0.885, lr=0.000258] Steps: 26%|██▌ | 130/500 [11:27<21:33, 3.50s/it, loss=0.885, lr=0.000258] Steps: 26%|██▌ | 130/500 [11:27<21:33, 3.50s/it, loss=0.923, lr=0.00026] Steps: 26%|██▌ | 131/500 [11:28<18:30, 3.01s/it, loss=0.923, lr=0.00026] Steps: 26%|██▌ | 131/500 [11:28<18:30, 3.01s/it, loss=0.812, lr=0.000262] Steps: 26%|██▋ | 132/500 [11:30<16:21, 2.67s/it, loss=0.812, lr=0.000262] Steps: 26%|██▋ | 132/500 [11:30<16:21, 2.67s/it, loss=0.986, lr=0.000264] Steps: 27%|██▋ | 133/500 [11:38<25:53, 4.23s/it, loss=0.986, lr=0.000264] Steps: 27%|██▋ | 133/500 [11:38<25:53, 4.23s/it, loss=0.823, lr=0.000266] Steps: 27%|██▋ | 134/500 [11:40<21:31, 3.53s/it, loss=0.823, lr=0.000266] Steps: 27%|██▋ | 134/500 [11:40<21:31, 3.53s/it, loss=1.06, lr=0.000268] Steps: 27%|██▋ | 135/500 [11:42<18:26, 3.03s/it, loss=1.06, lr=0.000268] Steps: 27%|██▋ | 135/500 [11:42<18:26, 3.03s/it, loss=1.07, lr=0.00027] Steps: 27%|██▋ | 136/500 [11:44<16:17, 2.69s/it, loss=1.07, lr=0.00027] Steps: 27%|██▋ | 136/500 [11:44<16:17, 2.69s/it, loss=0.961, lr=0.000272] Steps: 27%|██▋ | 137/500 [11:51<25:19, 4.19s/it, loss=0.961, lr=0.000272] Steps: 27%|██▋ | 137/500 [11:52<25:19, 4.19s/it, loss=0.842, lr=0.000274] Steps: 28%|██▊ | 138/500 [11:53<21:04, 3.49s/it, loss=0.842, lr=0.000274] Steps: 28%|██▊ | 138/500 [11:53<21:04, 3.49s/it, loss=0.952, lr=0.000276] Steps: 28%|██▊ | 139/500 [11:55<18:05, 3.01s/it, loss=0.952, lr=0.000276] Steps: 28%|██▊ | 139/500 [11:55<18:05, 3.01s/it, loss=0.901, lr=0.000278] Steps: 28%|██▊ | 140/500 [11:57<16:00, 2.67s/it, loss=0.901, lr=0.000278] Steps: 28%|██▊ | 140/500 [11:57<16:00, 2.67s/it, loss=0.926, lr=0.00028] Steps: 28%|██▊ | 141/500 [12:05<25:06, 4.20s/it, loss=0.926, lr=0.00028] Steps: 28%|██▊ | 141/500 [12:05<25:06, 4.20s/it, loss=0.808, lr=0.000282] Steps: 28%|██▊ | 142/500 [12:07<20:52, 3.50s/it, loss=0.808, lr=0.000282] Steps: 28%|██▊ | 142/500 [12:07<20:52, 3.50s/it, loss=0.926, lr=0.000284] Steps: 29%|██▊ | 143/500 [12:09<17:55, 3.01s/it, loss=0.926, lr=0.000284] Steps: 29%|██▊ | 143/500 [12:09<17:55, 3.01s/it, loss=0.951, lr=0.000286] Steps: 29%|██▉ | 144/500 [12:11<15:51, 2.67s/it, loss=0.951, lr=0.000286] Steps: 29%|██▉ | 144/500 [12:11<15:51, 2.67s/it, loss=0.911, lr=0.000288] Steps: 29%|██▉ | 145/500 [12:18<24:43, 4.18s/it, loss=0.911, lr=0.000288] Steps: 29%|██▉ | 145/500 [12:18<24:43, 4.18s/it, loss=0.806, lr=0.00029] Steps: 29%|██▉ | 146/500 [12:20<20:34, 3.49s/it, loss=0.806, lr=0.00029] Steps: 29%|██▉ | 146/500 [12:20<20:34, 3.49s/it, loss=0.901, lr=0.000292] Steps: 29%|██▉ | 147/500 [12:22<17:40, 3.00s/it, loss=0.901, lr=0.000292] Steps: 29%|██▉ | 147/500 [12:22<17:40, 3.00s/it, loss=0.847, lr=0.000294] Steps: 30%|██▉ | 148/500 [12:24<15:38, 2.67s/it, loss=0.847, lr=0.000294] Steps: 30%|██▉ | 148/500 [12:24<15:38, 2.67s/it, loss=0.963, lr=0.000296] Steps: 30%|██▉ | 149/500 [12:32<24:33, 4.20s/it, loss=0.963, lr=0.000296] Steps: 30%|██▉ | 149/500 [12:32<24:33, 4.20s/it, loss=1, lr=0.000298] Steps: 30%|███ | 150/500 [12:33<20:24, 3.50s/it, loss=1, lr=0.000298] Steps: 30%|███ | 150/500 [12:33<20:24, 3.50s/it, loss=0.897, lr=0.0003] Steps: 30%|███ | 151/500 [12:35<17:30, 3.01s/it, loss=0.897, lr=0.0003] Steps: 30%|███ | 151/500 [12:35<17:30, 3.01s/it, loss=0.842, lr=0.000302] Steps: 30%|███ | 152/500 [12:37<15:28, 2.67s/it, loss=0.842, lr=0.000302] Steps: 30%|███ | 152/500 [12:37<15:28, 2.67s/it, loss=1.07, lr=0.000304] Steps: 31%|███ | 153/500 [12:45<24:13, 4.19s/it, loss=1.07, lr=0.000304] Steps: 31%|███ | 153/500 [12:45<24:13, 4.19s/it, loss=0.861, lr=0.000306] Steps: 31%|███ | 154/500 [12:47<20:09, 3.49s/it, loss=0.861, lr=0.000306] Steps: 31%|███ | 154/500 [12:47<20:09, 3.49s/it, loss=0.903, lr=0.000308] Steps: 31%|███ | 155/500 [12:49<17:17, 3.01s/it, loss=0.903, lr=0.000308] Steps: 31%|███ | 155/500 [12:49<17:17, 3.01s/it, loss=0.86, lr=0.00031] Steps: 31%|███ | 156/500 [12:51<15:18, 2.67s/it, loss=0.86, lr=0.00031] Steps: 31%|███ | 156/500 [12:51<15:18, 2.67s/it, loss=0.904, lr=0.000312] Steps: 31%|███▏ | 157/500 [12:58<24:02, 4.21s/it, loss=0.904, lr=0.000312] Steps: 31%|███▏ | 157/500 [12:58<24:02, 4.21s/it, loss=1.05, lr=0.000314] Steps: 32%|███▏ | 158/500 [13:00<19:58, 3.51s/it, loss=1.05, lr=0.000314] Steps: 32%|███▏ | 158/500 [13:00<19:58, 3.51s/it, loss=1.02, lr=0.000316] Steps: 32%|███▏ | 159/500 [13:02<17:08, 3.02s/it, loss=1.02, lr=0.000316] Steps: 32%|███▏ | 159/500 [13:02<17:08, 3.02s/it, loss=0.964, lr=0.000318] Steps: 32%|███▏ | 160/500 [13:04<15:08, 2.67s/it, loss=0.964, lr=0.000318] Steps: 32%|███▏ | 160/500 [13:04<15:08, 2.67s/it, loss=0.909, lr=0.00032] Steps: 32%|███▏ | 161/500 [13:12<23:30, 4.16s/it, loss=0.909, lr=0.00032] Steps: 32%|███▏ | 161/500 [13:12<23:30, 4.16s/it, loss=0.874, lr=0.000322] Steps: 32%|███▏ | 162/500 [13:13<19:34, 3.47s/it, loss=0.874, lr=0.000322] Steps: 32%|███▏ | 162/500 [13:14<19:34, 3.47s/it, loss=0.932, lr=0.000324] Steps: 33%|███▎ | 163/500 [13:15<16:49, 2.99s/it, loss=0.932, lr=0.000324] Steps: 33%|███▎ | 163/500 [13:15<16:49, 2.99s/it, loss=0.917, lr=0.000326] Steps: 33%|███▎ | 164/500 [13:17<14:53, 2.66s/it, loss=0.917, lr=0.000326] Steps: 33%|███▎ | 164/500 [13:17<14:53, 2.66s/it, loss=1.07, lr=0.000328] Steps: 33%|███▎ | 165/500 [13:25<23:19, 4.18s/it, loss=1.07, lr=0.000328] Steps: 33%|███▎ | 165/500 [13:25<23:19, 4.18s/it, loss=0.855, lr=0.00033] Steps: 33%|███▎ | 166/500 [13:27<19:24, 3.49s/it, loss=0.855, lr=0.00033] Steps: 33%|███▎ | 166/500 [13:27<19:24, 3.49s/it, loss=0.986, lr=0.000332] Steps: 33%|███▎ | 167/500 [13:29<16:39, 3.00s/it, loss=0.986, lr=0.000332] Steps: 33%|███▎ | 167/500 [13:29<16:39, 3.00s/it, loss=0.814, lr=0.000334] Steps: 34%|███▎ | 168/500 [13:31<14:44, 2.66s/it, loss=0.814, lr=0.000334] Steps: 34%|███▎ | 168/500 [13:31<14:44, 2.66s/it, loss=0.92, lr=0.000336] Steps: 34%|███▍ | 169/500 [13:38<23:13, 4.21s/it, loss=0.92, lr=0.000336] Steps: 34%|███▍ | 169/500 [13:38<23:13, 4.21s/it, loss=0.835, lr=0.000338] Steps: 34%|███▍ | 170/500 [13:40<19:17, 3.51s/it, loss=0.835, lr=0.000338] Steps: 34%|███▍ | 170/500 [13:40<19:17, 3.51s/it, loss=1.08, lr=0.00034] Steps: 34%|███▍ | 171/500 [13:42<16:33, 3.02s/it, loss=1.08, lr=0.00034] Steps: 34%|███▍ | 171/500 [13:42<16:33, 3.02s/it, loss=0.988, lr=0.000342] Steps: 34%|███▍ | 172/500 [13:44<14:37, 2.68s/it, loss=0.988, lr=0.000342] Steps: 34%|███▍ | 172/500 [13:44<14:37, 2.68s/it, loss=1, lr=0.000344] Steps: 35%|███▍ | 173/500 [13:52<22:40, 4.16s/it, loss=1, lr=0.000344] Steps: 35%|███▍ | 173/500 [13:52<22:40, 4.16s/it, loss=1.04, lr=0.000346] Steps: 35%|███▍ | 174/500 [13:54<18:52, 3.47s/it, loss=1.04, lr=0.000346] Steps: 35%|███▍ | 174/500 [13:54<18:52, 3.47s/it, loss=1.05, lr=0.000348] Steps: 35%|███▌ | 175/500 [13:55<16:13, 2.99s/it, loss=1.05, lr=0.000348] Steps: 35%|███▌ | 175/500 [13:55<16:13, 2.99s/it, loss=0.996, lr=0.00035] Steps: 35%|███▌ | 176/500 [13:57<14:20, 2.66s/it, loss=0.996, lr=0.00035] Steps: 35%|███▌ | 176/500 [13:57<14:20, 2.66s/it, loss=1.06, lr=0.000352] Steps: 35%|███▌ | 177/500 [14:05<22:27, 4.17s/it, loss=1.06, lr=0.000352] Steps: 35%|███▌ | 177/500 [14:05<22:27, 4.17s/it, loss=0.994, lr=0.000354] Steps: 36%|███▌ | 178/500 [14:07<18:41, 3.48s/it, loss=0.994, lr=0.000354] Steps: 36%|███▌ | 178/500 [14:07<18:41, 3.48s/it, loss=0.987, lr=0.000356] Steps: 36%|███▌ | 179/500 [14:09<16:02, 3.00s/it, loss=0.987, lr=0.000356] Steps: 36%|███▌ | 179/500 [14:09<16:02, 3.00s/it, loss=0.81, lr=0.000358] Steps: 36%|███▌ | 180/500 [14:11<14:11, 2.66s/it, loss=0.81, lr=0.000358] Steps: 36%|███▌ | 180/500 [14:11<14:11, 2.66s/it, loss=0.944, lr=0.00036] Steps: 36%|███▌ | 181/500 [14:18<22:05, 4.16s/it, loss=0.944, lr=0.00036] Steps: 36%|███▌ | 181/500 [14:18<22:05, 4.16s/it, loss=0.856, lr=0.000362] Steps: 36%|███▋ | 182/500 [14:20<18:23, 3.47s/it, loss=0.856, lr=0.000362] Steps: 36%|███▋ | 182/500 [14:20<18:23, 3.47s/it, loss=0.956, lr=0.000364] Steps: 37%|███▋ | 183/500 [14:22<15:48, 2.99s/it, loss=0.956, lr=0.000364] Steps: 37%|███▋ | 183/500 [14:22<15:48, 2.99s/it, loss=0.823, lr=0.000366] Steps: 37%|███▋ | 184/500 [14:24<13:59, 2.66s/it, loss=0.823, lr=0.000366] Steps: 37%|███▋ | 184/500 [14:24<13:59, 2.66s/it, loss=0.963, lr=0.000368] Steps: 37%|███▋ | 185/500 [14:31<21:45, 4.15s/it, loss=0.963, lr=0.000368] Steps: 37%|███▋ | 185/500 [14:31<21:45, 4.15s/it, loss=0.971, lr=0.00037] Steps: 37%|███▋ | 186/500 [14:33<18:07, 3.46s/it, loss=0.971, lr=0.00037] Steps: 37%|███▋ | 186/500 [14:33<18:07, 3.46s/it, loss=1.01, lr=0.000372] Steps: 37%|███▋ | 187/500 [14:35<15:34, 2.99s/it, loss=1.01, lr=0.000372] Steps: 37%|███▋ | 187/500 [14:35<15:34, 2.99s/it, loss=0.855, lr=0.000374] Steps: 38%|███▊ | 188/500 [14:37<13:47, 2.65s/it, loss=0.855, lr=0.000374] Steps: 38%|███▊ | 188/500 [14:37<13:47, 2.65s/it, loss=1.06, lr=0.000376] Steps: 38%|███▊ | 189/500 [14:45<21:33, 4.16s/it, loss=1.06, lr=0.000376] Steps: 38%|███▊ | 189/500 [14:45<21:33, 4.16s/it, loss=0.906, lr=0.000378] Steps: 38%|███▊ | 190/500 [14:47<17:56, 3.47s/it, loss=0.906, lr=0.000378] Steps: 38%|███▊ | 190/500 [14:47<17:56, 3.47s/it, loss=0.957, lr=0.00038] Steps: 38%|███▊ | 191/500 [14:49<15:24, 2.99s/it, loss=0.957, lr=0.00038] Steps: 38%|███▊ | 191/500 [14:49<15:24, 2.99s/it, loss=0.874, lr=0.000382] Steps: 38%|███▊ | 192/500 [14:50<13:38, 2.66s/it, loss=0.874, lr=0.000382] Steps: 38%|███▊ | 192/500 [14:50<13:38, 2.66s/it, loss=0.902, lr=0.000384] Steps: 39%|███▊ | 193/500 [14:58<21:24, 4.19s/it, loss=0.902, lr=0.000384] Steps: 39%|███▊ | 193/500 [14:58<21:24, 4.19s/it, loss=1.06, lr=0.000386] Steps: 39%|███▉ | 194/500 [15:00<17:48, 3.49s/it, loss=1.06, lr=0.000386] Steps: 39%|███▉ | 194/500 [15:00<17:48, 3.49s/it, loss=0.955, lr=0.000388] Steps: 39%|███▉ | 195/500 [15:02<15:16, 3.01s/it, loss=0.955, lr=0.000388] Steps: 39%|███▉ | 195/500 [15:02<15:16, 3.01s/it, loss=0.808, lr=0.00039] Steps: 39%|███▉ | 196/500 [15:04<13:30, 2.66s/it, loss=0.808, lr=0.00039] Steps: 39%|███▉ | 196/500 [15:04<13:30, 2.66s/it, loss=0.925, lr=0.000392] Steps: 39%|███▉ | 197/500 [15:12<21:19, 4.22s/it, loss=0.925, lr=0.000392] Steps: 39%|███▉ | 197/500 [15:12<21:19, 4.22s/it, loss=0.869, lr=0.000394] Steps: 40%|███▉ | 198/500 [15:13<17:42, 3.52s/it, loss=0.869, lr=0.000394] Steps: 40%|███▉ | 198/500 [15:13<17:42, 3.52s/it, loss=1.08, lr=0.000396] Steps: 40%|███▉ | 199/500 [15:15<15:10, 3.02s/it, loss=1.08, lr=0.000396] Steps: 40%|███▉ | 199/500 [15:15<15:10, 3.02s/it, loss=0.829, lr=0.000398] Steps: 40%|████ | 200/500 [15:17<13:23, 2.68s/it, loss=0.829, lr=0.000398] Steps: 40%|████ | 200/500 [15:17<13:23, 2.68s/it, loss=1.05, lr=0.0004] Steps: 40%|████ | 201/500 [15:25<20:55, 4.20s/it, loss=1.05, lr=0.0004] Steps: 40%|████ | 201/500 [15:25<20:55, 4.20s/it, loss=1.03, lr=0.0004] Steps: 40%|████ | 202/500 [15:27<17:23, 3.50s/it, loss=1.03, lr=0.0004] Steps: 40%|████ | 202/500 [15:27<17:23, 3.50s/it, loss=1.06, lr=0.0004] Steps: 41%|████ | 203/500 [15:29<14:54, 3.01s/it, loss=1.06, lr=0.0004] Steps: 41%|████ | 203/500 [15:29<14:54, 3.01s/it, loss=0.846, lr=0.0004] Steps: 41%|████ | 204/500 [15:31<13:10, 2.67s/it, loss=0.846, lr=0.0004] Steps: 41%|████ | 204/500 [15:31<13:10, 2.67s/it, loss=0.921, lr=0.0004] Steps: 41%|████ | 205/500 [15:38<20:37, 4.19s/it, loss=0.921, lr=0.0004] Steps: 41%|████ | 205/500 [15:38<20:37, 4.19s/it, loss=0.856, lr=0.0004] Steps: 41%|████ | 206/500 [15:40<17:08, 3.50s/it, loss=0.856, lr=0.0004] Steps: 41%|████ | 206/500 [15:40<17:08, 3.50s/it, loss=1.06, lr=0.0004] Steps: 41%|████▏ | 207/500 [15:42<14:41, 3.01s/it, loss=1.06, lr=0.0004] Steps: 41%|████▏ | 207/500 [15:42<14:41, 3.01s/it, loss=0.81, lr=0.000399] Steps: 42%|████▏ | 208/500 [15:44<12:59, 2.67s/it, loss=0.81, lr=0.000399] Steps: 42%|████▏ | 208/500 [15:44<12:59, 2.67s/it, loss=0.961, lr=0.000399] Steps: 42%|████▏ | 209/500 [15:52<20:15, 4.18s/it, loss=0.961, lr=0.000399] Steps: 42%|████▏ | 209/500 [15:52<20:15, 4.18s/it, loss=0.809, lr=0.000399] Steps: 42%|████▏ | 210/500 [15:54<16:50, 3.48s/it, loss=0.809, lr=0.000399] Steps: 42%|████▏ | 210/500 [15:54<16:50, 3.48s/it, loss=0.983, lr=0.000399] Steps: 42%|████▏ | 211/500 [15:55<14:27, 3.00s/it, loss=0.983, lr=0.000399] Steps: 42%|████▏ | 211/500 [15:55<14:27, 3.00s/it, loss=0.865, lr=0.000399] Steps: 42%|████▏ | 212/500 [15:57<12:46, 2.66s/it, loss=0.865, lr=0.000399] Steps: 42%|████▏ | 212/500 [15:57<12:46, 2.66s/it, loss=0.927, lr=0.000398] Steps: 43%|████▎ | 213/500 [16:05<19:55, 4.17s/it, loss=0.927, lr=0.000398] Steps: 43%|████▎ | 213/500 [16:05<19:55, 4.17s/it, loss=0.799, lr=0.000398] Steps: 43%|████▎ | 214/500 [16:07<16:34, 3.48s/it, loss=0.799, lr=0.000398] Steps: 43%|████▎ | 214/500 [16:07<16:34, 3.48s/it, loss=1.02, lr=0.000398] Steps: 43%|████▎ | 215/500 [16:09<14:13, 3.00s/it, loss=1.02, lr=0.000398] Steps: 43%|████▎ | 215/500 [16:09<14:13, 3.00s/it, loss=0.864, lr=0.000398] Steps: 43%|████▎ | 216/500 [16:11<12:34, 2.66s/it, loss=0.864, lr=0.000398] Steps: 43%|████▎ | 216/500 [16:11<12:34, 2.66s/it, loss=0.976, lr=0.000397] Steps: 43%|████▎ | 217/500 [16:18<19:35, 4.15s/it, loss=0.976, lr=0.000397] Steps: 43%|████▎ | 217/500 [16:18<19:35, 4.15s/it, loss=0.859, lr=0.000397] Steps: 44%|████▎ | 218/500 [16:20<16:18, 3.47s/it, loss=0.859, lr=0.000397] Steps: 44%|████▎ | 218/500 [16:20<16:18, 3.47s/it, loss=0.9, lr=0.000396] Steps: 44%|████▍ | 219/500 [16:22<14:00, 2.99s/it, loss=0.9, lr=0.000396] Steps: 44%|████▍ | 219/500 [16:22<14:00, 2.99s/it, loss=0.935, lr=0.000396] Steps: 44%|████▍ | 220/500 [16:24<12:23, 2.66s/it, loss=0.935, lr=0.000396] Steps: 44%|████▍ | 220/500 [16:24<12:23, 2.66s/it, loss=0.919, lr=0.000396] Steps: 44%|████▍ | 221/500 [16:31<19:15, 4.14s/it, loss=0.919, lr=0.000396] Steps: 44%|████▍ | 221/500 [16:31<19:15, 4.14s/it, loss=0.849, lr=0.000395] Steps: 44%|████▍ | 222/500 [16:33<16:01, 3.46s/it, loss=0.849, lr=0.000395] Steps: 44%|████▍ | 222/500 [16:33<16:01, 3.46s/it, loss=0.985, lr=0.000395] Steps: 45%|████▍ | 223/500 [16:35<13:46, 2.98s/it, loss=0.985, lr=0.000395] Steps: 45%|████▍ | 223/500 [16:35<13:46, 2.98s/it, loss=0.798, lr=0.000394] Steps: 45%|████▍ | 224/500 [16:37<12:11, 2.65s/it, loss=0.798, lr=0.000394] Steps: 45%|████▍ | 224/500 [16:37<12:11, 2.65s/it, loss=0.896, lr=0.000394] Steps: 45%|████▌ | 225/500 [16:45<19:01, 4.15s/it, loss=0.896, lr=0.000394] Steps: 45%|████▌ | 225/500 [16:45<19:01, 4.15s/it, loss=0.772, lr=0.000393] Steps: 45%|████▌ | 226/500 [16:47<15:50, 3.47s/it, loss=0.772, lr=0.000393] Steps: 45%|████▌ | 226/500 [16:47<15:50, 3.47s/it, loss=0.968, lr=0.000393] Steps: 45%|████▌ | 227/500 [16:48<13:36, 2.99s/it, loss=0.968, lr=0.000393] Steps: 45%|████▌ | 227/500 [16:48<13:36, 2.99s/it, loss=0.943, lr=0.000392] Steps: 46%|████▌ | 228/500 [16:50<12:01, 2.65s/it, loss=0.943, lr=0.000392] Steps: 46%|████▌ | 228/500 [16:50<12:01, 2.65s/it, loss=0.951, lr=0.000391] Steps: 46%|████▌ | 229/500 [16:58<18:52, 4.18s/it, loss=0.951, lr=0.000391] Steps: 46%|████▌ | 229/500 [16:58<18:52, 4.18s/it, loss=0.839, lr=0.000391] Steps: 46%|████▌ | 230/500 [17:00<15:41, 3.49s/it, loss=0.839, lr=0.000391] Steps: 46%|████▌ | 230/500 [17:00<15:41, 3.49s/it, loss=1.02, lr=0.00039] Steps: 46%|████▌ | 231/500 [17:02<13:27, 3.00s/it, loss=1.02, lr=0.00039] Steps: 46%|████▌ | 231/500 [17:02<13:27, 3.00s/it, loss=0.854, lr=0.00039] Steps: 46%|████▋ | 232/500 [17:04<11:53, 2.66s/it, loss=0.854, lr=0.00039] Steps: 46%|████▋ | 232/500 [17:04<11:53, 2.66s/it, loss=0.958, lr=0.000389] Steps: 47%|████▋ | 233/500 [17:11<18:44, 4.21s/it, loss=0.958, lr=0.000389] Steps: 47%|████▋ | 233/500 [17:11<18:44, 4.21s/it, loss=1.06, lr=0.000388] Steps: 47%|████▋ | 234/500 [17:13<15:33, 3.51s/it, loss=1.06, lr=0.000388] Steps: 47%|████▋ | 234/500 [17:13<15:33, 3.51s/it, loss=1.07, lr=0.000387] Steps: 47%|████▋ | 235/500 [17:15<13:20, 3.02s/it, loss=1.07, lr=0.000387] Steps: 47%|████▋ | 235/500 [17:15<13:20, 3.02s/it, loss=0.996, lr=0.000387] Steps: 47%|████▋ | 236/500 [17:17<11:46, 2.68s/it, loss=0.996, lr=0.000387] Steps: 47%|████▋ | 236/500 [17:17<11:46, 2.68s/it, loss=0.889, lr=0.000386] Steps: 47%|████▋ | 237/500 [17:25<18:25, 4.20s/it, loss=0.889, lr=0.000386] Steps: 47%|████▋ | 237/500 [17:25<18:25, 4.20s/it, loss=0.789, lr=0.000385] Steps: 48%|████▊ | 238/500 [17:27<15:18, 3.51s/it, loss=0.789, lr=0.000385] Steps: 48%|████▊ | 238/500 [17:27<15:18, 3.51s/it, loss=1.04, lr=0.000384] Steps: 48%|████▊ | 239/500 [17:29<13:07, 3.02s/it, loss=1.04, lr=0.000384] Steps: 48%|████▊ | 239/500 [17:29<13:07, 3.02s/it, loss=0.85, lr=0.000384] Steps: 48%|████▊ | 240/500 [17:31<11:35, 2.67s/it, loss=0.85, lr=0.000384] Steps: 48%|████▊ | 240/500 [17:31<11:35, 2.67s/it, loss=0.976, lr=0.000383] Steps: 48%|████▊ | 241/500 [17:38<18:00, 4.17s/it, loss=0.976, lr=0.000383] Steps: 48%|████▊ | 241/500 [17:38<18:00, 4.17s/it, loss=0.842, lr=0.000382] Steps: 48%|████▊ | 242/500 [17:40<14:57, 3.48s/it, loss=0.842, lr=0.000382] Steps: 48%|████▊ | 242/500 [17:40<14:57, 3.48s/it, loss=1.01, lr=0.000381] Steps: 49%|████▊ | 243/500 [17:42<12:50, 3.00s/it, loss=1.01, lr=0.000381] Steps: 49%|████▊ | 243/500 [17:42<12:50, 3.00s/it, loss=0.848, lr=0.00038] Steps: 49%|████▉ | 244/500 [17:44<11:21, 2.66s/it, loss=0.848, lr=0.00038] Steps: 49%|████▉ | 244/500 [17:44<11:21, 2.66s/it, loss=1.07, lr=0.000379] Steps: 49%|████▉ | 245/500 [17:51<17:42, 4.17s/it, loss=1.07, lr=0.000379] Steps: 49%|████▉ | 245/500 [17:51<17:42, 4.17s/it, loss=1.05, lr=0.000378] Steps: 49%|████▉ | 246/500 [17:53<14:43, 3.48s/it, loss=1.05, lr=0.000378] Steps: 49%|████▉ | 246/500 [17:53<14:43, 3.48s/it, loss=0.908, lr=0.000377] Steps: 49%|████▉ | 247/500 [17:55<12:38, 3.00s/it, loss=0.908, lr=0.000377] Steps: 49%|████▉ | 247/500 [17:55<12:38, 3.00s/it, loss=0.8, lr=0.000376] Steps: 50%|████▉ | 248/500 [17:57<11:10, 2.66s/it, loss=0.8, lr=0.000376] Steps: 50%|████▉ | 248/500 [17:57<11:10, 2.66s/it, loss=1.07, lr=0.000375] Steps: 50%|████▉ | 249/500 [18:05<17:29, 4.18s/it, loss=1.07, lr=0.000375] Steps: 50%|████▉ | 249/500 [18:05<17:29, 4.18s/it, loss=0.966, lr=0.000374] Steps: 50%|█████ | 250/500 [18:07<14:31, 3.49s/it, loss=0.966, lr=0.000374] Steps: 50%|█████ | 250/500 [18:07<14:31, 3.49s/it, loss=1.07, lr=0.000373] Steps: 50%|█████ | 251/500 [18:09<12:27, 3.00s/it, loss=1.07, lr=0.000373] Steps: 50%|█████ | 251/500 [18:09<12:27, 3.00s/it, loss=0.789, lr=0.000372] Steps: 50%|█████ | 252/500 [18:10<11:00, 2.66s/it, loss=0.789, lr=0.000372] Steps: 50%|█████ | 252/500 [18:10<11:00, 2.66s/it, loss=0.945, lr=0.000371] Steps: 51%|█████ | 253/500 [18:18<17:08, 4.16s/it, loss=0.945, lr=0.000371] Steps: 51%|█████ | 253/500 [18:18<17:08, 4.16s/it, loss=0.83, lr=0.00037] Steps: 51%|█████ | 254/500 [18:20<14:15, 3.48s/it, loss=0.83, lr=0.00037] Steps: 51%|█████ | 254/500 [18:20<14:15, 3.48s/it, loss=0.999, lr=0.000369] Steps: 51%|█████ | 255/500 [18:22<12:13, 3.00s/it, loss=0.999, lr=0.000369] Steps: 51%|█████ | 255/500 [18:22<12:13, 3.00s/it, loss=0.883, lr=0.000368] Steps: 51%|█████ | 256/500 [18:24<10:48, 2.66s/it, loss=0.883, lr=0.000368] Steps: 51%|█████ | 256/500 [18:24<10:48, 2.66s/it, loss=1.07, lr=0.000367] Steps: 51%|█████▏ | 257/500 [18:32<17:04, 4.21s/it, loss=1.07, lr=0.000367] Steps: 51%|█████▏ | 257/500 [18:32<17:04, 4.21s/it, loss=0.81, lr=0.000365] Steps: 52%|█████▏ | 258/500 [18:33<14:10, 3.51s/it, loss=0.81, lr=0.000365] Steps: 52%|█████▏ | 258/500 [18:33<14:10, 3.51s/it, loss=0.94, lr=0.000364] Steps: 52%|█████▏ | 259/500 [18:35<12:08, 3.02s/it, loss=0.94, lr=0.000364] Steps: 52%|█████▏ | 259/500 [18:35<12:08, 3.02s/it, loss=0.963, lr=0.000363] Steps: 52%|█████▏ | 260/500 [18:37<10:42, 2.68s/it, loss=0.963, lr=0.000363] Steps: 52%|█████▏ | 260/500 [18:37<10:42, 2.68s/it, loss=0.942, lr=0.000362] Steps: 52%|█████▏ | 261/500 [18:45<16:53, 4.24s/it, loss=0.942, lr=0.000362] Steps: 52%|█████▏ | 261/500 [18:45<16:53, 4.24s/it, loss=0.962, lr=0.000361] Steps: 52%|█████▏ | 262/500 [18:47<14:00, 3.53s/it, loss=0.962, lr=0.000361] Steps: 52%|█████▏ | 262/500 [18:47<14:00, 3.53s/it, loss=0.922, lr=0.000359] Steps: 53%|█████▎ | 263/500 [18:49<11:58, 3.03s/it, loss=0.922, lr=0.000359] Steps: 53%|█████▎ | 263/500 [18:49<11:58, 3.03s/it, loss=0.8, lr=0.000358] Steps: 53%|█████▎ | 264/500 [18:51<10:33, 2.69s/it, loss=0.8, lr=0.000358] Steps: 53%|█████▎ | 264/500 [18:51<10:33, 2.69s/it, loss=0.954, lr=0.000357] Steps: 53%|█████▎ | 265/500 [18:58<16:26, 4.20s/it, loss=0.954, lr=0.000357] Steps: 53%|█████▎ | 265/500 [18:58<16:26, 4.20s/it, loss=0.852, lr=0.000355] Steps: 53%|█████▎ | 266/500 [19:00<13:39, 3.50s/it, loss=0.852, lr=0.000355] Steps: 53%|█████▎ | 266/500 [19:00<13:39, 3.50s/it, loss=0.9, lr=0.000354] Steps: 53%|█████▎ | 267/500 [19:02<11:42, 3.02s/it, loss=0.9, lr=0.000354] Steps: 53%|█████▎ | 267/500 [19:02<11:42, 3.02s/it, loss=0.838, lr=0.000353] Steps: 54%|█████▎ | 268/500 [19:04<10:20, 2.67s/it, loss=0.838, lr=0.000353] Steps: 54%|█████▎ | 268/500 [19:04<10:20, 2.67s/it, loss=1.07, lr=0.000351] Steps: 54%|█████▍ | 269/500 [19:12<16:02, 4.17s/it, loss=1.07, lr=0.000351] Steps: 54%|█████▍ | 269/500 [19:12<16:02, 4.17s/it, loss=0.983, lr=0.00035] Steps: 54%|█████▍ | 270/500 [19:14<13:20, 3.48s/it, loss=0.983, lr=0.00035] Steps: 54%|█████▍ | 270/500 [19:14<13:20, 3.48s/it, loss=0.957, lr=0.000349] Steps: 54%|█████▍ | 271/500 [19:15<11:26, 3.00s/it, loss=0.957, lr=0.000349] Steps: 54%|█████▍ | 271/500 [19:15<11:26, 3.00s/it, loss=0.828, lr=0.000347] Steps: 54%|█████▍ | 272/500 [19:17<10:06, 2.66s/it, loss=0.828, lr=0.000347] Steps: 54%|█████▍ | 272/500 [19:17<10:06, 2.66s/it, loss=0.946, lr=0.000346] Steps: 55%|█████▍ | 273/500 [19:25<15:43, 4.16s/it, loss=0.946, lr=0.000346] Steps: 55%|█████▍ | 273/500 [19:25<15:43, 4.16s/it, loss=1.01, lr=0.000344] Steps: 55%|█████▍ | 274/500 [19:27<13:04, 3.47s/it, loss=1.01, lr=0.000344] Steps: 55%|█████▍ | 274/500 [19:27<13:04, 3.47s/it, loss=0.915, lr=0.000343] Steps: 55%|█████▌ | 275/500 [19:29<11:13, 2.99s/it, loss=0.915, lr=0.000343] Steps: 55%|█████▌ | 275/500 [19:29<11:13, 2.99s/it, loss=0.881, lr=0.000341] Steps: 55%|█████▌ | 276/500 [19:31<09:55, 2.66s/it, loss=0.881, lr=0.000341] Steps: 55%|█████▌ | 276/500 [19:31<09:55, 2.66s/it, loss=0.896, lr=0.00034] Steps: 55%|█████▌ | 277/500 [19:38<15:23, 4.14s/it, loss=0.896, lr=0.00034] Steps: 55%|█████▌ | 277/500 [19:38<15:23, 4.14s/it, loss=0.863, lr=0.000338] Steps: 56%|█████▌ | 278/500 [19:40<12:48, 3.46s/it, loss=0.863, lr=0.000338] Steps: 56%|█████▌ | 278/500 [19:40<12:48, 3.46s/it, loss=0.968, lr=0.000337] Steps: 56%|█████▌ | 279/500 [19:42<10:59, 2.99s/it, loss=0.968, lr=0.000337] Steps: 56%|█████▌ | 279/500 [19:42<10:59, 2.99s/it, loss=0.817, lr=0.000335] Steps: 56%|█████▌ | 280/500 [19:44<09:43, 2.65s/it, loss=0.817, lr=0.000335] Steps: 56%|█████▌ | 280/500 [19:44<09:43, 2.65s/it, loss=1.07, lr=0.000334] Steps: 56%|█████▌ | 281/500 [19:51<15:08, 4.15s/it, loss=1.07, lr=0.000334] Steps: 56%|█████▌ | 281/500 [19:51<15:08, 4.15s/it, loss=0.795, lr=0.000332] Steps: 56%|█████▋ | 282/500 [19:53<12:35, 3.46s/it, loss=0.795, lr=0.000332] Steps: 56%|█████▋ | 282/500 [19:53<12:35, 3.46s/it, loss=0.99, lr=0.000331] Steps: 57%|█████▋ | 283/500 [19:55<10:48, 2.99s/it, loss=0.99, lr=0.000331] Steps: 57%|█████▋ | 283/500 [19:55<10:48, 2.99s/it, loss=0.844, lr=0.000329] Steps: 57%|█████▋ | 284/500 [19:57<09:32, 2.65s/it, loss=0.844, lr=0.000329] Steps: 57%|█████▋ | 284/500 [19:57<09:32, 2.65s/it, loss=0.94, lr=0.000327] Steps: 57%|█████▋ | 285/500 [20:05<14:54, 4.16s/it, loss=0.94, lr=0.000327] Steps: 57%|█████▋ | 285/500 [20:05<14:54, 4.16s/it, loss=0.9, lr=0.000326] Steps: 57%|█████▋ | 286/500 [20:07<12:24, 3.48s/it, loss=0.9, lr=0.000326] Steps: 57%|█████▋ | 286/500 [20:07<12:24, 3.48s/it, loss=1.06, lr=0.000324] Steps: 57%|█████▋ | 287/500 [20:09<10:38, 3.00s/it, loss=1.06, lr=0.000324] Steps: 57%|█████▋ | 287/500 [20:09<10:38, 3.00s/it, loss=1.02, lr=0.000323] Steps: 58%|█████▊ | 288/500 [20:10<09:23, 2.66s/it, loss=1.02, lr=0.000323] Steps: 58%|█████▊ | 288/500 [20:10<09:23, 2.66s/it, loss=1.03, lr=0.000321] Steps: 58%|█████▊ | 289/500 [20:18<14:35, 4.15s/it, loss=1.03, lr=0.000321] Steps: 58%|█████▊ | 289/500 [20:18<14:35, 4.15s/it, loss=1.05, lr=0.000319] Steps: 58%|█████▊ | 290/500 [20:20<12:07, 3.47s/it, loss=1.05, lr=0.000319] Steps: 58%|█████▊ | 290/500 [20:20<12:07, 3.47s/it, loss=0.899, lr=0.000318] Steps: 58%|█████▊ | 291/500 [20:22<10:24, 2.99s/it, loss=0.899, lr=0.000318] Steps: 58%|█████▊ | 291/500 [20:22<10:24, 2.99s/it, loss=1.03, lr=0.000316] Steps: 58%|█████▊ | 292/500 [20:24<09:11, 2.65s/it, loss=1.03, lr=0.000316] Steps: 58%|█████▊ | 292/500 [20:24<09:11, 2.65s/it, loss=1.03, lr=0.000314] Steps: 59%|█████▊ | 293/500 [20:31<14:28, 4.19s/it, loss=1.03, lr=0.000314] Steps: 59%|█████▊ | 293/500 [20:31<14:28, 4.19s/it, loss=0.821, lr=0.000312] Steps: 59%|█████▉ | 294/500 [20:33<12:00, 3.50s/it, loss=0.821, lr=0.000312] Steps: 59%|█████▉ | 294/500 [20:33<12:00, 3.50s/it, loss=0.884, lr=0.000311] Steps: 59%|█████▉ | 295/500 [20:35<10:17, 3.01s/it, loss=0.884, lr=0.000311] Steps: 59%|█████▉ | 295/500 [20:35<10:17, 3.01s/it, loss=0.792, lr=0.000309] Steps: 59%|█████▉ | 296/500 [20:37<09:04, 2.67s/it, loss=0.792, lr=0.000309] Steps: 59%|█████▉ | 296/500 [20:37<09:04, 2.67s/it, loss=1.01, lr=0.000307] Steps: 59%|█████▉ | 297/500 [20:45<14:02, 4.15s/it, loss=1.01, lr=0.000307] Steps: 59%|█████▉ | 297/500 [20:45<14:02, 4.15s/it, loss=0.787, lr=0.000305] Steps: 60%|█████▉ | 298/500 [20:47<11:40, 3.47s/it, loss=0.787, lr=0.000305] Steps: 60%|█████▉ | 298/500 [20:47<11:40, 3.47s/it, loss=0.909, lr=0.000304] Steps: 60%|█████▉ | 299/500 [20:48<10:00, 2.99s/it, loss=0.909, lr=0.000304] Steps: 60%|█████▉ | 299/500 [20:48<10:00, 2.99s/it, loss=0.832, lr=0.000302] Steps: 60%|██████ | 300/500 [20:50<08:50, 2.65s/it, loss=0.832, lr=0.000302] Steps: 60%|██████ | 300/500 [20:50<08:50, 2.65s/it, loss=0.945, lr=0.0003] Steps: 60%|██████ | 301/500 [20:58<13:47, 4.16s/it, loss=0.945, lr=0.0003] Steps: 60%|██████ | 301/500 [20:58<13:47, 4.16s/it, loss=0.866, lr=0.000298] Steps: 60%|██████ | 302/500 [21:00<11:27, 3.47s/it, loss=0.866, lr=0.000298] Steps: 60%|██████ | 302/500 [21:00<11:27, 3.47s/it, loss=0.905, lr=0.000296] Steps: 61%|██████ | 303/500 [21:02<09:49, 2.99s/it, loss=0.905, lr=0.000296] Steps: 61%|██████ | 303/500 [21:02<09:49, 2.99s/it, loss=0.818, lr=0.000295] Steps: 61%|██████ | 304/500 [21:04<08:40, 2.66s/it, loss=0.818, lr=0.000295] Steps: 61%|██████ | 304/500 [21:04<08:40, 2.66s/it, loss=0.912, lr=0.000293] Steps: 61%|██████ | 305/500 [21:11<13:32, 4.16s/it, loss=0.912, lr=0.000293] Steps: 61%|██████ | 305/500 [21:11<13:32, 4.16s/it, loss=0.784, lr=0.000291] Steps: 61%|██████ | 306/500 [21:13<11:14, 3.48s/it, loss=0.784, lr=0.000291] Steps: 61%|██████ | 306/500 [21:13<11:14, 3.48s/it, loss=1.03, lr=0.000289] Steps: 61%|██████▏ | 307/500 [21:15<09:38, 3.00s/it, loss=1.03, lr=0.000289] Steps: 61%|██████▏ | 307/500 [21:15<09:38, 3.00s/it, loss=1.05, lr=0.000287] Steps: 62%|██████▏ | 308/500 [21:17<08:30, 2.66s/it, loss=1.05, lr=0.000287] Steps: 62%|██████▏ | 308/500 [21:17<08:30, 2.66s/it, loss=1.04, lr=0.000285] Steps: 62%|██████▏ | 309/500 [21:25<13:15, 4.16s/it, loss=1.04, lr=0.000285] Steps: 62%|██████▏ | 309/500 [21:25<13:15, 4.16s/it, loss=1.06, lr=0.000283] Steps: 62%|██████▏ | 310/500 [21:26<11:00, 3.47s/it, loss=1.06, lr=0.000283] Steps: 62%|██████▏ | 310/500 [21:26<11:00, 3.47s/it, loss=0.99, lr=0.000281] Steps: 62%|██████▏ | 311/500 [21:28<09:25, 2.99s/it, loss=0.99, lr=0.000281] Steps: 62%|██████▏ | 311/500 [21:28<09:25, 2.99s/it, loss=0.86, lr=0.000279] Steps: 62%|██████▏ | 312/500 [21:30<08:19, 2.66s/it, loss=0.86, lr=0.000279] Steps: 62%|██████▏ | 312/500 [21:30<08:19, 2.66s/it, loss=0.877, lr=0.000278] Steps: 63%|██████▎ | 313/500 [21:38<13:00, 4.17s/it, loss=0.877, lr=0.000278] Steps: 63%|██████▎ | 313/500 [21:38<13:00, 4.17s/it, loss=0.82, lr=0.000276] Steps: 63%|██████▎ | 314/500 [21:40<10:47, 3.48s/it, loss=0.82, lr=0.000276] Steps: 63%|██████▎ | 314/500 [21:40<10:47, 3.48s/it, loss=0.89, lr=0.000274] Steps: 63%|██████▎ | 315/500 [21:42<09:15, 3.00s/it, loss=0.89, lr=0.000274] Steps: 63%|██████▎ | 315/500 [21:42<09:15, 3.00s/it, loss=0.855, lr=0.000272] Steps: 63%|██████▎ | 316/500 [21:43<08:09, 2.66s/it, loss=0.855, lr=0.000272] Steps: 63%|██████▎ | 316/500 [21:43<08:09, 2.66s/it, loss=1.01, lr=0.00027] Steps: 63%|██████▎ | 317/500 [21:51<12:43, 4.17s/it, loss=1.01, lr=0.00027] Steps: 63%|██████▎ | 317/500 [21:51<12:43, 4.17s/it, loss=0.9, lr=0.000268] Steps: 64%|██████▎ | 318/500 [21:53<10:33, 3.48s/it, loss=0.9, lr=0.000268] Steps: 64%|██████▎ | 318/500 [21:53<10:33, 3.48s/it, loss=0.966, lr=0.000266] Steps: 64%|██████▍ | 319/500 [21:55<09:02, 3.00s/it, loss=0.966, lr=0.000266] Steps: 64%|██████▍ | 319/500 [21:55<09:02, 3.00s/it, loss=0.968, lr=0.000264] Steps: 64%|██████▍ | 320/500 [21:57<07:58, 2.66s/it, loss=0.968, lr=0.000264] Steps: 64%|██████▍ | 320/500 [21:57<07:58, 2.66s/it, loss=0.891, lr=0.000262] Steps: 64%|██████▍ | 321/500 [22:04<12:20, 4.14s/it, loss=0.891, lr=0.000262] Steps: 64%|██████▍ | 321/500 [22:04<12:20, 4.14s/it, loss=0.787, lr=0.00026] Steps: 64%|██████▍ | 322/500 [22:06<10:15, 3.46s/it, loss=0.787, lr=0.00026] Steps: 64%|██████▍ | 322/500 [22:06<10:15, 3.46s/it, loss=0.878, lr=0.000258] Steps: 65%|██████▍ | 323/500 [22:08<08:47, 2.98s/it, loss=0.878, lr=0.000258] Steps: 65%|██████▍ | 323/500 [22:08<08:47, 2.98s/it, loss=0.852, lr=0.000256] Steps: 65%|██████▍ | 324/500 [22:10<07:46, 2.65s/it, loss=0.852, lr=0.000256] Steps: 65%|██████▍ | 324/500 [22:10<07:46, 2.65s/it, loss=1.02, lr=0.000254] Steps: 65%|██████▌ | 325/500 [22:18<12:03, 4.14s/it, loss=1.02, lr=0.000254] Steps: 65%|██████▌ | 325/500 [22:18<12:03, 4.14s/it, loss=0.878, lr=0.000252] Steps: 65%|██████▌ | 326/500 [22:19<10:01, 3.46s/it, loss=0.878, lr=0.000252] Steps: 65%|██████▌ | 326/500 [22:19<10:01, 3.46s/it, loss=0.878, lr=0.00025] Steps: 65%|██████▌ | 327/500 [22:21<08:35, 2.98s/it, loss=0.878, lr=0.00025] Steps: 65%|██████▌ | 327/500 [22:21<08:35, 2.98s/it, loss=0.845, lr=0.000248] Steps: 66%|██████▌ | 328/500 [22:23<07:35, 2.65s/it, loss=0.845, lr=0.000248] Steps: 66%|██████▌ | 328/500 [22:23<07:35, 2.65s/it, loss=0.905, lr=0.000246] Steps: 66%|██████▌ | 329/500 [22:31<11:55, 4.18s/it, loss=0.905, lr=0.000246] Steps: 66%|██████▌ | 329/500 [22:31<11:55, 4.18s/it, loss=1.05, lr=0.000244] Steps: 66%|██████▌ | 330/500 [22:33<09:53, 3.49s/it, loss=1.05, lr=0.000244] Steps: 66%|██████▌ | 330/500 [22:33<09:53, 3.49s/it, loss=0.936, lr=0.000242] Steps: 66%|██████▌ | 331/500 [22:35<08:27, 3.00s/it, loss=0.936, lr=0.000242] Steps: 66%|██████▌ | 331/500 [22:35<08:27, 3.00s/it, loss=0.834, lr=0.00024] Steps: 66%|██████▋ | 332/500 [22:37<07:27, 2.67s/it, loss=0.834, lr=0.00024] Steps: 66%|██████▋ | 332/500 [22:37<07:27, 2.67s/it, loss=1, lr=0.000237] Steps: 67%|██████▋ | 333/500 [22:44<11:31, 4.14s/it, loss=1, lr=0.000237] Steps: 67%|██████▋ | 333/500 [22:44<11:31, 4.14s/it, loss=0.791, lr=0.000235] Steps: 67%|██████▋ | 334/500 [22:46<09:34, 3.46s/it, loss=0.791, lr=0.000235] Steps: 67%|██████▋ | 334/500 [22:46<09:34, 3.46s/it, loss=0.893, lr=0.000233] Steps: 67%|██████▋ | 335/500 [22:48<08:12, 2.98s/it, loss=0.893, lr=0.000233] Steps: 67%|██████▋ | 335/500 [22:48<08:12, 2.98s/it, loss=1.03, lr=0.000231] Steps: 67%|██████▋ | 336/500 [22:50<07:14, 2.65s/it, loss=1.03, lr=0.000231] Steps: 67%|██████▋ | 336/500 [22:50<07:14, 2.65s/it, loss=1.03, lr=0.000229] Steps: 67%|██████▋ | 337/500 [22:58<11:23, 4.20s/it, loss=1.03, lr=0.000229] Steps: 67%|██████▋ | 337/500 [22:58<11:23, 4.20s/it, loss=1.03, lr=0.000227] Steps: 68%|██████▊ | 338/500 [22:59<09:26, 3.50s/it, loss=1.03, lr=0.000227] Steps: 68%|██████▊ | 338/500 [22:59<09:26, 3.50s/it, loss=0.882, lr=0.000225] Steps: 68%|██████▊ | 339/500 [23:01<08:04, 3.01s/it, loss=0.882, lr=0.000225] Steps: 68%|██████▊ | 339/500 [23:01<08:04, 3.01s/it, loss=0.792, lr=0.000223] Steps: 68%|██████▊ | 340/500 [23:03<07:07, 2.67s/it, loss=0.792, lr=0.000223] Steps: 68%|██████▊ | 340/500 [23:03<07:07, 2.67s/it, loss=0.974, lr=0.000221] Steps: 68%|██████▊ | 341/500 [23:11<10:57, 4.14s/it, loss=0.974, lr=0.000221] Steps: 68%|██████▊ | 341/500 [23:11<10:57, 4.14s/it, loss=0.83, lr=0.000219] Steps: 68%|██████▊ | 342/500 [23:13<09:06, 3.46s/it, loss=0.83, lr=0.000219] Steps: 68%|██████▊ | 342/500 [23:13<09:06, 3.46s/it, loss=0.874, lr=0.000217] Steps: 69%|██████▊ | 343/500 [23:14<07:48, 2.98s/it, loss=0.874, lr=0.000217] Steps: 69%|██████▊ | 343/500 [23:15<07:48, 2.98s/it, loss=0.789, lr=0.000215] Steps: 69%|██████▉ | 344/500 [23:16<06:53, 2.65s/it, loss=0.789, lr=0.000215] Steps: 69%|██████▉ | 344/500 [23:16<06:53, 2.65s/it, loss=0.975, lr=0.000213] Steps: 69%|██████▉ | 345/500 [23:24<10:41, 4.14s/it, loss=0.975, lr=0.000213] Steps: 69%|██████▉ | 345/500 [23:24<10:41, 4.14s/it, loss=0.838, lr=0.00021] Steps: 69%|██████▉ | 346/500 [23:26<08:52, 3.46s/it, loss=0.838, lr=0.00021] Steps: 69%|██████▉ | 346/500 [23:26<08:52, 3.46s/it, loss=1.02, lr=0.000208] Steps: 69%|██████▉ | 347/500 [23:28<07:36, 2.98s/it, loss=1.02, lr=0.000208] Steps: 69%|██████▉ | 347/500 [23:28<07:36, 2.98s/it, loss=0.815, lr=0.000206] Steps: 70%|██████▉ | 348/500 [23:30<06:43, 2.65s/it, loss=0.815, lr=0.000206] Steps: 70%|██████▉ | 348/500 [23:30<06:43, 2.65s/it, loss=0.865, lr=0.000204] Steps: 70%|██████▉ | 349/500 [23:37<10:27, 4.15s/it, loss=0.865, lr=0.000204] Steps: 70%|██████▉ | 349/500 [23:37<10:27, 4.15s/it, loss=0.806, lr=0.000202] Steps: 70%|███████ | 350/500 [23:39<08:40, 3.47s/it, loss=0.806, lr=0.000202] Steps: 70%|███████ | 350/500 [23:39<08:40, 3.47s/it, loss=0.869, lr=0.0002] Steps: 70%|███████ | 351/500 [23:41<07:25, 2.99s/it, loss=0.869, lr=0.0002] Steps: 70%|███████ | 351/500 [23:41<07:25, 2.99s/it, loss=0.812, lr=0.000198] Steps: 70%|███████ | 352/500 [23:43<06:33, 2.66s/it, loss=0.812, lr=0.000198] Steps: 70%|███████ | 352/500 [23:43<06:33, 2.66s/it, loss=1.01, lr=0.000196] Steps: 71%|███████ | 353/500 [23:51<10:15, 4.19s/it, loss=1.01, lr=0.000196] Steps: 71%|███████ | 353/500 [23:51<10:15, 4.19s/it, loss=1.01, lr=0.000194] Steps: 71%|███████ | 354/500 [23:53<08:29, 3.49s/it, loss=1.01, lr=0.000194] Steps: 71%|███████ | 354/500 [23:53<08:29, 3.49s/it, loss=0.951, lr=0.000192] Steps: 71%|███████ | 355/500 [23:54<07:15, 3.01s/it, loss=0.951, lr=0.000192] Steps: 71%|███████ | 355/500 [23:54<07:15, 3.01s/it, loss=0.849, lr=0.00019] Steps: 71%|███████ | 356/500 [23:56<06:23, 2.67s/it, loss=0.849, lr=0.00019] Steps: 71%|███████ | 356/500 [23:56<06:23, 2.67s/it, loss=1.06, lr=0.000187] Steps: 71%|███████▏ | 357/500 [24:04<09:51, 4.14s/it, loss=1.06, lr=0.000187] Steps: 71%|███████▏ | 357/500 [24:04<09:51, 4.14s/it, loss=1.03, lr=0.000185] Steps: 72%|███████▏ | 358/500 [24:06<08:10, 3.46s/it, loss=1.03, lr=0.000185] Steps: 72%|███████▏ | 358/500 [24:06<08:10, 3.46s/it, loss=0.889, lr=0.000183] Steps: 72%|███████▏ | 359/500 [24:08<07:00, 2.98s/it, loss=0.889, lr=0.000183] Steps: 72%|███████▏ | 359/500 [24:08<07:00, 2.98s/it, loss=0.818, lr=0.000181] Steps: 72%|███████▏ | 360/500 [24:09<06:10, 2.65s/it, loss=0.818, lr=0.000181] Steps: 72%|███████▏ | 360/500 [24:09<06:10, 2.65s/it, loss=1, lr=0.000179] Steps: 72%|███████▏ | 361/500 [24:17<09:34, 4.13s/it, loss=1, lr=0.000179] Steps: 72%|███████▏ | 361/500 [24:17<09:34, 4.13s/it, loss=0.996, lr=0.000177] Steps: 72%|███████▏ | 362/500 [24:19<07:57, 3.46s/it, loss=0.996, lr=0.000177] Steps: 72%|███████▏ | 362/500 [24:19<07:57, 3.46s/it, loss=1.04, lr=0.000175] Steps: 73%|███████▎ | 363/500 [24:21<06:48, 2.98s/it, loss=1.04, lr=0.000175] Steps: 73%|███████▎ | 363/500 [24:21<06:48, 2.98s/it, loss=0.784, lr=0.000173] Steps: 73%|███████▎ | 364/500 [24:23<06:00, 2.65s/it, loss=0.784, lr=0.000173] Steps: 73%|███████▎ | 364/500 [24:23<06:00, 2.65s/it, loss=0.997, lr=0.000171] Steps: 73%|███████▎ | 365/500 [24:30<09:22, 4.17s/it, loss=0.997, lr=0.000171] Steps: 73%|███████▎ | 365/500 [24:30<09:22, 4.17s/it, loss=0.794, lr=0.000169] Steps: 73%|███████▎ | 366/500 [24:32<07:45, 3.48s/it, loss=0.794, lr=0.000169] Steps: 73%|███████▎ | 366/500 [24:32<07:45, 3.48s/it, loss=0.874, lr=0.000167] Steps: 73%|███████▎ | 367/500 [24:34<06:38, 3.00s/it, loss=0.874, lr=0.000167] Steps: 73%|███████▎ | 367/500 [24:34<06:38, 3.00s/it, loss=0.848, lr=0.000165] Steps: 74%|███████▎ | 368/500 [24:36<05:50, 2.66s/it, loss=0.848, lr=0.000165] Steps: 74%|███████▎ | 368/500 [24:36<05:50, 2.66s/it, loss=0.964, lr=0.000163] Steps: 74%|███████▍ | 369/500 [24:44<09:07, 4.18s/it, loss=0.964, lr=0.000163] Steps: 74%|███████▍ | 369/500 [24:44<09:07, 4.18s/it, loss=0.778, lr=0.00016] Steps: 74%|███████▍ | 370/500 [24:46<07:33, 3.49s/it, loss=0.778, lr=0.00016] Steps: 74%|███████▍ | 370/500 [24:46<07:33, 3.49s/it, loss=1.04, lr=0.000158] Steps: 74%|███████▍ | 371/500 [24:47<06:27, 3.00s/it, loss=1.04, lr=0.000158] Steps: 74%|███████▍ | 371/500 [24:47<06:27, 3.00s/it, loss=1, lr=0.000156] Steps: 74%|███████▍ | 372/500 [24:49<05:41, 2.67s/it, loss=1, lr=0.000156] Steps: 74%|███████▍ | 372/500 [24:49<05:41, 2.67s/it, loss=0.937, lr=0.000154] Steps: 75%|███████▍ | 373/500 [24:57<08:47, 4.15s/it, loss=0.937, lr=0.000154] Steps: 75%|███████▍ | 373/500 [24:57<08:47, 4.15s/it, loss=1.05, lr=0.000152] Steps: 75%|███████▍ | 374/500 [24:59<07:17, 3.47s/it, loss=1.05, lr=0.000152] Steps: 75%|███████▍ | 374/500 [24:59<07:17, 3.47s/it, loss=0.894, lr=0.00015] Steps: 75%|███████▌ | 375/500 [25:01<06:13, 2.99s/it, loss=0.894, lr=0.00015] Steps: 75%|███████▌ | 375/500 [25:01<06:13, 2.99s/it, loss=0.821, lr=0.000148] Steps: 75%|███████▌ | 376/500 [25:03<05:29, 2.66s/it, loss=0.821, lr=0.000148] Steps: 75%|███████▌ | 376/500 [25:03<05:29, 2.66s/it, loss=1.04, lr=0.000146] Steps: 75%|███████▌ | 377/500 [25:10<08:34, 4.19s/it, loss=1.04, lr=0.000146] Steps: 75%|███████▌ | 377/500 [25:10<08:34, 4.19s/it, loss=0.978, lr=0.000144] Steps: 76%|███████▌ | 378/500 [25:12<07:05, 3.49s/it, loss=0.978, lr=0.000144] Steps: 76%|███████▌ | 378/500 [25:12<07:05, 3.49s/it, loss=0.943, lr=0.000142] Steps: 76%|███████▌ | 379/500 [25:14<06:03, 3.01s/it, loss=0.943, lr=0.000142] Steps: 76%|███████▌ | 379/500 [25:14<06:03, 3.01s/it, loss=1.05, lr=0.00014] Steps: 76%|███████▌ | 380/500 [25:16<05:20, 2.67s/it, loss=1.05, lr=0.00014] Steps: 76%|███████▌ | 380/500 [25:16<05:20, 2.67s/it, loss=0.892, lr=0.000138] Steps: 76%|███████▌ | 381/500 [25:24<08:12, 4.14s/it, loss=0.892, lr=0.000138] Steps: 76%|███████▌ | 381/500 [25:24<08:12, 4.14s/it, loss=0.82, lr=0.000136] Steps: 76%|███████▋ | 382/500 [25:25<06:48, 3.46s/it, loss=0.82, lr=0.000136] Steps: 76%|███████▋ | 382/500 [25:25<06:48, 3.46s/it, loss=1.02, lr=0.000134] Steps: 77%|███████▋ | 383/500 [25:27<05:49, 2.98s/it, loss=1.02, lr=0.000134] Steps: 77%|███████▋ | 383/500 [25:27<05:49, 2.98s/it, loss=0.785, lr=0.000132] Steps: 77%|███████▋ | 384/500 [25:29<05:07, 2.65s/it, loss=0.785, lr=0.000132] Steps: 77%|███████▋ | 384/500 [25:29<05:07, 2.65s/it, loss=0.898, lr=0.00013] Steps: 77%|███████▋ | 385/500 [25:37<07:56, 4.14s/it, loss=0.898, lr=0.00013] Steps: 77%|███████▋ | 385/500 [25:37<07:56, 4.14s/it, loss=0.836, lr=0.000128] Steps: 77%|███████▋ | 386/500 [25:39<06:34, 3.46s/it, loss=0.836, lr=0.000128] Steps: 77%|███████▋ | 386/500 [25:39<06:34, 3.46s/it, loss=0.894, lr=0.000126] Steps: 77%|███████▋ | 387/500 [25:41<05:37, 2.98s/it, loss=0.894, lr=0.000126] Steps: 77%|███████▋ | 387/500 [25:41<05:37, 2.98s/it, loss=0.776, lr=0.000124] Steps: 78%|███████▊ | 388/500 [25:42<04:56, 2.65s/it, loss=0.776, lr=0.000124] Steps: 78%|███████▊ | 388/500 [25:42<04:56, 2.65s/it, loss=1.06, lr=0.000122] Steps: 78%|███████▊ | 389/500 [25:50<07:38, 4.14s/it, loss=1.06, lr=0.000122] Steps: 78%|███████▊ | 389/500 [25:50<07:38, 4.14s/it, loss=0.79, lr=0.000121] Steps: 78%|███████▊ | 390/500 [25:52<06:20, 3.46s/it, loss=0.79, lr=0.000121] Steps: 78%|███████▊ | 390/500 [25:52<06:20, 3.46s/it, loss=0.867, lr=0.000119] Steps: 78%|███████▊ | 391/500 [25:54<05:25, 2.98s/it, loss=0.867, lr=0.000119] Steps: 78%|███████▊ | 391/500 [25:54<05:25, 2.98s/it, loss=0.79, lr=0.000117] Steps: 78%|███████▊ | 392/500 [25:56<04:46, 2.65s/it, loss=0.79, lr=0.000117] Steps: 78%|███████▊ | 392/500 [25:56<04:46, 2.65s/it, loss=0.867, lr=0.000115] Steps: 79%|███████▊ | 393/500 [26:03<07:21, 4.12s/it, loss=0.867, lr=0.000115] Steps: 79%|███████▊ | 393/500 [26:03<07:21, 4.12s/it, loss=0.818, lr=0.000113] Steps: 79%|███████▉ | 394/500 [26:05<06:05, 3.45s/it, loss=0.818, lr=0.000113] Steps: 79%|███████▉ | 394/500 [26:05<06:05, 3.45s/it, loss=0.931, lr=0.000111] Steps: 79%|███████▉ | 395/500 [26:07<05:12, 2.97s/it, loss=0.931, lr=0.000111] Steps: 79%|███████▉ | 395/500 [26:07<05:12, 2.97s/it, loss=0.821, lr=0.000109] Steps: 79%|███████▉ | 396/500 [26:09<04:35, 2.65s/it, loss=0.821, lr=0.000109] Steps: 79%|███████▉ | 396/500 [26:09<04:35, 2.65s/it, loss=0.916, lr=0.000107] Steps: 79%|███████▉ | 397/500 [26:17<07:10, 4.18s/it, loss=0.916, lr=0.000107] Steps: 79%|███████▉ | 397/500 [26:17<07:10, 4.18s/it, loss=0.805, lr=0.000105] Steps: 80%|███████▉ | 398/500 [26:18<05:55, 3.49s/it, loss=0.805, lr=0.000105] Steps: 80%|███████▉ | 398/500 [26:18<05:55, 3.49s/it, loss=1.06, lr=0.000104] Steps: 80%|███████▉ | 399/500 [26:20<05:03, 3.00s/it, loss=1.06, lr=0.000104] Steps: 80%|███████▉ | 399/500 [26:20<05:03, 3.00s/it, loss=0.812, lr=0.000102] Steps: 80%|████████ | 400/500 [26:22<04:26, 2.66s/it, loss=0.812, lr=0.000102] Steps: 80%|████████ | 400/500 [26:22<04:26, 2.66s/it, loss=0.863, lr=0.0001] Steps: 80%|████████ | 401/500 [26:30<06:54, 4.19s/it, loss=0.863, lr=0.0001] Steps: 80%|████████ | 401/500 [26:30<06:54, 4.19s/it, loss=0.843, lr=9.82e-5] Steps: 80%|████████ | 402/500 [26:32<05:42, 3.49s/it, loss=0.843, lr=9.82e-5] Steps: 80%|████████ | 402/500 [26:32<05:42, 3.49s/it, loss=0.926, lr=9.64e-5] Steps: 81%|████████ | 403/500 [26:34<04:51, 3.01s/it, loss=0.926, lr=9.64e-5] Steps: 81%|████████ | 403/500 [26:34<04:51, 3.01s/it, loss=0.953, lr=9.46e-5] Steps: 81%|████████ | 404/500 [26:36<04:15, 2.67s/it, loss=0.953, lr=9.46e-5] Steps: 81%|████████ | 404/500 [26:36<04:15, 2.67s/it, loss=1.01, lr=9.28e-5] Steps: 81%|████████ | 405/500 [26:43<06:40, 4.22s/it, loss=1.01, lr=9.28e-5] Steps: 81%|████████ | 405/500 [26:43<06:40, 4.22s/it, loss=0.825, lr=9.11e-5] Steps: 81%|████████ | 406/500 [26:45<05:30, 3.52s/it, loss=0.825, lr=9.11e-5] Steps: 81%|████████ | 406/500 [26:45<05:30, 3.52s/it, loss=0.909, lr=8.93e-5] Steps: 81%|████████▏ | 407/500 [26:47<04:41, 3.02s/it, loss=0.909, lr=8.93e-5] Steps: 81%|████████▏ | 407/500 [26:47<04:41, 3.02s/it, loss=0.781, lr=8.76e-5] Steps: 82%|████████▏ | 408/500 [26:49<04:06, 2.68s/it, loss=0.781, lr=8.76e-5] Steps: 82%|████████▏ | 408/500 [26:49<04:06, 2.68s/it, loss=0.862, lr=8.59e-5] Steps: 82%|████████▏ | 409/500 [26:57<06:22, 4.20s/it, loss=0.862, lr=8.59e-5] Steps: 82%|████████▏ | 409/500 [26:57<06:22, 4.20s/it, loss=0.822, lr=8.41e-5] Steps: 82%|████████▏ | 410/500 [26:59<05:15, 3.50s/it, loss=0.822, lr=8.41e-5] Steps: 82%|████████▏ | 410/500 [26:59<05:15, 3.50s/it, loss=1.01, lr=8.24e-5] Steps: 82%|████████▏ | 411/500 [27:01<04:28, 3.01s/it, loss=1.01, lr=8.24e-5] Steps: 82%|████████▏ | 411/500 [27:01<04:28, 3.01s/it, loss=0.829, lr=8.08e-5] Steps: 82%|████████▏ | 412/500 [27:02<03:55, 2.67s/it, loss=0.829, lr=8.08e-5] Steps: 82%|████████▏ | 412/500 [27:02<03:55, 2.67s/it, loss=0.92, lr=7.91e-5] Steps: 83%|████████▎ | 413/500 [27:10<06:02, 4.16s/it, loss=0.92, lr=7.91e-5] Steps: 83%|████████▎ | 413/500 [27:10<06:02, 4.16s/it, loss=0.789, lr=7.74e-5] Steps: 83%|████████▎ | 414/500 [27:12<04:59, 3.48s/it, loss=0.789, lr=7.74e-5] Steps: 83%|████████▎ | 414/500 [27:12<04:59, 3.48s/it, loss=0.979, lr=7.58e-5] Steps: 83%|████████▎ | 415/500 [27:14<04:14, 3.00s/it, loss=0.979, lr=7.58e-5] Steps: 83%|████████▎ | 415/500 [27:14<04:14, 3.00s/it, loss=0.963, lr=7.41e-5] Steps: 83%|████████▎ | 416/500 [27:16<03:43, 2.66s/it, loss=0.963, lr=7.41e-5] Steps: 83%|████████▎ | 416/500 [27:16<03:43, 2.66s/it, loss=0.903, lr=7.25e-5] Steps: 83%|████████▎ | 417/500 [27:23<05:42, 4.13s/it, loss=0.903, lr=7.25e-5] Steps: 83%|████████▎ | 417/500 [27:23<05:42, 4.13s/it, loss=0.803, lr=7.09e-5] Steps: 84%|████████▎ | 418/500 [27:25<04:42, 3.45s/it, loss=0.803, lr=7.09e-5] Steps: 84%|████████▎ | 418/500 [27:25<04:42, 3.45s/it, loss=0.924, lr=6.93e-5] Steps: 84%|████████▍ | 419/500 [27:27<04:01, 2.98s/it, loss=0.924, lr=6.93e-5] Steps: 84%|████████▍ | 419/500 [27:27<04:01, 2.98s/it, loss=0.769, lr=6.77e-5] Steps: 84%|████████▍ | 420/500 [27:29<03:31, 2.64s/it, loss=0.769, lr=6.77e-5] Steps: 84%|████████▍ | 420/500 [27:29<03:31, 2.64s/it, loss=0.99, lr=6.62e-5] Steps: 84%|████████▍ | 421/500 [27:37<05:28, 4.16s/it, loss=0.99, lr=6.62e-5] Steps: 84%|████████▍ | 421/500 [27:37<05:28, 4.16s/it, loss=0.778, lr=6.46e-5] Steps: 84%|████████▍ | 422/500 [27:38<04:31, 3.48s/it, loss=0.778, lr=6.46e-5] Steps: 84%|████████▍ | 422/500 [27:38<04:31, 3.48s/it, loss=0.994, lr=6.31e-5] Steps: 85%|████████▍ | 423/500 [27:40<03:50, 3.00s/it, loss=0.994, lr=6.31e-5] Steps: 85%|████████▍ | 423/500 [27:40<03:50, 3.00s/it, loss=0.845, lr=6.16e-5] Steps: 85%|████████▍ | 424/500 [27:42<03:22, 2.66s/it, loss=0.845, lr=6.16e-5] Steps: 85%|████████▍ | 424/500 [27:42<03:22, 2.66s/it, loss=0.944, lr=6.01e-5] Steps: 85%|████████▌ | 425/500 [27:50<05:13, 4.18s/it, loss=0.944, lr=6.01e-5] Steps: 85%|████████▌ | 425/500 [27:50<05:13, 4.18s/it, loss=0.771, lr=5.86e-5] Steps: 85%|████████▌ | 426/500 [27:52<04:17, 3.49s/it, loss=0.771, lr=5.86e-5] Steps: 85%|████████▌ | 426/500 [27:52<04:17, 3.49s/it, loss=0.932, lr=5.71e-5] Steps: 85%|████████▌ | 427/500 [27:54<03:39, 3.00s/it, loss=0.932, lr=5.71e-5] Steps: 85%|████████▌ | 427/500 [27:54<03:39, 3.00s/it, loss=0.771, lr=5.56e-5] Steps: 86%|████████▌ | 428/500 [27:55<03:11, 2.66s/it, loss=0.771, lr=5.56e-5] Steps: 86%|████████▌ | 428/500 [27:55<03:11, 2.66s/it, loss=0.861, lr=5.42e-5] Steps: 86%|████████▌ | 429/500 [28:03<04:57, 4.19s/it, loss=0.861, lr=5.42e-5] Steps: 86%|████████▌ | 429/500 [28:03<04:57, 4.19s/it, loss=0.836, lr=5.28e-5] Steps: 86%|████████▌ | 430/500 [28:05<04:04, 3.50s/it, loss=0.836, lr=5.28e-5] Steps: 86%|████████▌ | 430/500 [28:05<04:04, 3.50s/it, loss=0.99, lr=5.14e-5] Steps: 86%|████████▌ | 431/500 [28:07<03:27, 3.01s/it, loss=0.99, lr=5.14e-5] Steps: 86%|████████▌ | 431/500 [28:07<03:27, 3.01s/it, loss=0.804, lr=5e-5] Steps: 86%|████████▋ | 432/500 [28:09<03:01, 2.67s/it, loss=0.804, lr=5e-5] Steps: 86%|████████▋ | 432/500 [28:09<03:01, 2.67s/it, loss=0.885, lr=4.86e-5] Steps: 87%|████████▋ | 433/500 [28:17<04:42, 4.22s/it, loss=0.885, lr=4.86e-5] Steps: 87%|████████▋ | 433/500 [28:17<04:42, 4.22s/it, loss=1, lr=4.72e-5] Steps: 87%|████████▋ | 434/500 [28:19<03:51, 3.51s/it, loss=1, lr=4.72e-5] Steps: 87%|████████▋ | 434/500 [28:19<03:51, 3.51s/it, loss=0.999, lr=4.59e-5] Steps: 87%|████████▋ | 435/500 [28:20<03:16, 3.02s/it, loss=0.999, lr=4.59e-5] Steps: 87%|████████▋ | 435/500 [28:20<03:16, 3.02s/it, loss=0.774, lr=4.46e-5] Steps: 87%|████████▋ | 436/500 [28:22<02:51, 2.68s/it, loss=0.774, lr=4.46e-5] Steps: 87%|████████▋ | 436/500 [28:22<02:51, 2.68s/it, loss=0.946, lr=4.33e-5] Steps: 87%|████████▋ | 437/500 [28:30<04:21, 4.15s/it, loss=0.946, lr=4.33e-5] Steps: 87%|████████▋ | 437/500 [28:30<04:21, 4.15s/it, loss=0.841, lr=4.2e-5] Steps: 88%|████████▊ | 438/500 [28:32<03:34, 3.46s/it, loss=0.841, lr=4.2e-5] Steps: 88%|████████▊ | 438/500 [28:32<03:34, 3.46s/it, loss=1.01, lr=4.07e-5] Steps: 88%|████████▊ | 439/500 [28:34<03:02, 2.99s/it, loss=1.01, lr=4.07e-5] Steps: 88%|████████▊ | 439/500 [28:34<03:02, 2.99s/it, loss=0.882, lr=3.94e-5] Steps: 88%|████████▊ | 440/500 [28:36<02:39, 2.65s/it, loss=0.882, lr=3.94e-5] Steps: 88%|████████▊ | 440/500 [28:36<02:39, 2.65s/it, loss=0.937, lr=3.82e-5] Steps: 88%|████████▊ | 441/500 [28:44<04:11, 4.26s/it, loss=0.937, lr=3.82e-5] Steps: 88%|████████▊ | 441/500 [28:44<04:11, 4.26s/it, loss=0.851, lr=3.7e-5] Steps: 88%|████████▊ | 442/500 [28:45<03:25, 3.54s/it, loss=0.851, lr=3.7e-5] Steps: 88%|████████▊ | 442/500 [28:45<03:25, 3.54s/it, loss=0.866, lr=3.58e-5] Steps: 89%|████████▊ | 443/500 [28:47<02:53, 3.04s/it, loss=0.866, lr=3.58e-5] Steps: 89%|████████▊ | 443/500 [28:47<02:53, 3.04s/it, loss=0.918, lr=3.46e-5] Steps: 89%|████████▉ | 444/500 [28:49<02:30, 2.69s/it, loss=0.918, lr=3.46e-5] Steps: 89%|████████▉ | 444/500 [28:49<02:30, 2.69s/it, loss=0.892, lr=3.34e-5] Steps: 89%|████████▉ | 445/500 [28:57<03:49, 4.18s/it, loss=0.892, lr=3.34e-5] Steps: 89%|████████▉ | 445/500 [28:57<03:49, 4.18s/it, loss=0.787, lr=3.23e-5] Steps: 89%|████████▉ | 446/500 [28:59<03:08, 3.49s/it, loss=0.787, lr=3.23e-5] Steps: 89%|████████▉ | 446/500 [28:59<03:08, 3.49s/it, loss=0.89, lr=3.11e-5] Steps: 89%|████████▉ | 447/500 [29:01<02:39, 3.00s/it, loss=0.89, lr=3.11e-5] Steps: 89%|████████▉ | 447/500 [29:01<02:39, 3.00s/it, loss=1.05, lr=3e-5] Steps: 90%|████████▉ | 448/500 [29:02<02:18, 2.66s/it, loss=1.05, lr=3e-5] Steps: 90%|████████▉ | 448/500 [29:02<02:18, 2.66s/it, loss=1.01, lr=2.89e-5] Steps: 90%|████████▉ | 449/500 [29:10<03:31, 4.15s/it, loss=1.01, lr=2.89e-5] Steps: 90%|████████▉ | 449/500 [29:10<03:31, 4.15s/it, loss=0.801, lr=2.79e-5] Steps: 90%|█████████ | 450/500 [29:12<02:53, 3.47s/it, loss=0.801, lr=2.79e-5] Steps: 90%|█████████ | 450/500 [29:12<02:53, 3.47s/it, loss=0.908, lr=2.68e-5] Steps: 90%|█████████ | 451/500 [29:14<02:26, 2.99s/it, loss=0.908, lr=2.68e-5] Steps: 90%|█████████ | 451/500 [29:14<02:26, 2.99s/it, loss=0.758, lr=2.58e-5] Steps: 90%|█████████ | 452/500 [29:16<02:07, 2.65s/it, loss=0.758, lr=2.58e-5] Steps: 90%|█████████ | 452/500 [29:16<02:07, 2.65s/it, loss=0.872, lr=2.47e-5] Steps: 91%|█████████ | 453/500 [29:23<03:15, 4.15s/it, loss=0.872, lr=2.47e-5] Steps: 91%|█████████ | 453/500 [29:23<03:15, 4.15s/it, loss=0.784, lr=2.37e-5] Steps: 91%|█████████ | 454/500 [29:25<02:39, 3.47s/it, loss=0.784, lr=2.37e-5] Steps: 91%|█████████ | 454/500 [29:25<02:39, 3.47s/it, loss=0.86, lr=2.28e-5] Steps: 91%|█████████ | 455/500 [29:27<02:14, 2.99s/it, loss=0.86, lr=2.28e-5] Steps: 91%|█████████ | 455/500 [29:27<02:14, 2.99s/it, loss=0.839, lr=2.18e-5] Steps: 91%|█████████ | 456/500 [29:29<01:56, 2.66s/it, loss=0.839, lr=2.18e-5] Steps: 91%|█████████ | 456/500 [29:29<01:56, 2.66s/it, loss=1, lr=2.09e-5] Steps: 91%|█████████▏| 457/500 [29:37<02:59, 4.18s/it, loss=1, lr=2.09e-5] Steps: 91%|█████████▏| 457/500 [29:37<02:59, 4.18s/it, loss=1.05, lr=1.99e-5] Steps: 92%|█████████▏| 458/500 [29:39<02:26, 3.49s/it, loss=1.05, lr=1.99e-5] Steps: 92%|█████████▏| 458/500 [29:39<02:26, 3.49s/it, loss=0.987, lr=1.9e-5] Steps: 92%|█████████▏| 459/500 [29:40<02:03, 3.00s/it, loss=0.987, lr=1.9e-5] Steps: 92%|█████████▏| 459/500 [29:40<02:03, 3.00s/it, loss=0.831, lr=1.82e-5] Steps: 92%|█████████▏| 460/500 [29:42<01:46, 2.66s/it, loss=0.831, lr=1.82e-5] Steps: 92%|█████████▏| 460/500 [29:42<01:46, 2.66s/it, loss=0.996, lr=1.73e-5] Steps: 92%|█████████▏| 461/500 [29:50<02:42, 4.17s/it, loss=0.996, lr=1.73e-5] Steps: 92%|█████████▏| 461/500 [29:50<02:42, 4.17s/it, loss=0.805, lr=1.64e-5] Steps: 92%|█████████▏| 462/500 [29:52<02:12, 3.48s/it, loss=0.805, lr=1.64e-5] Steps: 92%|█████████▏| 462/500 [29:52<02:12, 3.48s/it, loss=1.05, lr=1.56e-5] Steps: 93%|█████████▎| 463/500 [29:54<01:50, 3.00s/it, loss=1.05, lr=1.56e-5] Steps: 93%|█████████▎| 463/500 [29:54<01:50, 3.00s/it, loss=0.837, lr=1.48e-5] Steps: 93%|█████████▎| 464/500 [29:56<01:35, 2.66s/it, loss=0.837, lr=1.48e-5] Steps: 93%|█████████▎| 464/500 [29:56<01:35, 2.66s/it, loss=0.87, lr=1.4e-5] Steps: 93%|█████████▎| 465/500 [30:03<02:24, 4.13s/it, loss=0.87, lr=1.4e-5] Steps: 93%|█████████▎| 465/500 [30:03<02:24, 4.13s/it, loss=0.973, lr=1.33e-5] Steps: 93%|█████████▎| 466/500 [30:05<01:57, 3.45s/it, loss=0.973, lr=1.33e-5] Steps: 93%|█████████▎| 466/500 [30:05<01:57, 3.45s/it, loss=0.984, lr=1.25e-5] Steps: 93%|█████████▎| 467/500 [30:07<01:38, 2.98s/it, loss=0.984, lr=1.25e-5] Steps: 93%|█████████▎| 467/500 [30:07<01:38, 2.98s/it, loss=0.84, lr=1.18e-5] Steps: 94%|█████████▎| 468/500 [30:09<01:24, 2.65s/it, loss=0.84, lr=1.18e-5] Steps: 94%|█████████▎| 468/500 [30:09<01:24, 2.65s/it, loss=0.917, lr=1.11e-5] Steps: 94%|█████████▍| 469/500 [30:16<02:08, 4.15s/it, loss=0.917, lr=1.11e-5] Steps: 94%|█████████▍| 469/500 [30:16<02:08, 4.15s/it, loss=0.838, lr=1.04e-5] Steps: 94%|█████████▍| 470/500 [30:18<01:43, 3.46s/it, loss=0.838, lr=1.04e-5] Steps: 94%|█████████▍| 470/500 [30:18<01:43, 3.46s/it, loss=0.949, lr=9.79e-6] Steps: 94%|█████████▍| 471/500 [30:20<01:26, 2.99s/it, loss=0.949, lr=9.79e-6] Steps: 94%|█████████▍| 471/500 [30:20<01:26, 2.99s/it, loss=0.809, lr=9.15e-6] Steps: 94%|█████████▍| 472/500 [30:22<01:14, 2.65s/it, loss=0.809, lr=9.15e-6] Steps: 94%|█████████▍| 472/500 [30:22<01:14, 2.65s/it, loss=1.05, lr=8.54e-6] Steps: 95%|█████████▍| 473/500 [30:30<01:52, 4.17s/it, loss=1.05, lr=8.54e-6] Steps: 95%|█████████▍| 473/500 [30:30<01:52, 4.17s/it, loss=0.781, lr=7.94e-6] Steps: 95%|█████████▍| 474/500 [30:32<01:30, 3.48s/it, loss=0.781, lr=7.94e-6] Steps: 95%|█████████▍| 474/500 [30:32<01:30, 3.48s/it, loss=1.02, lr=7.37e-6] Steps: 95%|█████████▌| 475/500 [30:33<01:14, 3.00s/it, loss=1.02, lr=7.37e-6] Steps: 95%|█████████▌| 475/500 [30:33<01:14, 3.00s/it, loss=0.77, lr=6.81e-6] Steps: 95%|█████████▌| 476/500 [30:35<01:03, 2.66s/it, loss=0.77, lr=6.81e-6] Steps: 95%|█████████▌| 476/500 [30:35<01:03, 2.66s/it, loss=0.964, lr=6.28e-6] Steps: 95%|█████████▌| 477/500 [30:43<01:35, 4.16s/it, loss=0.964, lr=6.28e-6] Steps: 95%|█████████▌| 477/500 [30:43<01:35, 4.16s/it, loss=0.813, lr=5.77e-6] Steps: 96%|█████████▌| 478/500 [30:45<01:16, 3.47s/it, loss=0.813, lr=5.77e-6] Steps: 96%|█████████▌| 478/500 [30:45<01:16, 3.47s/it, loss=1.06, lr=5.28e-6] Steps: 96%|█████████▌| 479/500 [30:47<01:02, 3.00s/it, loss=1.06, lr=5.28e-6] Steps: 96%|█████████▌| 479/500 [30:47<01:02, 3.00s/it, loss=0.959, lr=4.82e-6] Steps: 96%|█████████▌| 480/500 [30:49<00:53, 2.66s/it, loss=0.959, lr=4.82e-6] Steps: 96%|█████████▌| 480/500 [30:49<00:53, 2.66s/it, loss=0.859, lr=4.37e-6] Steps: 96%|█████████▌| 481/500 [30:56<01:19, 4.17s/it, loss=0.859, lr=4.37e-6] Steps: 96%|█████████▌| 481/500 [30:56<01:19, 4.17s/it, loss=1.03, lr=3.95e-6] Steps: 96%|█████████▋| 482/500 [30:58<01:02, 3.48s/it, loss=1.03, lr=3.95e-6] Steps: 96%|█████████▋| 482/500 [30:58<01:02, 3.48s/it, loss=1.07, lr=3.54e-6] Steps: 97%|█████████▋| 483/500 [31:00<00:50, 3.00s/it, loss=1.07, lr=3.54e-6] Steps: 97%|█████████▋| 483/500 [31:00<00:50, 3.00s/it, loss=0.791, lr=3.16e-6] Steps: 97%|█████████▋| 484/500 [31:02<00:42, 2.66s/it, loss=0.791, lr=3.16e-6] Steps: 97%|█████████▋| 484/500 [31:02<00:42, 2.66s/it, loss=0.934, lr=2.8e-6] Steps: 97%|█████████▋| 485/500 [31:10<01:02, 4.15s/it, loss=0.934, lr=2.8e-6] Steps: 97%|█████████▋| 485/500 [31:10<01:02, 4.15s/it, loss=0.777, lr=2.46e-6] Steps: 97%|█████████▋| 486/500 [31:11<00:48, 3.47s/it, loss=0.777, lr=2.46e-6] Steps: 97%|█████████▋| 486/500 [31:11<00:48, 3.47s/it, loss=0.957, lr=2.15e-6] Steps: 97%|█████████▋| 487/500 [31:13<00:38, 2.99s/it, loss=0.957, lr=2.15e-6] Steps: 97%|█████████▋| 487/500 [31:13<00:38, 2.99s/it, loss=1.04, lr=1.85e-6] Steps: 98%|█████████▊| 488/500 [31:15<00:31, 2.65s/it, loss=1.04, lr=1.85e-6] Steps: 98%|█████████▊| 488/500 [31:15<00:31, 2.65s/it, loss=1.01, lr=1.58e-6] Steps: 98%|█████████▊| 489/500 [31:23<00:45, 4.12s/it, loss=1.01, lr=1.58e-6] Steps: 98%|█████████▊| 489/500 [31:23<00:45, 4.12s/it, loss=0.779, lr=1.33e-6] Steps: 98%|█████████▊| 490/500 [31:25<00:34, 3.45s/it, loss=0.779, lr=1.33e-6] Steps: 98%|█████████▊| 490/500 [31:25<00:34, 3.45s/it, loss=0.955, lr=1.1e-6] Steps: 98%|█████████▊| 491/500 [31:26<00:26, 2.97s/it, loss=0.955, lr=1.1e-6] Steps: 98%|█████████▊| 491/500 [31:26<00:26, 2.97s/it, loss=0.839, lr=8.88e-7] Steps: 98%|█████████▊| 492/500 [31:28<00:21, 2.64s/it, loss=0.839, lr=8.88e-7] Steps: 98%|█████████▊| 492/500 [31:28<00:21, 2.64s/it, loss=1.06, lr=7.01e-7] Steps: 99%|█████████▊| 493/500 [31:36<00:29, 4.19s/it, loss=1.06, lr=7.01e-7] Steps: 99%|█████████▊| 493/500 [31:36<00:29, 4.19s/it, loss=0.975, lr=5.37e-7] Steps: 99%|█████████▉| 494/500 [31:38<00:20, 3.49s/it, loss=0.975, lr=5.37e-7] Steps: 99%|█████████▉| 494/500 [31:38<00:20, 3.49s/it, loss=0.987, lr=3.95e-7] Steps: 99%|█████████▉| 495/500 [31:40<00:15, 3.01s/it, loss=0.987, lr=3.95e-7] Steps: 99%|█████████▉| 495/500 [31:40<00:15, 3.01s/it, loss=0.783, lr=2.74e-7] Steps: 99%|█████████▉| 496/500 [31:42<00:10, 2.67s/it, loss=0.783, lr=2.74e-7] Steps: 99%|█████████▉| 496/500 [31:42<00:10, 2.67s/it, loss=0.941, lr=1.75e-7] Steps: 99%|█████████▉| 497/500 [31:49<00:12, 4.18s/it, loss=0.941, lr=1.75e-7] Steps: 99%|█████████▉| 497/500 [31:49<00:12, 4.18s/it, loss=0.78, lr=9.87e-8] Steps: 100%|█████████▉| 498/500 [31:51<00:06, 3.49s/it, loss=0.78, lr=9.87e-8] Steps: 100%|█████████▉| 498/500 [31:51<00:06, 3.49s/it, loss=0.928, lr=4.39e-8] Steps: 100%|█████████▉| 499/500 [31:53<00:03, 3.00s/it, loss=0.928, lr=4.39e-8] Steps: 100%|█████████▉| 499/500 [31:53<00:03, 3.00s/it, loss=1.03, lr=1.1e-8] Steps: 100%|██████████| 500/500 [31:55<00:00, 2.66s/it, loss=1.03, lr=1.1e-8] Steps: 100%|██████████| 500/500 [31:55<00:00, 2.66s/it, loss=0.906, lr=0] Steps: 100%|██████████| 500/500 [31:59<00:00, 3.84s/it, loss=0.906, lr=0] ---Tar up output directory--- mochi-lora/ mochi-lora/pytorch_lora_weights.safetensors Uploading to Hugging Face: lucataco/mochi-lora-vhs HF Repo URL: https://huggingface.co/lucataco/mochi-lora-vhs pytorch_lora_weights.safetensors: 0%| | 0.00/76.1M [00:00<?, ?B/s] pytorch_lora_weights.safetensors: 10%|▉ | 7.34M/76.1M [00:00<00:00, 73.4MB/s] pytorch_lora_weights.safetensors: 21%|██ | 16.0M/76.1M [00:00<00:01, 42.3MB/s] pytorch_lora_weights.safetensors: 42%|████▏ | 32.0M/76.1M [00:00<00:00, 46.4MB/s] pytorch_lora_weights.safetensors: 63%|██████▎ | 48.0M/76.1M [00:00<00:00, 54.6MB/s] pytorch_lora_weights.safetensors: 84%|████████▍ | 64.0M/76.1M [00:01<00:00, 57.3MB/s] pytorch_lora_weights.safetensors: 100%|██████████| 76.1M/76.1M [00:01<00:00, 54.6MB/s] Successfully uploaded model to https://huggingface.co/lucataco/mochi-lora-vhs
Prediction
genmoai/mochi-1-lora-trainer:9f7c62841ad839d7091201100d03dd778e2a78ca702fa65addd69ecbd511f64eIDjnmeawnxn1rma0ckqgvr6t5q28StatusSucceededSourceWebHardwareH100Total durationCreatedInput
- seed
- 42
- steps
- 500
- hf_token
- ████████████████████
This value was redacted after being sent to the model.
- optimizer
- adamw
- batch_size
- 1
- hf_repo_id
- lucataco/mochi-lora-melty
- compile_dit
- input_videos
- melty-3.zip
- learning_rate
- 0.0004
- trim_and_crop
- caption_dropout
- 0.1
{ "seed": 42, "steps": 500, "hf_token": "[REDACTED]", "optimizer": "adamw", "batch_size": 1, "hf_repo_id": "lucataco/mochi-lora-melty", "compile_dit": true, "input_videos": "https://replicate.delivery/pbxt/M8I1KDz03YouBVLkP2ughdR6l60xX1EYQ9x5Cza2lcvKDmLY/melty-3.zip", "learning_rate": 0.0004, "trim_and_crop": true, "caption_dropout": 0.1 }
Install Replicate’s Node.js client library:npm install replicate
Import and set up the client:import Replicate from "replicate"; const replicate = new Replicate({ auth: process.env.REPLICATE_API_TOKEN, });
Run genmoai/mochi-1-lora-trainer using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "genmoai/mochi-1-lora-trainer:9f7c62841ad839d7091201100d03dd778e2a78ca702fa65addd69ecbd511f64e", { input: { seed: 42, steps: 500, hf_token: "[REDACTED]", optimizer: "adamw", batch_size: 1, hf_repo_id: "lucataco/mochi-lora-melty", compile_dit: true, input_videos: "https://replicate.delivery/pbxt/M8I1KDz03YouBVLkP2ughdR6l60xX1EYQ9x5Cza2lcvKDmLY/melty-3.zip", learning_rate: 0.0004, trim_and_crop: true, caption_dropout: 0.1 } } ); console.log(output);
To learn more, take a look at the guide on getting started with Node.js.
Install Replicate’s Python client library:pip install replicate
Import the client:import replicate
Run genmoai/mochi-1-lora-trainer using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "genmoai/mochi-1-lora-trainer:9f7c62841ad839d7091201100d03dd778e2a78ca702fa65addd69ecbd511f64e", input={ "seed": 42, "steps": 500, "hf_token": "[REDACTED]", "optimizer": "adamw", "batch_size": 1, "hf_repo_id": "lucataco/mochi-lora-melty", "compile_dit": True, "input_videos": "https://replicate.delivery/pbxt/M8I1KDz03YouBVLkP2ughdR6l60xX1EYQ9x5Cza2lcvKDmLY/melty-3.zip", "learning_rate": 0.0004, "trim_and_crop": True, "caption_dropout": 0.1 } ) print(output)
To learn more, take a look at the guide on getting started with Python.
Run genmoai/mochi-1-lora-trainer using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
curl -s -X POST \ -H "Authorization: Bearer $REPLICATE_API_TOKEN" \ -H "Content-Type: application/json" \ -H "Prefer: wait" \ -d $'{ "version": "9f7c62841ad839d7091201100d03dd778e2a78ca702fa65addd69ecbd511f64e", "input": { "seed": 42, "steps": 500, "hf_token": "[REDACTED]", "optimizer": "adamw", "batch_size": 1, "hf_repo_id": "lucataco/mochi-lora-melty", "compile_dit": true, "input_videos": "https://replicate.delivery/pbxt/M8I1KDz03YouBVLkP2ughdR6l60xX1EYQ9x5Cza2lcvKDmLY/melty-3.zip", "learning_rate": 0.0004, "trim_and_crop": true, "caption_dropout": 0.1 } }' \ https://api.replicate.com/v1/predictions
To learn more, take a look at Replicate’s HTTP API reference docs.
Output
{ "completed_at": "2024-12-12T19:33:33.783873Z", "created_at": "2024-12-12T18:53:09.928000Z", "data_removed": false, "error": null, "id": "jnmeawnxn1rma0ckqgvr6t5q28", "input": { "seed": 42, "steps": 500, "hf_token": "[REDACTED]", "optimizer": "adamw", "batch_size": 1, "hf_repo_id": "lucataco/mochi-lora-melty", "compile_dit": true, "input_videos": "https://replicate.delivery/pbxt/M8I1KDz03YouBVLkP2ughdR6l60xX1EYQ9x5Cza2lcvKDmLY/melty-3.zip", "learning_rate": 0.0004, "trim_and_crop": true, "caption_dropout": 0.1 }, "logs": "Cleaning up previous runs\nExtracted 6 files from zip to videos_input\n---Starting to Trim input videos---\nProcessing: videos_input/cat.mov\nvideos_input/cat.mov as target resolution 480x848 is larger than input 668x650. So, upsampling the video.\nCopied videos_input/cat.txt to videos_prepared/cat.txt\nMoviepy - Building video videos_prepared/cat.mp4.\nMoviepy - Writing video videos_prepared/cat.mp4\n 0%| | 0/3 [00:00<?, ?it/s]\n0%| | 0/3 [00:00<?, ?it/s]\n 0%| | 0/3 [00:00<?, ?it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 65%|██████▌ | 26/40 [00:00<00:00, 254.70it/s, now=None]\u001b[A\n \u001b[A\n0%| | 0/3 [00:00<?, ?it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/cat.mp4\n 0%| | 0/3 [00:00<?, ?it/s]\nProcessing: videos_input/glove.mov\nCopied videos_input/glove.txt to videos_prepared/glove.txt\nMoviepy - Building video videos_prepared/glove.mp4.\nMoviepy - Writing video videos_prepared/glove.mp4\n 33%|███▎ | 1/3 [00:00<00:00, 2.34it/s]\n33%|███▎ | 1/3 [00:00<00:00, 2.34it/s]\n 33%|███▎ | 1/3 [00:00<00:00, 2.34it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 15%|█▌ | 6/40 [00:00<00:00, 58.08it/s, now=None]\u001b[A\nt: 30%|███ | 12/40 [00:00<00:00, 53.57it/s, now=None]\u001b[A\nt: 45%|████▌ | 18/40 [00:00<00:00, 52.35it/s, now=None]\u001b[A\nt: 60%|██████ | 24/40 [00:00<00:00, 51.47it/s, now=None]\u001b[A\nt: 75%|███████▌ | 30/40 [00:00<00:00, 50.92it/s, now=None]\u001b[A\nt: 90%|█████████ | 36/40 [00:00<00:00, 50.87it/s, now=None]\u001b[A\n \u001b[A\n33%|███▎ | 1/3 [00:01<00:00, 2.34it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/glove.mp4\n 33%|███▎ | 1/3 [00:01<00:00, 2.34it/s]\nProcessing: videos_input/straws.mov\nvideos_input/straws.mov as target resolution 480x848 is larger than input 668x514. So, upsampling the video.\nCopied videos_input/straws.txt to videos_prepared/straws.txt\nMoviepy - Building video videos_prepared/straws.mp4.\nMoviepy - Writing video videos_prepared/straws.mp4\n 67%|██████▋ | 2/3 [00:01<00:00, 1.17it/s]\n67%|██████▋ | 2/3 [00:01<00:00, 1.17it/s]\n 67%|██████▋ | 2/3 [00:01<00:00, 1.17it/s]\nt: 0%| | 0/40 [00:00<?, ?it/s, now=None]\u001b[A\nt: 72%|███████▎ | 29/40 [00:00<00:00, 286.01it/s, now=None]\u001b[A\n \u001b[A\n67%|██████▋ | 2/3 [00:01<00:00, 1.17it/s]\nMoviepy - Done !\nMoviepy - video ready videos_prepared/straws.mp4\n 67%|██████▋ | 2/3 [00:01<00:00, 1.17it/s]\n100%|██████████| 3/3 [00:01<00:00, 1.59it/s]\n100%|██████████| 3/3 [00:01<00:00, 1.54it/s]\n---Starting to Embed videos---\nLoading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s]\nLoading checkpoint shards: 50%|█████ | 1/2 [00:00<00:00, 1.84it/s]\nLoading checkpoint shards: 100%|██████████| 2/2 [00:01<00:00, 1.88it/s]\nLoading checkpoint shards: 100%|██████████| 2/2 [00:01<00:00, 1.87it/s]\nLoading pipeline components...: 0%| | 0/3 [00:00<?, ?it/s]\nLoading pipeline components...: 100%|██████████| 3/3 [00:00<00:00, 670.55it/s]\nProcessing videos_prepared/cat.mp4\nTrimmed video from 40 to first 37 frames\n0it [00:00, ?it/s]\nProcessing videos_prepared/glove.mp4\nTrimmed video from 40 to first 37 frames\n1it [00:01, 1.64s/it]\nProcessing videos_prepared/straws.mp4\nTrimmed video from 40 to first 37 frames\n2it [00:02, 1.30s/it]\n3it [00:03, 1.15s/it]\n3it [00:03, 1.23s/it]\n---Starting training---\nFound 3 training videos in videos_prepared\nLoaded 3/3 valid file pairs.\n===== Memory before training =====\nmemory_allocated=18.903 GB\nmax_memory_allocated=18.903 GB\nmax_memory_reserved=28.078 GB\n***** Running training *****\nNum trainable parameters = 19005440\nNum examples = 3\nNum batches each epoch = 3\nNum epochs = 167\nInstantaneous batch size per device = 1\nTotal train batch size (w. parallel, distributed & accumulation) = 1\nTotal optimization steps = 500\nSteps: 0%| | 0/500 [00:00<?, ?it/s]W1212 18:59:29.908000 138667870809600 torch/fx/experimental/symbolic_shapes.py:4449] [0/0] xindex is not in var_ranges, defaulting to unknown range.\nW1212 18:59:29.922000 138667870809600 torch/fx/experimental/symbolic_shapes.py:4449] [0/0] xindex is not in var_ranges, defaulting to unknown range.\nW1212 18:59:30.058000 138667870809600 torch/fx/experimental/symbolic_shapes.py:4449] [0/0] xindex is not in var_ranges, defaulting to unknown range.\nSteps: 0%| | 1/500 [04:22<36:20:40, 262.21s/it]\nSteps: 0%| | 1/500 [04:22<36:20:40, 262.21s/it, loss=0.758, lr=2e-6]\nSteps: 0%| | 2/500 [04:24<15:05:07, 109.05s/it, loss=0.758, lr=2e-6]\nSteps: 0%| | 2/500 [04:24<15:05:07, 109.05s/it, loss=0.492, lr=4e-6]\nSteps: 1%| | 3/500 [04:25<8:17:55, 60.11s/it, loss=0.492, lr=4e-6] \nSteps: 1%| | 3/500 [04:25<8:17:55, 60.11s/it, loss=0.291, lr=6e-6]\nSteps: 1%| | 4/500 [04:34<5:27:59, 39.68s/it, loss=0.291, lr=6e-6]\nSteps: 1%| | 4/500 [04:34<5:27:59, 39.68s/it, loss=0.581, lr=8e-6]\nSteps: 1%| | 5/500 [04:36<3:34:51, 26.04s/it, loss=0.581, lr=8e-6]\nSteps: 1%| | 5/500 [04:36<3:34:51, 26.04s/it, loss=0.675, lr=1e-5]\nSteps: 1%| | 6/500 [04:38<2:26:44, 17.82s/it, loss=0.675, lr=1e-5]\nSteps: 1%| | 6/500 [04:38<2:26:44, 17.82s/it, loss=0.387, lr=1.2e-5]\nSteps: 1%|▏ | 7/500 [04:46<2:02:39, 14.93s/it, loss=0.387, lr=1.2e-5]\nSteps: 1%|▏ | 7/500 [04:46<2:02:39, 14.93s/it, loss=1.16, lr=1.4e-5] \nSteps: 2%|▏ | 8/500 [04:48<1:28:19, 10.77s/it, loss=1.16, lr=1.4e-5]\nSteps: 2%|▏ | 8/500 [04:48<1:28:19, 10.77s/it, loss=0.585, lr=1.6e-5]\nSteps: 2%|▏ | 9/500 [04:50<1:05:23, 7.99s/it, loss=0.585, lr=1.6e-5]\nSteps: 2%|▏ | 9/500 [04:50<1:05:23, 7.99s/it, loss=0.294, lr=1.8e-5]\nSteps: 2%|▏ | 10/500 [04:59<1:06:54, 8.19s/it, loss=0.294, lr=1.8e-5]\nSteps: 2%|▏ | 10/500 [04:59<1:06:54, 8.19s/it, loss=0.598, lr=2e-5] \nSteps: 2%|▏ | 11/500 [05:01<51:00, 6.26s/it, loss=0.598, lr=2e-5] \nSteps: 2%|▏ | 11/500 [05:01<51:00, 6.26s/it, loss=0.949, lr=2.2e-5]\nSteps: 2%|▏ | 12/500 [05:03<40:03, 4.93s/it, loss=0.949, lr=2.2e-5]\nSteps: 2%|▏ | 12/500 [05:03<40:03, 4.93s/it, loss=0.543, lr=2.4e-5]\nSteps: 3%|▎ | 13/500 [05:10<46:42, 5.75s/it, loss=0.543, lr=2.4e-5]\nSteps: 3%|▎ | 13/500 [05:10<46:42, 5.75s/it, loss=0.574, lr=2.6e-5]\nSteps: 3%|▎ | 14/500 [05:12<37:08, 4.58s/it, loss=0.574, lr=2.6e-5]\nSteps: 3%|▎ | 14/500 [05:12<37:08, 4.58s/it, loss=0.671, lr=2.8e-5]\nSteps: 3%|▎ | 15/500 [05:14<30:27, 3.77s/it, loss=0.671, lr=2.8e-5]\nSteps: 3%|▎ | 15/500 [05:14<30:27, 3.77s/it, loss=0.666, lr=3e-5] \nSteps: 3%|▎ | 16/500 [05:22<39:55, 4.95s/it, loss=0.666, lr=3e-5]\nSteps: 3%|▎ | 16/500 [05:22<39:55, 4.95s/it, loss=1.08, lr=3.2e-5]\nSteps: 3%|▎ | 17/500 [05:24<32:23, 4.02s/it, loss=1.08, lr=3.2e-5]\nSteps: 3%|▎ | 17/500 [05:24<32:23, 4.02s/it, loss=1.05, lr=3.4e-5]\nSteps: 4%|▎ | 18/500 [05:25<27:08, 3.38s/it, loss=1.05, lr=3.4e-5]\nSteps: 4%|▎ | 18/500 [05:25<27:08, 3.38s/it, loss=0.292, lr=3.6e-5]\nSteps: 4%|▍ | 19/500 [05:33<37:18, 4.65s/it, loss=0.292, lr=3.6e-5]\nSteps: 4%|▍ | 19/500 [05:33<37:18, 4.65s/it, loss=0.574, lr=3.8e-5]\nSteps: 4%|▍ | 20/500 [05:35<30:34, 3.82s/it, loss=0.574, lr=3.8e-5]\nSteps: 4%|▍ | 20/500 [05:35<30:34, 3.82s/it, loss=0.575, lr=4e-5] \nSteps: 4%|▍ | 21/500 [05:37<25:51, 3.24s/it, loss=0.575, lr=4e-5]\nSteps: 4%|▍ | 21/500 [05:37<25:51, 3.24s/it, loss=0.855, lr=4.2e-5]\nSteps: 4%|▍ | 22/500 [05:45<36:30, 4.58s/it, loss=0.855, lr=4.2e-5]\nSteps: 4%|▍ | 22/500 [05:45<36:30, 4.58s/it, loss=0.625, lr=4.4e-5]\nSteps: 5%|▍ | 23/500 [05:46<29:59, 3.77s/it, loss=0.625, lr=4.4e-5]\nSteps: 5%|▍ | 23/500 [05:46<29:59, 3.77s/it, loss=0.851, lr=4.6e-5]\nSteps: 5%|▍ | 24/500 [05:48<25:25, 3.20s/it, loss=0.851, lr=4.6e-5]\nSteps: 5%|▍ | 24/500 [05:48<25:25, 3.20s/it, loss=0.878, lr=4.8e-5]\nSteps: 5%|▌ | 25/500 [05:56<35:52, 4.53s/it, loss=0.878, lr=4.8e-5]\nSteps: 5%|▌ | 25/500 [05:56<35:52, 4.53s/it, loss=0.626, lr=5e-5] \nSteps: 5%|▌ | 26/500 [05:58<29:30, 3.73s/it, loss=0.626, lr=5e-5]\nSteps: 5%|▌ | 26/500 [05:58<29:30, 3.73s/it, loss=0.46, lr=5.2e-5]\nSteps: 5%|▌ | 27/500 [06:00<25:04, 3.18s/it, loss=0.46, lr=5.2e-5]\nSteps: 5%|▌ | 27/500 [06:00<25:04, 3.18s/it, loss=0.653, lr=5.4e-5]\nSteps: 6%|▌ | 28/500 [06:07<35:41, 4.54s/it, loss=0.653, lr=5.4e-5]\nSteps: 6%|▌ | 28/500 [06:07<35:41, 4.54s/it, loss=0.579, lr=5.6e-5]\nSteps: 6%|▌ | 29/500 [06:09<29:20, 3.74s/it, loss=0.579, lr=5.6e-5]\nSteps: 6%|▌ | 29/500 [06:09<29:20, 3.74s/it, loss=0.698, lr=5.8e-5]\nSteps: 6%|▌ | 30/500 [06:11<24:54, 3.18s/it, loss=0.698, lr=5.8e-5]\nSteps: 6%|▌ | 30/500 [06:11<24:54, 3.18s/it, loss=0.557, lr=6e-5] \nSteps: 6%|▌ | 31/500 [06:19<35:33, 4.55s/it, loss=0.557, lr=6e-5]\nSteps: 6%|▌ | 31/500 [06:19<35:33, 4.55s/it, loss=1.19, lr=6.2e-5]\nSteps: 6%|▋ | 32/500 [06:21<29:13, 3.75s/it, loss=1.19, lr=6.2e-5]\nSteps: 6%|▋ | 32/500 [06:21<29:13, 3.75s/it, loss=0.738, lr=6.4e-5]\nSteps: 7%|▋ | 33/500 [06:23<24:48, 3.19s/it, loss=0.738, lr=6.4e-5]\nSteps: 7%|▋ | 33/500 [06:23<24:48, 3.19s/it, loss=0.294, lr=6.6e-5]\nSteps: 7%|▋ | 34/500 [06:30<35:21, 4.55s/it, loss=0.294, lr=6.6e-5]\nSteps: 7%|▋ | 34/500 [06:30<35:21, 4.55s/it, loss=0.585, lr=6.8e-5]\nSteps: 7%|▋ | 35/500 [06:32<29:03, 3.75s/it, loss=0.585, lr=6.8e-5]\nSteps: 7%|▋ | 35/500 [06:32<29:03, 3.75s/it, loss=0.594, lr=7e-5] \nSteps: 7%|▋ | 36/500 [06:34<24:40, 3.19s/it, loss=0.594, lr=7e-5]\nSteps: 7%|▋ | 36/500 [06:34<24:40, 3.19s/it, loss=0.303, lr=7.2e-5]\nSteps: 7%|▋ | 37/500 [06:42<34:55, 4.52s/it, loss=0.303, lr=7.2e-5]\nSteps: 7%|▋ | 37/500 [06:42<34:55, 4.52s/it, loss=0.561, lr=7.4e-5]\nSteps: 8%|▊ | 38/500 [06:44<28:43, 3.73s/it, loss=0.561, lr=7.4e-5]\nSteps: 8%|▊ | 38/500 [06:44<28:43, 3.73s/it, loss=0.658, lr=7.6e-5]\nSteps: 8%|▊ | 39/500 [06:46<24:23, 3.18s/it, loss=0.658, lr=7.6e-5]\nSteps: 8%|▊ | 39/500 [06:46<24:23, 3.18s/it, loss=0.299, lr=7.8e-5]\nSteps: 8%|▊ | 40/500 [06:53<34:23, 4.49s/it, loss=0.299, lr=7.8e-5]\nSteps: 8%|▊ | 40/500 [06:53<34:23, 4.49s/it, loss=0.773, lr=8e-5] \nSteps: 8%|▊ | 41/500 [06:55<28:19, 3.70s/it, loss=0.773, lr=8e-5]\nSteps: 8%|▊ | 41/500 [06:55<28:19, 3.70s/it, loss=0.454, lr=8.2e-5]\nSteps: 8%|▊ | 42/500 [06:57<24:05, 3.16s/it, loss=0.454, lr=8.2e-5]\nSteps: 8%|▊ | 42/500 [06:57<24:05, 3.16s/it, loss=0.344, lr=8.4e-5]\nSteps: 9%|▊ | 43/500 [07:05<34:21, 4.51s/it, loss=0.344, lr=8.4e-5]\nSteps: 9%|▊ | 43/500 [07:05<34:21, 4.51s/it, loss=0.608, lr=8.6e-5]\nSteps: 9%|▉ | 44/500 [07:06<28:17, 3.72s/it, loss=0.608, lr=8.6e-5]\nSteps: 9%|▉ | 44/500 [07:06<28:17, 3.72s/it, loss=0.461, lr=8.8e-5]\nSteps: 9%|▉ | 45/500 [07:08<24:01, 3.17s/it, loss=0.461, lr=8.8e-5]\nSteps: 9%|▉ | 45/500 [07:08<24:01, 3.17s/it, loss=0.291, lr=9e-5] \nSteps: 9%|▉ | 46/500 [07:16<34:17, 4.53s/it, loss=0.291, lr=9e-5]\nSteps: 9%|▉ | 46/500 [07:16<34:17, 4.53s/it, loss=0.671, lr=9.2e-5]\nSteps: 9%|▉ | 47/500 [07:18<28:11, 3.73s/it, loss=0.671, lr=9.2e-5]\nSteps: 9%|▉ | 47/500 [07:18<28:11, 3.73s/it, loss=0.6, lr=9.4e-5] \nSteps: 10%|▉ | 48/500 [07:20<23:56, 3.18s/it, loss=0.6, lr=9.4e-5]\nSteps: 10%|▉ | 48/500 [07:20<23:56, 3.18s/it, loss=0.285, lr=9.6e-5]\nSteps: 10%|▉ | 49/500 [07:27<34:01, 4.53s/it, loss=0.285, lr=9.6e-5]\nSteps: 10%|▉ | 49/500 [07:27<34:01, 4.53s/it, loss=1.07, lr=9.8e-5] \nSteps: 10%|█ | 50/500 [07:29<27:59, 3.73s/it, loss=1.07, lr=9.8e-5]\nSteps: 10%|█ | 50/500 [07:29<27:59, 3.73s/it, loss=0.633, lr=0.0001]\nSteps: 10%|█ | 51/500 [07:31<23:46, 3.18s/it, loss=0.633, lr=0.0001]\nSteps: 10%|█ | 51/500 [07:31<23:46, 3.18s/it, loss=0.29, lr=0.000102]\nSteps: 10%|█ | 52/500 [07:39<33:35, 4.50s/it, loss=0.29, lr=0.000102]\nSteps: 10%|█ | 52/500 [07:39<33:35, 4.50s/it, loss=0.555, lr=0.000104]\nSteps: 11%|█ | 53/500 [07:41<27:39, 3.71s/it, loss=0.555, lr=0.000104]\nSteps: 11%|█ | 53/500 [07:41<27:39, 3.71s/it, loss=0.457, lr=0.000106]\nSteps: 11%|█ | 54/500 [07:43<23:30, 3.16s/it, loss=0.457, lr=0.000106]\nSteps: 11%|█ | 54/500 [07:43<23:30, 3.16s/it, loss=0.365, lr=0.000108]\nSteps: 11%|█ | 55/500 [07:50<33:32, 4.52s/it, loss=0.365, lr=0.000108]\nSteps: 11%|█ | 55/500 [07:50<33:32, 4.52s/it, loss=0.618, lr=0.00011] \nSteps: 11%|█ | 56/500 [07:52<27:35, 3.73s/it, loss=0.618, lr=0.00011]\nSteps: 11%|█ | 56/500 [07:52<27:35, 3.73s/it, loss=0.464, lr=0.000112]\nSteps: 11%|█▏ | 57/500 [07:54<23:25, 3.17s/it, loss=0.464, lr=0.000112]\nSteps: 11%|█▏ | 57/500 [07:54<23:25, 3.17s/it, loss=0.283, lr=0.000114]\nSteps: 12%|█▏ | 58/500 [08:02<33:09, 4.50s/it, loss=0.283, lr=0.000114]\nSteps: 12%|█▏ | 58/500 [08:02<33:09, 4.50s/it, loss=0.575, lr=0.000116]\nSteps: 12%|█▏ | 59/500 [08:03<27:18, 3.72s/it, loss=0.575, lr=0.000116]\nSteps: 12%|█▏ | 59/500 [08:04<27:18, 3.72s/it, loss=0.539, lr=0.000118]\nSteps: 12%|█▏ | 60/500 [08:05<23:12, 3.16s/it, loss=0.539, lr=0.000118]\nSteps: 12%|█▏ | 60/500 [08:05<23:12, 3.16s/it, loss=0.595, lr=0.00012] \nSteps: 12%|█▏ | 61/500 [08:13<33:02, 4.52s/it, loss=0.595, lr=0.00012]\nSteps: 12%|█▏ | 61/500 [08:13<33:02, 4.52s/it, loss=1.18, lr=0.000122]\nSteps: 12%|█▏ | 62/500 [08:15<27:11, 3.73s/it, loss=1.18, lr=0.000122]\nSteps: 12%|█▏ | 62/500 [08:15<27:11, 3.73s/it, loss=0.447, lr=0.000124]\nSteps: 13%|█▎ | 63/500 [08:17<23:06, 3.17s/it, loss=0.447, lr=0.000124]\nSteps: 13%|█▎ | 63/500 [08:17<23:06, 3.17s/it, loss=0.352, lr=0.000126]\nSteps: 13%|█▎ | 64/500 [08:24<32:43, 4.50s/it, loss=0.352, lr=0.000126]\nSteps: 13%|█▎ | 64/500 [08:24<32:43, 4.50s/it, loss=0.631, lr=0.000128]\nSteps: 13%|█▎ | 65/500 [08:26<26:56, 3.72s/it, loss=0.631, lr=0.000128]\nSteps: 13%|█▎ | 65/500 [08:26<26:56, 3.72s/it, loss=0.571, lr=0.00013] \nSteps: 13%|█▎ | 66/500 [08:28<22:54, 3.17s/it, loss=0.571, lr=0.00013]\nSteps: 13%|█▎ | 66/500 [08:28<22:54, 3.17s/it, loss=0.544, lr=0.000132]\nSteps: 13%|█▎ | 67/500 [08:36<32:43, 4.53s/it, loss=0.544, lr=0.000132]\nSteps: 13%|█▎ | 67/500 [08:36<32:43, 4.53s/it, loss=0.549, lr=0.000134]\nSteps: 14%|█▎ | 68/500 [08:38<26:55, 3.74s/it, loss=0.549, lr=0.000134]\nSteps: 14%|█▎ | 68/500 [08:38<26:55, 3.74s/it, loss=0.488, lr=0.000136]\nSteps: 14%|█▍ | 69/500 [08:40<22:51, 3.18s/it, loss=0.488, lr=0.000136]\nSteps: 14%|█▍ | 69/500 [08:40<22:51, 3.18s/it, loss=0.983, lr=0.000138]\nSteps: 14%|█▍ | 70/500 [08:47<32:09, 4.49s/it, loss=0.983, lr=0.000138]\nSteps: 14%|█▍ | 70/500 [08:47<32:09, 4.49s/it, loss=0.544, lr=0.00014] \nSteps: 14%|█▍ | 71/500 [08:49<26:28, 3.70s/it, loss=0.544, lr=0.00014]\nSteps: 14%|█▍ | 71/500 [08:49<26:28, 3.70s/it, loss=0.832, lr=0.000142]\nSteps: 14%|█▍ | 72/500 [08:51<22:31, 3.16s/it, loss=0.832, lr=0.000142]\nSteps: 14%|█▍ | 72/500 [08:51<22:31, 3.16s/it, loss=0.288, lr=0.000144]\nSteps: 15%|█▍ | 73/500 [08:59<32:10, 4.52s/it, loss=0.288, lr=0.000144]\nSteps: 15%|█▍ | 73/500 [08:59<32:10, 4.52s/it, loss=0.844, lr=0.000146]\nSteps: 15%|█▍ | 74/500 [09:01<26:29, 3.73s/it, loss=0.844, lr=0.000146]\nSteps: 15%|█▍ | 74/500 [09:01<26:29, 3.73s/it, loss=0.457, lr=0.000148]\nSteps: 15%|█▌ | 75/500 [09:02<22:29, 3.18s/it, loss=0.457, lr=0.000148]\nSteps: 15%|█▌ | 75/500 [09:02<22:29, 3.18s/it, loss=0.39, lr=0.00015] \nSteps: 15%|█▌ | 76/500 [09:10<31:42, 4.49s/it, loss=0.39, lr=0.00015]\nSteps: 15%|█▌ | 76/500 [09:10<31:42, 4.49s/it, loss=0.668, lr=0.000152]\nSteps: 15%|█▌ | 77/500 [09:12<26:07, 3.70s/it, loss=0.668, lr=0.000152]\nSteps: 15%|█▌ | 77/500 [09:12<26:07, 3.70s/it, loss=1, lr=0.000154] \nSteps: 16%|█▌ | 78/500 [09:14<22:12, 3.16s/it, loss=1, lr=0.000154]\nSteps: 16%|█▌ | 78/500 [09:14<22:12, 3.16s/it, loss=0.336, lr=0.000156]\nSteps: 16%|█▌ | 79/500 [09:21<31:36, 4.50s/it, loss=0.336, lr=0.000156]\nSteps: 16%|█▌ | 79/500 [09:21<31:36, 4.50s/it, loss=0.569, lr=0.000158]\nSteps: 16%|█▌ | 80/500 [09:23<26:01, 3.72s/it, loss=0.569, lr=0.000158]\nSteps: 16%|█▌ | 80/500 [09:23<26:01, 3.72s/it, loss=0.543, lr=0.00016] \nSteps: 16%|█▌ | 81/500 [09:25<22:08, 3.17s/it, loss=0.543, lr=0.00016]\nSteps: 16%|█▌ | 81/500 [09:25<22:08, 3.17s/it, loss=0.703, lr=0.000162]\nSteps: 16%|█▋ | 82/500 [09:33<31:23, 4.51s/it, loss=0.703, lr=0.000162]\nSteps: 16%|█▋ | 82/500 [09:33<31:23, 4.51s/it, loss=0.54, lr=0.000164] \nSteps: 17%|█▋ | 83/500 [09:35<25:50, 3.72s/it, loss=0.54, lr=0.000164]\nSteps: 17%|█▋ | 83/500 [09:35<25:50, 3.72s/it, loss=0.485, lr=0.000166]\nSteps: 17%|█▋ | 84/500 [09:37<21:58, 3.17s/it, loss=0.485, lr=0.000166]\nSteps: 17%|█▋ | 84/500 [09:37<21:58, 3.17s/it, loss=0.348, lr=0.000168]\nSteps: 17%|█▋ | 85/500 [09:44<31:12, 4.51s/it, loss=0.348, lr=0.000168]\nSteps: 17%|█▋ | 85/500 [09:44<31:12, 4.51s/it, loss=0.587, lr=0.00017] \nSteps: 17%|█▋ | 86/500 [09:46<25:41, 3.72s/it, loss=0.587, lr=0.00017]\nSteps: 17%|█▋ | 86/500 [09:46<25:41, 3.72s/it, loss=0.626, lr=0.000172]\nSteps: 17%|█▋ | 87/500 [09:48<21:50, 3.17s/it, loss=0.626, lr=0.000172]\nSteps: 17%|█▋ | 87/500 [09:48<21:50, 3.17s/it, loss=0.827, lr=0.000174]\nSteps: 18%|█▊ | 88/500 [09:56<31:01, 4.52s/it, loss=0.827, lr=0.000174]\nSteps: 18%|█▊ | 88/500 [09:56<31:01, 4.52s/it, loss=0.877, lr=0.000176]\nSteps: 18%|█▊ | 89/500 [09:57<25:31, 3.73s/it, loss=0.877, lr=0.000176]\nSteps: 18%|█▊ | 89/500 [09:58<25:31, 3.73s/it, loss=0.614, lr=0.000178]\nSteps: 18%|█▊ | 90/500 [09:59<21:41, 3.17s/it, loss=0.614, lr=0.000178]\nSteps: 18%|█▊ | 90/500 [09:59<21:41, 3.17s/it, loss=0.416, lr=0.00018] \nSteps: 18%|█▊ | 91/500 [10:07<30:52, 4.53s/it, loss=0.416, lr=0.00018]\nSteps: 18%|█▊ | 91/500 [10:07<30:52, 4.53s/it, loss=0.654, lr=0.000182]\nSteps: 18%|█▊ | 92/500 [10:09<25:23, 3.73s/it, loss=0.654, lr=0.000182]\nSteps: 18%|█▊ | 92/500 [10:09<25:23, 3.73s/it, loss=0.689, lr=0.000184]\nSteps: 19%|█▊ | 93/500 [10:11<21:33, 3.18s/it, loss=0.689, lr=0.000184]\nSteps: 19%|█▊ | 93/500 [10:11<21:33, 3.18s/it, loss=0.295, lr=0.000186]\nSteps: 19%|█▉ | 94/500 [10:19<30:42, 4.54s/it, loss=0.295, lr=0.000186]\nSteps: 19%|█▉ | 94/500 [10:19<30:42, 4.54s/it, loss=0.975, lr=0.000188]\nSteps: 19%|█▉ | 95/500 [10:20<25:15, 3.74s/it, loss=0.975, lr=0.000188]\nSteps: 19%|█▉ | 95/500 [10:20<25:15, 3.74s/it, loss=0.585, lr=0.00019] \nSteps: 19%|█▉ | 96/500 [10:22<21:26, 3.18s/it, loss=0.585, lr=0.00019]\nSteps: 19%|█▉ | 96/500 [10:22<21:26, 3.18s/it, loss=1.01, lr=0.000192]\nSteps: 19%|█▉ | 97/500 [10:30<30:34, 4.55s/it, loss=1.01, lr=0.000192]\nSteps: 19%|█▉ | 97/500 [10:30<30:34, 4.55s/it, loss=0.605, lr=0.000194]\nSteps: 20%|█▉ | 98/500 [10:32<25:08, 3.75s/it, loss=0.605, lr=0.000194]\nSteps: 20%|█▉ | 98/500 [10:32<25:08, 3.75s/it, loss=0.681, lr=0.000196]\nSteps: 20%|█▉ | 99/500 [10:34<21:18, 3.19s/it, loss=0.681, lr=0.000196]\nSteps: 20%|█▉ | 99/500 [10:34<21:18, 3.19s/it, loss=0.431, lr=0.000198]\nSteps: 20%|██ | 100/500 [10:42<30:25, 4.56s/it, loss=0.431, lr=0.000198]\nSteps: 20%|██ | 100/500 [10:42<30:25, 4.56s/it, loss=0.549, lr=0.0002] \nSteps: 20%|██ | 101/500 [10:43<24:59, 3.76s/it, loss=0.549, lr=0.0002]\nSteps: 20%|██ | 101/500 [10:43<24:59, 3.76s/it, loss=0.541, lr=0.000202]\nSteps: 20%|██ | 102/500 [10:45<21:12, 3.20s/it, loss=0.541, lr=0.000202]\nSteps: 20%|██ | 102/500 [10:45<21:12, 3.20s/it, loss=0.434, lr=0.000204]\nSteps: 21%|██ | 103/500 [10:53<30:07, 4.55s/it, loss=0.434, lr=0.000204]\nSteps: 21%|██ | 103/500 [10:53<30:07, 4.55s/it, loss=0.813, lr=0.000206]\nSteps: 21%|██ | 104/500 [10:55<24:46, 3.75s/it, loss=0.813, lr=0.000206]\nSteps: 21%|██ | 104/500 [10:55<24:46, 3.75s/it, loss=0.48, lr=0.000208] \nSteps: 21%|██ | 105/500 [10:57<21:00, 3.19s/it, loss=0.48, lr=0.000208]\nSteps: 21%|██ | 105/500 [10:57<21:00, 3.19s/it, loss=0.568, lr=0.00021]\nSteps: 21%|██ | 106/500 [11:04<29:45, 4.53s/it, loss=0.568, lr=0.00021]\nSteps: 21%|██ | 106/500 [11:05<29:45, 4.53s/it, loss=0.535, lr=0.000212]\nSteps: 21%|██▏ | 107/500 [11:06<24:28, 3.74s/it, loss=0.535, lr=0.000212]\nSteps: 21%|██▏ | 107/500 [11:06<24:28, 3.74s/it, loss=1.08, lr=0.000214] \nSteps: 22%|██▏ | 108/500 [11:08<20:46, 3.18s/it, loss=1.08, lr=0.000214]\nSteps: 22%|██▏ | 108/500 [11:08<20:46, 3.18s/it, loss=0.41, lr=0.000216]\nSteps: 22%|██▏ | 109/500 [11:16<29:24, 4.51s/it, loss=0.41, lr=0.000216]\nSteps: 22%|██▏ | 109/500 [11:16<29:24, 4.51s/it, loss=0.535, lr=0.000218]\nSteps: 22%|██▏ | 110/500 [11:18<24:12, 3.72s/it, loss=0.535, lr=0.000218]\nSteps: 22%|██▏ | 110/500 [11:18<24:12, 3.72s/it, loss=0.543, lr=0.00022] \nSteps: 22%|██▏ | 111/500 [11:20<20:34, 3.17s/it, loss=0.543, lr=0.00022]\nSteps: 22%|██▏ | 111/500 [11:20<20:34, 3.17s/it, loss=0.978, lr=0.000222]\nSteps: 22%|██▏ | 112/500 [11:27<29:19, 4.53s/it, loss=0.978, lr=0.000222]\nSteps: 22%|██▏ | 112/500 [11:27<29:19, 4.53s/it, loss=0.569, lr=0.000224]\nSteps: 23%|██▎ | 113/500 [11:29<24:07, 3.74s/it, loss=0.569, lr=0.000224]\nSteps: 23%|██▎ | 113/500 [11:29<24:07, 3.74s/it, loss=0.497, lr=0.000226]\nSteps: 23%|██▎ | 114/500 [11:31<20:28, 3.18s/it, loss=0.497, lr=0.000226]\nSteps: 23%|██▎ | 114/500 [11:31<20:28, 3.18s/it, loss=0.59, lr=0.000228] \nSteps: 23%|██▎ | 115/500 [11:39<29:06, 4.54s/it, loss=0.59, lr=0.000228]\nSteps: 23%|██▎ | 115/500 [11:39<29:06, 4.54s/it, loss=0.532, lr=0.00023]\nSteps: 23%|██▎ | 116/500 [11:41<23:56, 3.74s/it, loss=0.532, lr=0.00023]\nSteps: 23%|██▎ | 116/500 [11:41<23:56, 3.74s/it, loss=0.543, lr=0.000232]\nSteps: 23%|██▎ | 117/500 [11:43<20:18, 3.18s/it, loss=0.543, lr=0.000232]\nSteps: 23%|██▎ | 117/500 [11:43<20:18, 3.18s/it, loss=0.781, lr=0.000234]\nSteps: 24%|██▎ | 118/500 [11:50<28:49, 4.53s/it, loss=0.781, lr=0.000234]\nSteps: 24%|██▎ | 118/500 [11:50<28:49, 4.53s/it, loss=0.533, lr=0.000236]\nSteps: 24%|██▍ | 119/500 [11:52<23:42, 3.73s/it, loss=0.533, lr=0.000236]\nSteps: 24%|██▍ | 119/500 [11:52<23:42, 3.73s/it, loss=0.45, lr=0.000238] \nSteps: 24%|██▍ | 120/500 [11:54<20:07, 3.18s/it, loss=0.45, lr=0.000238]\nSteps: 24%|██▍ | 120/500 [11:54<20:07, 3.18s/it, loss=0.285, lr=0.00024]\nSteps: 24%|██▍ | 121/500 [12:02<28:36, 4.53s/it, loss=0.285, lr=0.00024]\nSteps: 24%|██▍ | 121/500 [12:02<28:36, 4.53s/it, loss=0.544, lr=0.000242]\nSteps: 24%|██▍ | 122/500 [12:04<23:31, 3.74s/it, loss=0.544, lr=0.000242]\nSteps: 24%|██▍ | 122/500 [12:04<23:31, 3.74s/it, loss=1.03, lr=0.000244] \nSteps: 25%|██▍ | 123/500 [12:05<19:59, 3.18s/it, loss=1.03, lr=0.000244]\nSteps: 25%|██▍ | 123/500 [12:05<19:59, 3.18s/it, loss=0.448, lr=0.000246]\nSteps: 25%|██▍ | 124/500 [12:13<28:26, 4.54s/it, loss=0.448, lr=0.000246]\nSteps: 25%|██▍ | 124/500 [12:13<28:26, 4.54s/it, loss=0.662, lr=0.000248]\nSteps: 25%|██▌ | 125/500 [12:15<23:22, 3.74s/it, loss=0.662, lr=0.000248]\nSteps: 25%|██▌ | 125/500 [12:15<23:22, 3.74s/it, loss=0.6, lr=0.00025] \nSteps: 25%|██▌ | 126/500 [12:17<19:51, 3.19s/it, loss=0.6, lr=0.00025]\nSteps: 25%|██▌ | 126/500 [12:17<19:51, 3.19s/it, loss=0.291, lr=0.000252]\nSteps: 25%|██▌ | 127/500 [12:25<28:19, 4.56s/it, loss=0.291, lr=0.000252]\nSteps: 25%|██▌ | 127/500 [12:25<28:19, 4.56s/it, loss=0.999, lr=0.000254]\nSteps: 26%|██▌ | 128/500 [12:27<23:17, 3.76s/it, loss=0.999, lr=0.000254]\nSteps: 26%|██▌ | 128/500 [12:27<23:17, 3.76s/it, loss=0.446, lr=0.000256]\nSteps: 26%|██▌ | 129/500 [12:28<19:45, 3.19s/it, loss=0.446, lr=0.000256]\nSteps: 26%|██▌ | 129/500 [12:28<19:45, 3.19s/it, loss=0.297, lr=0.000258]\nSteps: 26%|██▌ | 130/500 [12:36<27:53, 4.52s/it, loss=0.297, lr=0.000258]\nSteps: 26%|██▌ | 130/500 [12:36<27:53, 4.52s/it, loss=0.8, lr=0.00026] \nSteps: 26%|██▌ | 131/500 [12:38<22:57, 3.73s/it, loss=0.8, lr=0.00026]\nSteps: 26%|██▌ | 131/500 [12:38<22:57, 3.73s/it, loss=0.899, lr=0.000262]\nSteps: 26%|██▋ | 132/500 [12:40<19:29, 3.18s/it, loss=0.899, lr=0.000262]\nSteps: 26%|██▋ | 132/500 [12:40<19:29, 3.18s/it, loss=0.369, lr=0.000264]\nSteps: 27%|██▋ | 133/500 [12:48<27:38, 4.52s/it, loss=0.369, lr=0.000264]\nSteps: 27%|██▋ | 133/500 [12:48<27:38, 4.52s/it, loss=0.545, lr=0.000266]\nSteps: 27%|██▋ | 134/500 [12:49<22:45, 3.73s/it, loss=0.545, lr=0.000266]\nSteps: 27%|██▋ | 134/500 [12:49<22:45, 3.73s/it, loss=0.45, lr=0.000268] \nSteps: 27%|██▋ | 135/500 [12:51<19:19, 3.18s/it, loss=0.45, lr=0.000268]\nSteps: 27%|██▋ | 135/500 [12:51<19:19, 3.18s/it, loss=0.634, lr=0.00027]\nSteps: 27%|██▋ | 136/500 [12:59<27:35, 4.55s/it, loss=0.634, lr=0.00027]\nSteps: 27%|██▋ | 136/500 [12:59<27:35, 4.55s/it, loss=0.848, lr=0.000272]\nSteps: 27%|██▋ | 137/500 [13:01<22:40, 3.75s/it, loss=0.848, lr=0.000272]\nSteps: 27%|██▋ | 137/500 [13:01<22:40, 3.75s/it, loss=0.462, lr=0.000274]\nSteps: 28%|██▊ | 138/500 [13:03<19:14, 3.19s/it, loss=0.462, lr=0.000274]\nSteps: 28%|██▊ | 138/500 [13:03<19:14, 3.19s/it, loss=0.277, lr=0.000276]\nSteps: 28%|██▊ | 139/500 [13:10<26:59, 4.49s/it, loss=0.277, lr=0.000276]\nSteps: 28%|██▊ | 139/500 [13:10<26:59, 4.49s/it, loss=0.874, lr=0.000278]\nSteps: 28%|██▊ | 140/500 [13:12<22:13, 3.71s/it, loss=0.874, lr=0.000278]\nSteps: 28%|██▊ | 140/500 [13:12<22:13, 3.71s/it, loss=0.527, lr=0.00028] \nSteps: 28%|██▊ | 141/500 [13:14<18:53, 3.16s/it, loss=0.527, lr=0.00028]\nSteps: 28%|██▊ | 141/500 [13:14<18:53, 3.16s/it, loss=1, lr=0.000282] \nSteps: 28%|██▊ | 142/500 [13:22<26:42, 4.48s/it, loss=1, lr=0.000282]\nSteps: 28%|██▊ | 142/500 [13:22<26:42, 4.48s/it, loss=0.889, lr=0.000284]\nSteps: 29%|██▊ | 143/500 [13:24<22:00, 3.70s/it, loss=0.889, lr=0.000284]\nSteps: 29%|██▊ | 143/500 [13:24<22:00, 3.70s/it, loss=0.439, lr=0.000286]\nSteps: 29%|██▉ | 144/500 [13:25<18:43, 3.15s/it, loss=0.439, lr=0.000286]\nSteps: 29%|██▉ | 144/500 [13:25<18:43, 3.15s/it, loss=0.725, lr=0.000288]\nSteps: 29%|██▉ | 145/500 [13:33<26:32, 4.49s/it, loss=0.725, lr=0.000288]\nSteps: 29%|██▉ | 145/500 [13:33<26:32, 4.49s/it, loss=0.523, lr=0.00029] \nSteps: 29%|██▉ | 146/500 [13:35<21:51, 3.71s/it, loss=0.523, lr=0.00029]\nSteps: 29%|██▉ | 146/500 [13:35<21:51, 3.71s/it, loss=0.829, lr=0.000292]\nSteps: 29%|██▉ | 147/500 [13:37<18:35, 3.16s/it, loss=0.829, lr=0.000292]\nSteps: 29%|██▉ | 147/500 [13:37<18:35, 3.16s/it, loss=0.412, lr=0.000294]\nSteps: 30%|██▉ | 148/500 [13:44<26:23, 4.50s/it, loss=0.412, lr=0.000294]\nSteps: 30%|██▉ | 148/500 [13:44<26:23, 4.50s/it, loss=0.812, lr=0.000296]\nSteps: 30%|██▉ | 149/500 [13:46<21:43, 3.71s/it, loss=0.812, lr=0.000296]\nSteps: 30%|██▉ | 149/500 [13:46<21:43, 3.71s/it, loss=0.48, lr=0.000298] \nSteps: 30%|███ | 150/500 [13:48<18:27, 3.17s/it, loss=0.48, lr=0.000298]\nSteps: 30%|███ | 150/500 [13:48<18:27, 3.17s/it, loss=0.815, lr=0.0003] \nSteps: 30%|███ | 151/500 [13:56<26:07, 4.49s/it, loss=0.815, lr=0.0003]\nSteps: 30%|███ | 151/500 [13:56<26:07, 4.49s/it, loss=0.841, lr=0.000302]\nSteps: 30%|███ | 152/500 [13:58<21:30, 3.71s/it, loss=0.841, lr=0.000302]\nSteps: 30%|███ | 152/500 [13:58<21:30, 3.71s/it, loss=0.652, lr=0.000304]\nSteps: 31%|███ | 153/500 [14:00<18:16, 3.16s/it, loss=0.652, lr=0.000304]\nSteps: 31%|███ | 153/500 [14:00<18:16, 3.16s/it, loss=0.356, lr=0.000306]\nSteps: 31%|███ | 154/500 [14:07<26:10, 4.54s/it, loss=0.356, lr=0.000306]\nSteps: 31%|███ | 154/500 [14:07<26:10, 4.54s/it, loss=0.994, lr=0.000308]\nSteps: 31%|███ | 155/500 [14:09<21:31, 3.74s/it, loss=0.994, lr=0.000308]\nSteps: 31%|███ | 155/500 [14:09<21:31, 3.74s/it, loss=0.458, lr=0.00031] \nSteps: 31%|███ | 156/500 [14:11<18:15, 3.18s/it, loss=0.458, lr=0.00031]\nSteps: 31%|███ | 156/500 [14:11<18:15, 3.18s/it, loss=1.01, lr=0.000312]\nSteps: 31%|███▏ | 157/500 [14:19<25:50, 4.52s/it, loss=1.01, lr=0.000312]\nSteps: 31%|███▏ | 157/500 [14:19<25:50, 4.52s/it, loss=0.67, lr=0.000314]\nSteps: 32%|███▏ | 158/500 [14:21<21:15, 3.73s/it, loss=0.67, lr=0.000314]\nSteps: 32%|███▏ | 158/500 [14:21<21:15, 3.73s/it, loss=0.468, lr=0.000316]\nSteps: 32%|███▏ | 159/500 [14:22<18:02, 3.17s/it, loss=0.468, lr=0.000316]\nSteps: 32%|███▏ | 159/500 [14:22<18:02, 3.17s/it, loss=0.272, lr=0.000318]\nSteps: 32%|███▏ | 160/500 [14:30<25:56, 4.58s/it, loss=0.272, lr=0.000318]\nSteps: 32%|███▏ | 160/500 [14:30<25:56, 4.58s/it, loss=0.746, lr=0.00032] \nSteps: 32%|███▏ | 161/500 [14:32<21:18, 3.77s/it, loss=0.746, lr=0.00032]\nSteps: 32%|███▏ | 161/500 [14:32<21:18, 3.77s/it, loss=0.688, lr=0.000322]\nSteps: 32%|███▏ | 162/500 [14:34<18:03, 3.20s/it, loss=0.688, lr=0.000322]\nSteps: 32%|███▏ | 162/500 [14:34<18:03, 3.20s/it, loss=0.411, lr=0.000324]\nSteps: 33%|███▎ | 163/500 [14:42<25:19, 4.51s/it, loss=0.411, lr=0.000324]\nSteps: 33%|███▎ | 163/500 [14:42<25:19, 4.51s/it, loss=0.568, lr=0.000326]\nSteps: 33%|███▎ | 164/500 [14:43<20:50, 3.72s/it, loss=0.568, lr=0.000326]\nSteps: 33%|███▎ | 164/500 [14:43<20:50, 3.72s/it, loss=0.46, lr=0.000328] \nSteps: 33%|███▎ | 165/500 [14:45<17:41, 3.17s/it, loss=0.46, lr=0.000328]\nSteps: 33%|███▎ | 165/500 [14:45<17:41, 3.17s/it, loss=0.553, lr=0.00033]\nSteps: 33%|███▎ | 166/500 [14:53<25:05, 4.51s/it, loss=0.553, lr=0.00033]\nSteps: 33%|███▎ | 166/500 [14:53<25:05, 4.51s/it, loss=0.581, lr=0.000332]\nSteps: 33%|███▎ | 167/500 [14:55<20:38, 3.72s/it, loss=0.581, lr=0.000332]\nSteps: 33%|███▎ | 167/500 [14:55<20:38, 3.72s/it, loss=0.502, lr=0.000334]\nSteps: 34%|███▎ | 168/500 [14:57<17:32, 3.17s/it, loss=0.502, lr=0.000334]\nSteps: 34%|███▎ | 168/500 [14:57<17:32, 3.17s/it, loss=0.337, lr=0.000336]\nSteps: 34%|███▍ | 169/500 [15:04<24:49, 4.50s/it, loss=0.337, lr=0.000336]\nSteps: 34%|███▍ | 169/500 [15:04<24:49, 4.50s/it, loss=0.648, lr=0.000338]\nSteps: 34%|███▍ | 170/500 [15:06<20:25, 3.72s/it, loss=0.648, lr=0.000338]\nSteps: 34%|███▍ | 170/500 [15:06<20:25, 3.72s/it, loss=0.443, lr=0.00034] \nSteps: 34%|███▍ | 171/500 [15:08<17:21, 3.17s/it, loss=0.443, lr=0.00034]\nSteps: 34%|███▍ | 171/500 [15:08<17:21, 3.17s/it, loss=0.39, lr=0.000342]\nSteps: 34%|███▍ | 172/500 [15:16<24:38, 4.51s/it, loss=0.39, lr=0.000342]\nSteps: 34%|███▍ | 172/500 [15:16<24:38, 4.51s/it, loss=0.984, lr=0.000344]\nSteps: 35%|███▍ | 173/500 [15:18<20:16, 3.72s/it, loss=0.984, lr=0.000344]\nSteps: 35%|███▍ | 173/500 [15:18<20:16, 3.72s/it, loss=0.758, lr=0.000346]\nSteps: 35%|███▍ | 174/500 [15:20<17:13, 3.17s/it, loss=0.758, lr=0.000346]\nSteps: 35%|███▍ | 174/500 [15:20<17:13, 3.17s/it, loss=0.343, lr=0.000348]\nSteps: 35%|███▌ | 175/500 [15:27<24:27, 4.52s/it, loss=0.343, lr=0.000348]\nSteps: 35%|███▌ | 175/500 [15:27<24:27, 4.52s/it, loss=0.686, lr=0.00035] \nSteps: 35%|███▌ | 176/500 [15:29<20:06, 3.73s/it, loss=0.686, lr=0.00035]\nSteps: 35%|███▌ | 176/500 [15:29<20:06, 3.73s/it, loss=1, lr=0.000352] \nSteps: 35%|███▌ | 177/500 [15:31<17:04, 3.17s/it, loss=1, lr=0.000352]\nSteps: 35%|███▌ | 177/500 [15:31<17:04, 3.17s/it, loss=0.293, lr=0.000354]\nSteps: 36%|███▌ | 178/500 [15:39<24:18, 4.53s/it, loss=0.293, lr=0.000354]\nSteps: 36%|███▌ | 178/500 [15:39<24:18, 4.53s/it, loss=0.58, lr=0.000356] \nSteps: 36%|███▌ | 179/500 [15:41<19:59, 3.74s/it, loss=0.58, lr=0.000356]\nSteps: 36%|███▌ | 179/500 [15:41<19:59, 3.74s/it, loss=0.978, lr=0.000358]\nSteps: 36%|███▌ | 180/500 [15:42<16:58, 3.18s/it, loss=0.978, lr=0.000358]\nSteps: 36%|███▌ | 180/500 [15:42<16:58, 3.18s/it, loss=0.312, lr=0.00036] \nSteps: 36%|███▌ | 181/500 [15:50<24:04, 4.53s/it, loss=0.312, lr=0.00036]\nSteps: 36%|███▌ | 181/500 [15:50<24:04, 4.53s/it, loss=0.996, lr=0.000362]\nSteps: 36%|███▋ | 182/500 [15:52<19:47, 3.73s/it, loss=0.996, lr=0.000362]\nSteps: 36%|███▋ | 182/500 [15:52<19:47, 3.73s/it, loss=0.481, lr=0.000364]\nSteps: 37%|███▋ | 183/500 [15:54<16:47, 3.18s/it, loss=0.481, lr=0.000364]\nSteps: 37%|███▋ | 183/500 [15:54<16:47, 3.18s/it, loss=0.923, lr=0.000366]\nSteps: 37%|███▋ | 184/500 [16:02<23:48, 4.52s/it, loss=0.923, lr=0.000366]\nSteps: 37%|███▋ | 184/500 [16:02<23:48, 4.52s/it, loss=0.949, lr=0.000368]\nSteps: 37%|███▋ | 185/500 [16:03<19:35, 3.73s/it, loss=0.949, lr=0.000368]\nSteps: 37%|███▋ | 185/500 [16:03<19:35, 3.73s/it, loss=0.996, lr=0.00037] \nSteps: 37%|███▋ | 186/500 [16:05<16:37, 3.18s/it, loss=0.996, lr=0.00037]\nSteps: 37%|███▋ | 186/500 [16:05<16:37, 3.18s/it, loss=1.01, lr=0.000372]\nSteps: 37%|███▋ | 187/500 [16:13<23:41, 4.54s/it, loss=1.01, lr=0.000372]\nSteps: 37%|███▋ | 187/500 [16:13<23:41, 4.54s/it, loss=0.628, lr=0.000374]\nSteps: 38%|███▊ | 188/500 [16:15<19:28, 3.74s/it, loss=0.628, lr=0.000374]\nSteps: 38%|███▊ | 188/500 [16:15<19:28, 3.74s/it, loss=0.465, lr=0.000376]\nSteps: 38%|███▊ | 189/500 [16:17<16:30, 3.19s/it, loss=0.465, lr=0.000376]\nSteps: 38%|███▊ | 189/500 [16:17<16:30, 3.19s/it, loss=0.51, lr=0.000378] \nSteps: 38%|███▊ | 190/500 [16:24<23:25, 4.53s/it, loss=0.51, lr=0.000378]\nSteps: 38%|███▊ | 190/500 [16:24<23:25, 4.53s/it, loss=0.547, lr=0.00038]\nSteps: 38%|███▊ | 191/500 [16:26<19:16, 3.74s/it, loss=0.547, lr=0.00038]\nSteps: 38%|███▊ | 191/500 [16:26<19:16, 3.74s/it, loss=0.441, lr=0.000382]\nSteps: 38%|███▊ | 192/500 [16:28<16:20, 3.18s/it, loss=0.441, lr=0.000382]\nSteps: 38%|███▊ | 192/500 [16:28<16:20, 3.18s/it, loss=0.73, lr=0.000384] \nSteps: 39%|███▊ | 193/500 [16:36<22:57, 4.49s/it, loss=0.73, lr=0.000384]\nSteps: 39%|███▊ | 193/500 [16:36<22:57, 4.49s/it, loss=0.541, lr=0.000386]\nSteps: 39%|███▉ | 194/500 [16:38<18:53, 3.70s/it, loss=0.541, lr=0.000386]\nSteps: 39%|███▉ | 194/500 [16:38<18:53, 3.70s/it, loss=0.805, lr=0.000388]\nSteps: 39%|███▉ | 195/500 [16:40<16:02, 3.16s/it, loss=0.805, lr=0.000388]\nSteps: 39%|███▉ | 195/500 [16:40<16:02, 3.16s/it, loss=0.501, lr=0.00039] \nSteps: 39%|███▉ | 196/500 [16:47<22:53, 4.52s/it, loss=0.501, lr=0.00039]\nSteps: 39%|███▉ | 196/500 [16:47<22:53, 4.52s/it, loss=0.607, lr=0.000392]\nSteps: 39%|███▉ | 197/500 [16:49<18:49, 3.73s/it, loss=0.607, lr=0.000392]\nSteps: 39%|███▉ | 197/500 [16:49<18:49, 3.73s/it, loss=1.03, lr=0.000394] \nSteps: 40%|███▉ | 198/500 [16:51<15:58, 3.17s/it, loss=1.03, lr=0.000394]\nSteps: 40%|███▉ | 198/500 [16:51<15:58, 3.17s/it, loss=0.45, lr=0.000396]\nSteps: 40%|███▉ | 199/500 [16:59<22:29, 4.48s/it, loss=0.45, lr=0.000396]\nSteps: 40%|███▉ | 199/500 [16:59<22:29, 4.48s/it, loss=0.579, lr=0.000398]\nSteps: 40%|████ | 200/500 [17:00<18:30, 3.70s/it, loss=0.579, lr=0.000398]\nSteps: 40%|████ | 200/500 [17:00<18:30, 3.70s/it, loss=0.451, lr=0.0004] \nSteps: 40%|████ | 201/500 [17:02<15:43, 3.16s/it, loss=0.451, lr=0.0004]\nSteps: 40%|████ | 201/500 [17:02<15:43, 3.16s/it, loss=0.468, lr=0.0004]\nSteps: 40%|████ | 202/500 [17:10<22:23, 4.51s/it, loss=0.468, lr=0.0004]\nSteps: 40%|████ | 202/500 [17:10<22:23, 4.51s/it, loss=1.06, lr=0.0004] \nSteps: 41%|████ | 203/500 [17:12<18:25, 3.72s/it, loss=1.06, lr=0.0004]\nSteps: 41%|████ | 203/500 [17:12<18:25, 3.72s/it, loss=0.462, lr=0.0004]\nSteps: 41%|████ | 204/500 [17:14<15:38, 3.17s/it, loss=0.462, lr=0.0004]\nSteps: 41%|████ | 204/500 [17:14<15:38, 3.17s/it, loss=0.405, lr=0.0004]\nSteps: 41%|████ | 205/500 [17:21<22:15, 4.53s/it, loss=0.405, lr=0.0004]\nSteps: 41%|████ | 205/500 [17:21<22:15, 4.53s/it, loss=0.921, lr=0.0004]\nSteps: 41%|████ | 206/500 [17:23<18:17, 3.73s/it, loss=0.921, lr=0.0004]\nSteps: 41%|████ | 206/500 [17:23<18:17, 3.73s/it, loss=0.462, lr=0.0004]\nSteps: 41%|████▏ | 207/500 [17:25<15:31, 3.18s/it, loss=0.462, lr=0.0004]\nSteps: 41%|████▏ | 207/500 [17:25<15:31, 3.18s/it, loss=0.799, lr=0.000399]\nSteps: 42%|████▏ | 208/500 [17:33<22:01, 4.53s/it, loss=0.799, lr=0.000399]\nSteps: 42%|████▏ | 208/500 [17:33<22:01, 4.53s/it, loss=1.22, lr=0.000399] \nSteps: 42%|████▏ | 209/500 [17:35<18:05, 3.73s/it, loss=1.22, lr=0.000399]\nSteps: 42%|████▏ | 209/500 [17:35<18:05, 3.73s/it, loss=0.448, lr=0.000399]\nSteps: 42%|████▏ | 210/500 [17:37<15:21, 3.18s/it, loss=0.448, lr=0.000399]\nSteps: 42%|████▏ | 210/500 [17:37<15:21, 3.18s/it, loss=0.317, lr=0.000399]\nSteps: 42%|████▏ | 211/500 [17:44<21:46, 4.52s/it, loss=0.317, lr=0.000399]\nSteps: 42%|████▏ | 211/500 [17:44<21:46, 4.52s/it, loss=0.61, lr=0.000399] \nSteps: 42%|████▏ | 212/500 [17:46<17:53, 3.73s/it, loss=0.61, lr=0.000399]\nSteps: 42%|████▏ | 212/500 [17:46<17:53, 3.73s/it, loss=0.751, lr=0.000398]\nSteps: 43%|████▎ | 213/500 [17:48<15:11, 3.18s/it, loss=0.751, lr=0.000398]\nSteps: 43%|████▎ | 213/500 [17:48<15:11, 3.18s/it, loss=0.31, lr=0.000398] \nSteps: 43%|████▎ | 214/500 [17:56<21:43, 4.56s/it, loss=0.31, lr=0.000398]\nSteps: 43%|████▎ | 214/500 [17:56<21:43, 4.56s/it, loss=0.919, lr=0.000398]\nSteps: 43%|████▎ | 215/500 [17:58<17:50, 3.76s/it, loss=0.919, lr=0.000398]\nSteps: 43%|████▎ | 215/500 [17:58<17:50, 3.76s/it, loss=0.454, lr=0.000398]\nSteps: 43%|████▎ | 216/500 [18:00<15:07, 3.19s/it, loss=0.454, lr=0.000398]\nSteps: 43%|████▎ | 216/500 [18:00<15:07, 3.19s/it, loss=0.395, lr=0.000397]\nSteps: 43%|████▎ | 217/500 [18:07<21:27, 4.55s/it, loss=0.395, lr=0.000397]\nSteps: 43%|████▎ | 217/500 [18:07<21:27, 4.55s/it, loss=0.579, lr=0.000397]\nSteps: 44%|████▎ | 218/500 [18:09<17:37, 3.75s/it, loss=0.579, lr=0.000397]\nSteps: 44%|████▎ | 218/500 [18:09<17:37, 3.75s/it, loss=0.641, lr=0.000396]\nSteps: 44%|████▍ | 219/500 [18:11<14:56, 3.19s/it, loss=0.641, lr=0.000396]\nSteps: 44%|████▍ | 219/500 [18:11<14:56, 3.19s/it, loss=0.817, lr=0.000396]\nSteps: 44%|████▍ | 220/500 [18:19<21:07, 4.53s/it, loss=0.817, lr=0.000396]\nSteps: 44%|████▍ | 220/500 [18:19<21:07, 4.53s/it, loss=0.93, lr=0.000396] \nSteps: 44%|████▍ | 221/500 [18:21<17:21, 3.73s/it, loss=0.93, lr=0.000396]\nSteps: 44%|████▍ | 221/500 [18:21<17:21, 3.73s/it, loss=0.556, lr=0.000395]\nSteps: 44%|████▍ | 222/500 [18:22<14:43, 3.18s/it, loss=0.556, lr=0.000395]\nSteps: 44%|████▍ | 222/500 [18:22<14:43, 3.18s/it, loss=0.666, lr=0.000395]\nSteps: 45%|████▍ | 223/500 [18:30<20:56, 4.53s/it, loss=0.666, lr=0.000395]\nSteps: 45%|████▍ | 223/500 [18:30<20:56, 4.53s/it, loss=0.991, lr=0.000394]\nSteps: 45%|████▍ | 224/500 [18:32<17:11, 3.74s/it, loss=0.991, lr=0.000394]\nSteps: 45%|████▍ | 224/500 [18:32<17:11, 3.74s/it, loss=0.529, lr=0.000394]\nSteps: 45%|████▌ | 225/500 [18:34<14:34, 3.18s/it, loss=0.529, lr=0.000394]\nSteps: 45%|████▌ | 225/500 [18:34<14:34, 3.18s/it, loss=0.307, lr=0.000393]\nSteps: 45%|████▌ | 226/500 [18:42<20:45, 4.55s/it, loss=0.307, lr=0.000393]\nSteps: 45%|████▌ | 226/500 [18:42<20:45, 4.55s/it, loss=0.905, lr=0.000393]\nSteps: 45%|████▌ | 227/500 [18:44<17:03, 3.75s/it, loss=0.905, lr=0.000393]\nSteps: 45%|████▌ | 227/500 [18:44<17:03, 3.75s/it, loss=1.03, lr=0.000392] \nSteps: 46%|████▌ | 228/500 [18:45<14:27, 3.19s/it, loss=1.03, lr=0.000392]\nSteps: 46%|████▌ | 228/500 [18:45<14:27, 3.19s/it, loss=0.808, lr=0.000391]\nSteps: 46%|████▌ | 229/500 [18:53<20:31, 4.55s/it, loss=0.808, lr=0.000391]\nSteps: 46%|████▌ | 229/500 [18:53<20:31, 4.55s/it, loss=0.572, lr=0.000391]\nSteps: 46%|████▌ | 230/500 [18:55<16:51, 3.75s/it, loss=0.572, lr=0.000391]\nSteps: 46%|████▌ | 230/500 [18:55<16:51, 3.75s/it, loss=0.452, lr=0.00039] \nSteps: 46%|████▌ | 231/500 [18:57<14:17, 3.19s/it, loss=0.452, lr=0.00039]\nSteps: 46%|████▌ | 231/500 [18:57<14:17, 3.19s/it, loss=0.638, lr=0.00039]\nSteps: 46%|████▋ | 232/500 [19:05<20:16, 4.54s/it, loss=0.638, lr=0.00039]\nSteps: 46%|████▋ | 232/500 [19:05<20:16, 4.54s/it, loss=0.549, lr=0.000389]\nSteps: 47%|████▋ | 233/500 [19:06<16:39, 3.74s/it, loss=0.549, lr=0.000389]\nSteps: 47%|████▋ | 233/500 [19:06<16:39, 3.74s/it, loss=0.448, lr=0.000388]\nSteps: 47%|████▋ | 234/500 [19:08<14:07, 3.19s/it, loss=0.448, lr=0.000388]\nSteps: 47%|████▋ | 234/500 [19:08<14:07, 3.19s/it, loss=0.284, lr=0.000387]\nSteps: 47%|████▋ | 235/500 [19:16<20:07, 4.56s/it, loss=0.284, lr=0.000387]\nSteps: 47%|████▋ | 235/500 [19:16<20:07, 4.56s/it, loss=0.538, lr=0.000387]\nSteps: 47%|████▋ | 236/500 [19:18<16:31, 3.75s/it, loss=0.538, lr=0.000387]\nSteps: 47%|████▋ | 236/500 [19:18<16:31, 3.75s/it, loss=0.485, lr=0.000386]\nSteps: 47%|████▋ | 237/500 [19:20<13:59, 3.19s/it, loss=0.485, lr=0.000386]\nSteps: 47%|████▋ | 237/500 [19:20<13:59, 3.19s/it, loss=0.848, lr=0.000385]\nSteps: 48%|████▊ | 238/500 [19:28<19:54, 4.56s/it, loss=0.848, lr=0.000385]\nSteps: 48%|████▊ | 238/500 [19:28<19:54, 4.56s/it, loss=1.03, lr=0.000384] \nSteps: 48%|████▊ | 239/500 [19:30<16:20, 3.76s/it, loss=1.03, lr=0.000384]\nSteps: 48%|████▊ | 239/500 [19:30<16:20, 3.76s/it, loss=0.848, lr=0.000384]\nSteps: 48%|████▊ | 240/500 [19:31<13:50, 3.19s/it, loss=0.848, lr=0.000384]\nSteps: 48%|████▊ | 240/500 [19:31<13:50, 3.19s/it, loss=0.973, lr=0.000383]\nSteps: 48%|████▊ | 241/500 [19:39<19:31, 4.52s/it, loss=0.973, lr=0.000383]\nSteps: 48%|████▊ | 241/500 [19:39<19:31, 4.52s/it, loss=0.888, lr=0.000382]\nSteps: 48%|████▊ | 242/500 [19:41<16:02, 3.73s/it, loss=0.888, lr=0.000382]\nSteps: 48%|████▊ | 242/500 [19:41<16:02, 3.73s/it, loss=0.717, lr=0.000381]\nSteps: 49%|████▊ | 243/500 [19:43<13:36, 3.18s/it, loss=0.717, lr=0.000381]\nSteps: 49%|████▊ | 243/500 [19:43<13:36, 3.18s/it, loss=0.284, lr=0.00038] \nSteps: 49%|████▉ | 244/500 [19:50<19:09, 4.49s/it, loss=0.284, lr=0.00038]\nSteps: 49%|████▉ | 244/500 [19:50<19:09, 4.49s/it, loss=0.639, lr=0.000379]\nSteps: 49%|████▉ | 245/500 [19:52<15:44, 3.71s/it, loss=0.639, lr=0.000379]\nSteps: 49%|████▉ | 245/500 [19:52<15:44, 3.71s/it, loss=0.885, lr=0.000378]\nSteps: 49%|████▉ | 246/500 [19:54<13:22, 3.16s/it, loss=0.885, lr=0.000378]\nSteps: 49%|████▉ | 246/500 [19:54<13:22, 3.16s/it, loss=0.32, lr=0.000377] \nSteps: 49%|████▉ | 247/500 [20:02<18:58, 4.50s/it, loss=0.32, lr=0.000377]\nSteps: 49%|████▉ | 247/500 [20:02<18:58, 4.50s/it, loss=0.567, lr=0.000376]\nSteps: 50%|████▉ | 248/500 [20:04<15:36, 3.72s/it, loss=0.567, lr=0.000376]\nSteps: 50%|████▉ | 248/500 [20:04<15:36, 3.72s/it, loss=0.506, lr=0.000375]\nSteps: 50%|████▉ | 249/500 [20:05<13:14, 3.17s/it, loss=0.506, lr=0.000375]\nSteps: 50%|████▉ | 249/500 [20:06<13:14, 3.17s/it, loss=0.478, lr=0.000374]\nSteps: 50%|█████ | 250/500 [20:13<18:52, 4.53s/it, loss=0.478, lr=0.000374]\nSteps: 50%|█████ | 250/500 [20:13<18:52, 4.53s/it, loss=0.97, lr=0.000373] \nSteps: 50%|█████ | 251/500 [20:15<15:30, 3.74s/it, loss=0.97, lr=0.000373]\nSteps: 50%|█████ | 251/500 [20:15<15:30, 3.74s/it, loss=0.446, lr=0.000372]\nSteps: 50%|█████ | 252/500 [20:17<13:08, 3.18s/it, loss=0.446, lr=0.000372]\nSteps: 50%|█████ | 252/500 [20:17<13:08, 3.18s/it, loss=0.295, lr=0.000371]\nSteps: 51%|█████ | 253/500 [20:25<18:39, 4.53s/it, loss=0.295, lr=0.000371]\nSteps: 51%|█████ | 253/500 [20:25<18:39, 4.53s/it, loss=0.732, lr=0.00037] \nSteps: 51%|█████ | 254/500 [20:27<15:19, 3.74s/it, loss=0.732, lr=0.00037]\nSteps: 51%|█████ | 254/500 [20:27<15:19, 3.74s/it, loss=0.527, lr=0.000369]\nSteps: 51%|█████ | 255/500 [20:28<12:59, 3.18s/it, loss=0.527, lr=0.000369]\nSteps: 51%|█████ | 255/500 [20:28<12:59, 3.18s/it, loss=0.658, lr=0.000368]\nSteps: 51%|█████ | 256/500 [20:36<18:23, 4.52s/it, loss=0.658, lr=0.000368]\nSteps: 51%|█████ | 256/500 [20:36<18:23, 4.52s/it, loss=0.673, lr=0.000367]\nSteps: 51%|█████▏ | 257/500 [20:38<15:05, 3.73s/it, loss=0.673, lr=0.000367]\nSteps: 51%|█████▏ | 257/500 [20:38<15:05, 3.73s/it, loss=0.984, lr=0.000365]\nSteps: 52%|█████▏ | 258/500 [20:40<12:48, 3.18s/it, loss=0.984, lr=0.000365]\nSteps: 52%|█████▏ | 258/500 [20:40<12:48, 3.18s/it, loss=0.774, lr=0.000364]\nSteps: 52%|█████▏ | 259/500 [20:47<18:07, 4.51s/it, loss=0.774, lr=0.000364]\nSteps: 52%|█████▏ | 259/500 [20:47<18:07, 4.51s/it, loss=0.542, lr=0.000363]\nSteps: 52%|█████▏ | 260/500 [20:49<14:53, 3.72s/it, loss=0.542, lr=0.000363]\nSteps: 52%|█████▏ | 260/500 [20:49<14:53, 3.72s/it, loss=1.01, lr=0.000362] \nSteps: 52%|█████▏ | 261/500 [20:51<12:37, 3.17s/it, loss=1.01, lr=0.000362]\nSteps: 52%|█████▏ | 261/500 [20:51<12:37, 3.17s/it, loss=0.554, lr=0.000361]\nSteps: 52%|█████▏ | 262/500 [20:59<17:55, 4.52s/it, loss=0.554, lr=0.000361]\nSteps: 52%|█████▏ | 262/500 [20:59<17:55, 4.52s/it, loss=0.589, lr=0.000359]\nSteps: 53%|█████▎ | 263/500 [21:01<14:43, 3.73s/it, loss=0.589, lr=0.000359]\nSteps: 53%|█████▎ | 263/500 [21:01<14:43, 3.73s/it, loss=0.699, lr=0.000358]\nSteps: 53%|█████▎ | 264/500 [21:03<12:29, 3.18s/it, loss=0.699, lr=0.000358]\nSteps: 53%|█████▎ | 264/500 [21:03<12:29, 3.18s/it, loss=0.283, lr=0.000357]\nSteps: 53%|█████▎ | 265/500 [21:10<17:44, 4.53s/it, loss=0.283, lr=0.000357]\nSteps: 53%|█████▎ | 265/500 [21:10<17:44, 4.53s/it, loss=0.661, lr=0.000355]\nSteps: 53%|█████▎ | 266/500 [21:12<14:34, 3.74s/it, loss=0.661, lr=0.000355]\nSteps: 53%|█████▎ | 266/500 [21:12<14:34, 3.74s/it, loss=0.492, lr=0.000354]\nSteps: 53%|█████▎ | 267/500 [21:14<12:20, 3.18s/it, loss=0.492, lr=0.000354]\nSteps: 53%|█████▎ | 267/500 [21:14<12:20, 3.18s/it, loss=0.592, lr=0.000353]\nSteps: 54%|█████▎ | 268/500 [21:22<17:23, 4.50s/it, loss=0.592, lr=0.000353]\nSteps: 54%|█████▎ | 268/500 [21:22<17:23, 4.50s/it, loss=0.575, lr=0.000351]\nSteps: 54%|█████▍ | 269/500 [21:24<14:17, 3.71s/it, loss=0.575, lr=0.000351]\nSteps: 54%|█████▍ | 269/500 [21:24<14:17, 3.71s/it, loss=0.493, lr=0.00035] \nSteps: 54%|█████▍ | 270/500 [21:25<12:07, 3.16s/it, loss=0.493, lr=0.00035]\nSteps: 54%|█████▍ | 270/500 [21:25<12:07, 3.16s/it, loss=0.783, lr=0.000349]\nSteps: 54%|█████▍ | 271/500 [21:33<17:13, 4.51s/it, loss=0.783, lr=0.000349]\nSteps: 54%|█████▍ | 271/500 [21:33<17:13, 4.51s/it, loss=0.56, lr=0.000347] \nSteps: 54%|█████▍ | 272/500 [21:35<14:08, 3.72s/it, loss=0.56, lr=0.000347]\nSteps: 54%|█████▍ | 272/500 [21:35<14:08, 3.72s/it, loss=0.823, lr=0.000346]\nSteps: 55%|█████▍ | 273/500 [21:37<11:59, 3.17s/it, loss=0.823, lr=0.000346]\nSteps: 55%|█████▍ | 273/500 [21:37<11:59, 3.17s/it, loss=0.536, lr=0.000344]\nSteps: 55%|█████▍ | 274/500 [21:44<16:59, 4.51s/it, loss=0.536, lr=0.000344]\nSteps: 55%|█████▍ | 274/500 [21:45<16:59, 4.51s/it, loss=0.589, lr=0.000343]\nSteps: 55%|█████▌ | 275/500 [21:46<13:57, 3.72s/it, loss=0.589, lr=0.000343]\nSteps: 55%|█████▌ | 275/500 [21:46<13:57, 3.72s/it, loss=0.437, lr=0.000341]\nSteps: 55%|█████▌ | 276/500 [21:48<11:50, 3.17s/it, loss=0.437, lr=0.000341]\nSteps: 55%|█████▌ | 276/500 [21:48<11:50, 3.17s/it, loss=0.693, lr=0.00034] \nSteps: 55%|█████▌ | 277/500 [21:56<16:53, 4.55s/it, loss=0.693, lr=0.00034]\nSteps: 55%|█████▌ | 277/500 [21:56<16:53, 4.55s/it, loss=0.601, lr=0.000338]\nSteps: 56%|█████▌ | 278/500 [21:58<13:51, 3.75s/it, loss=0.601, lr=0.000338]\nSteps: 56%|█████▌ | 278/500 [21:58<13:51, 3.75s/it, loss=0.609, lr=0.000337]\nSteps: 56%|█████▌ | 279/500 [22:00<11:44, 3.19s/it, loss=0.609, lr=0.000337]\nSteps: 56%|█████▌ | 279/500 [22:00<11:44, 3.19s/it, loss=0.619, lr=0.000335]\nSteps: 56%|█████▌ | 280/500 [22:07<16:36, 4.53s/it, loss=0.619, lr=0.000335]\nSteps: 56%|█████▌ | 280/500 [22:07<16:36, 4.53s/it, loss=0.653, lr=0.000334]\nSteps: 56%|█████▌ | 281/500 [22:09<13:37, 3.73s/it, loss=0.653, lr=0.000334]\nSteps: 56%|█████▌ | 281/500 [22:09<13:37, 3.73s/it, loss=0.946, lr=0.000332]\nSteps: 56%|█████▋ | 282/500 [22:11<11:32, 3.18s/it, loss=0.946, lr=0.000332]\nSteps: 56%|█████▋ | 282/500 [22:11<11:32, 3.18s/it, loss=0.272, lr=0.000331]\nSteps: 57%|█████▋ | 283/500 [22:19<16:26, 4.55s/it, loss=0.272, lr=0.000331]\nSteps: 57%|█████▋ | 283/500 [22:19<16:26, 4.55s/it, loss=0.567, lr=0.000329]\nSteps: 57%|█████▋ | 284/500 [22:21<13:29, 3.75s/it, loss=0.567, lr=0.000329]\nSteps: 57%|█████▋ | 284/500 [22:21<13:29, 3.75s/it, loss=0.45, lr=0.000327] \nSteps: 57%|█████▋ | 285/500 [22:23<11:25, 3.19s/it, loss=0.45, lr=0.000327]\nSteps: 57%|█████▋ | 285/500 [22:23<11:25, 3.19s/it, loss=0.299, lr=0.000326]\nSteps: 57%|█████▋ | 286/500 [22:30<16:08, 4.53s/it, loss=0.299, lr=0.000326]\nSteps: 57%|█████▋ | 286/500 [22:30<16:08, 4.53s/it, loss=0.687, lr=0.000324]\nSteps: 57%|█████▋ | 287/500 [22:32<13:15, 3.73s/it, loss=0.687, lr=0.000324]\nSteps: 57%|█████▋ | 287/500 [22:32<13:15, 3.73s/it, loss=0.457, lr=0.000323]\nSteps: 58%|█████▊ | 288/500 [22:34<11:13, 3.18s/it, loss=0.457, lr=0.000323]\nSteps: 58%|█████▊ | 288/500 [22:34<11:13, 3.18s/it, loss=0.333, lr=0.000321]\nSteps: 58%|█████▊ | 289/500 [22:42<15:55, 4.53s/it, loss=0.333, lr=0.000321]\nSteps: 58%|█████▊ | 289/500 [22:42<15:55, 4.53s/it, loss=0.936, lr=0.000319]\nSteps: 58%|█████▊ | 290/500 [22:44<13:04, 3.74s/it, loss=0.936, lr=0.000319]\nSteps: 58%|█████▊ | 290/500 [22:44<13:04, 3.74s/it, loss=0.6, lr=0.000318] \nSteps: 58%|█████▊ | 291/500 [22:46<11:04, 3.18s/it, loss=0.6, lr=0.000318]\nSteps: 58%|█████▊ | 291/500 [22:46<11:04, 3.18s/it, loss=0.434, lr=0.000316]\nSteps: 58%|█████▊ | 292/500 [22:53<15:44, 4.54s/it, loss=0.434, lr=0.000316]\nSteps: 58%|█████▊ | 292/500 [22:53<15:44, 4.54s/it, loss=0.562, lr=0.000314]\nSteps: 59%|█████▊ | 293/500 [22:55<12:54, 3.74s/it, loss=0.562, lr=0.000314]\nSteps: 59%|█████▊ | 293/500 [22:55<12:54, 3.74s/it, loss=0.462, lr=0.000312]\nSteps: 59%|█████▉ | 294/500 [22:57<10:56, 3.19s/it, loss=0.462, lr=0.000312]\nSteps: 59%|█████▉ | 294/500 [22:57<10:56, 3.19s/it, loss=0.667, lr=0.000311]\nSteps: 59%|█████▉ | 295/500 [23:05<15:33, 4.55s/it, loss=0.667, lr=0.000311]\nSteps: 59%|█████▉ | 295/500 [23:05<15:33, 4.55s/it, loss=0.514, lr=0.000309]\nSteps: 59%|█████▉ | 296/500 [23:07<12:45, 3.75s/it, loss=0.514, lr=0.000309]\nSteps: 59%|█████▉ | 296/500 [23:07<12:45, 3.75s/it, loss=1.03, lr=0.000307] \nSteps: 59%|█████▉ | 297/500 [23:09<10:47, 3.19s/it, loss=1.03, lr=0.000307]\nSteps: 59%|█████▉ | 297/500 [23:09<10:47, 3.19s/it, loss=0.667, lr=0.000305]\nSteps: 60%|█████▉ | 298/500 [23:16<15:13, 4.52s/it, loss=0.667, lr=0.000305]\nSteps: 60%|█████▉ | 298/500 [23:16<15:13, 4.52s/it, loss=0.792, lr=0.000304]\nSteps: 60%|█████▉ | 299/500 [23:18<12:29, 3.73s/it, loss=0.792, lr=0.000304]\nSteps: 60%|█████▉ | 299/500 [23:18<12:29, 3.73s/it, loss=0.582, lr=0.000302]\nSteps: 60%|██████ | 300/500 [23:20<10:35, 3.18s/it, loss=0.582, lr=0.000302]\nSteps: 60%|██████ | 300/500 [23:20<10:35, 3.18s/it, loss=0.754, lr=0.0003] \nSteps: 60%|██████ | 301/500 [23:28<15:09, 4.57s/it, loss=0.754, lr=0.0003]\nSteps: 60%|██████ | 301/500 [23:28<15:09, 4.57s/it, loss=0.508, lr=0.000298]\nSteps: 60%|██████ | 302/500 [23:30<12:25, 3.76s/it, loss=0.508, lr=0.000298]\nSteps: 60%|██████ | 302/500 [23:30<12:25, 3.76s/it, loss=0.472, lr=0.000296]\nSteps: 61%|██████ | 303/500 [23:32<10:30, 3.20s/it, loss=0.472, lr=0.000296]\nSteps: 61%|██████ | 303/500 [23:32<10:30, 3.20s/it, loss=0.28, lr=0.000295] \nSteps: 61%|██████ | 304/500 [23:39<14:54, 4.57s/it, loss=0.28, lr=0.000295]\nSteps: 61%|██████ | 304/500 [23:39<14:54, 4.57s/it, loss=0.529, lr=0.000293]\nSteps: 61%|██████ | 305/500 [23:41<12:13, 3.76s/it, loss=0.529, lr=0.000293]\nSteps: 61%|██████ | 305/500 [23:41<12:13, 3.76s/it, loss=0.757, lr=0.000291]\nSteps: 61%|██████ | 306/500 [23:43<10:20, 3.20s/it, loss=0.757, lr=0.000291]\nSteps: 61%|██████ | 306/500 [23:43<10:20, 3.20s/it, loss=0.314, lr=0.000289]\nSteps: 61%|██████▏ | 307/500 [23:51<14:47, 4.60s/it, loss=0.314, lr=0.000289]\nSteps: 61%|██████▏ | 307/500 [23:51<14:47, 4.60s/it, loss=0.878, lr=0.000287]\nSteps: 62%|██████▏ | 308/500 [23:53<12:06, 3.78s/it, loss=0.878, lr=0.000287]\nSteps: 62%|██████▏ | 308/500 [23:53<12:06, 3.78s/it, loss=0.701, lr=0.000285]\nSteps: 62%|██████▏ | 309/500 [23:55<10:13, 3.21s/it, loss=0.701, lr=0.000285]\nSteps: 62%|██████▏ | 309/500 [23:55<10:13, 3.21s/it, loss=0.363, lr=0.000283]\nSteps: 62%|██████▏ | 310/500 [24:02<14:25, 4.56s/it, loss=0.363, lr=0.000283]\nSteps: 62%|██████▏ | 310/500 [24:02<14:25, 4.56s/it, loss=1.04, lr=0.000281] \nSteps: 62%|██████▏ | 311/500 [24:04<11:49, 3.75s/it, loss=1.04, lr=0.000281]\nSteps: 62%|██████▏ | 311/500 [24:04<11:49, 3.75s/it, loss=0.431, lr=0.000279]\nSteps: 62%|██████▏ | 312/500 [24:06<10:00, 3.19s/it, loss=0.431, lr=0.000279]\nSteps: 62%|██████▏ | 312/500 [24:06<10:00, 3.19s/it, loss=0.693, lr=0.000278]\nSteps: 63%|██████▎ | 313/500 [24:14<14:06, 4.52s/it, loss=0.693, lr=0.000278]\nSteps: 63%|██████▎ | 313/500 [24:14<14:06, 4.52s/it, loss=0.872, lr=0.000276]\nSteps: 63%|██████▎ | 314/500 [24:16<11:34, 3.73s/it, loss=0.872, lr=0.000276]\nSteps: 63%|██████▎ | 314/500 [24:16<11:34, 3.73s/it, loss=0.551, lr=0.000274]\nSteps: 63%|██████▎ | 315/500 [24:18<09:47, 3.18s/it, loss=0.551, lr=0.000274]\nSteps: 63%|██████▎ | 315/500 [24:18<09:47, 3.18s/it, loss=0.497, lr=0.000272]\nSteps: 63%|██████▎ | 316/500 [24:25<13:52, 4.52s/it, loss=0.497, lr=0.000272]\nSteps: 63%|██████▎ | 316/500 [24:25<13:52, 4.52s/it, loss=0.579, lr=0.00027] \nSteps: 63%|██████▎ | 317/500 [24:27<11:22, 3.73s/it, loss=0.579, lr=0.00027]\nSteps: 63%|██████▎ | 317/500 [24:27<11:22, 3.73s/it, loss=0.668, lr=0.000268]\nSteps: 64%|██████▎ | 318/500 [24:29<09:38, 3.18s/it, loss=0.668, lr=0.000268]\nSteps: 64%|██████▎ | 318/500 [24:29<09:38, 3.18s/it, loss=0.846, lr=0.000266]\nSteps: 64%|██████▍ | 319/500 [24:37<13:41, 4.54s/it, loss=0.846, lr=0.000266]\nSteps: 64%|██████▍ | 319/500 [24:37<13:41, 4.54s/it, loss=1.03, lr=0.000264] \nSteps: 64%|██████▍ | 320/500 [24:39<11:13, 3.74s/it, loss=1.03, lr=0.000264]\nSteps: 64%|██████▍ | 320/500 [24:39<11:13, 3.74s/it, loss=1.01, lr=0.000262]\nSteps: 64%|██████▍ | 321/500 [24:40<09:30, 3.19s/it, loss=1.01, lr=0.000262]\nSteps: 64%|██████▍ | 321/500 [24:40<09:30, 3.19s/it, loss=0.334, lr=0.00026]\nSteps: 64%|██████▍ | 322/500 [24:48<13:27, 4.53s/it, loss=0.334, lr=0.00026]\nSteps: 64%|██████▍ | 322/500 [24:48<13:27, 4.53s/it, loss=0.601, lr=0.000258]\nSteps: 65%|██████▍ | 323/500 [24:50<11:01, 3.74s/it, loss=0.601, lr=0.000258]\nSteps: 65%|██████▍ | 323/500 [24:50<11:01, 3.74s/it, loss=1.01, lr=0.000256] \nSteps: 65%|██████▍ | 324/500 [24:52<09:20, 3.18s/it, loss=1.01, lr=0.000256]\nSteps: 65%|██████▍ | 324/500 [24:52<09:20, 3.18s/it, loss=0.628, lr=0.000254]\nSteps: 65%|██████▌ | 325/500 [25:00<13:11, 4.52s/it, loss=0.628, lr=0.000254]\nSteps: 65%|██████▌ | 325/500 [25:00<13:11, 4.52s/it, loss=0.52, lr=0.000252] \nSteps: 65%|██████▌ | 326/500 [25:01<10:49, 3.73s/it, loss=0.52, lr=0.000252]\nSteps: 65%|██████▌ | 326/500 [25:01<10:49, 3.73s/it, loss=0.422, lr=0.00025]\nSteps: 65%|██████▌ | 327/500 [25:03<09:09, 3.18s/it, loss=0.422, lr=0.00025]\nSteps: 65%|██████▌ | 327/500 [25:03<09:09, 3.18s/it, loss=0.941, lr=0.000248]\nSteps: 66%|██████▌ | 328/500 [25:11<12:56, 4.51s/it, loss=0.941, lr=0.000248]\nSteps: 66%|██████▌ | 328/500 [25:11<12:56, 4.51s/it, loss=1.02, lr=0.000246] \nSteps: 66%|██████▌ | 329/500 [25:13<10:36, 3.73s/it, loss=1.02, lr=0.000246]\nSteps: 66%|██████▌ | 329/500 [25:13<10:36, 3.73s/it, loss=0.62, lr=0.000244]\nSteps: 66%|██████▌ | 330/500 [25:15<08:59, 3.17s/it, loss=0.62, lr=0.000244]\nSteps: 66%|██████▌ | 330/500 [25:15<08:59, 3.17s/it, loss=0.628, lr=0.000242]\nSteps: 66%|██████▌ | 331/500 [25:22<12:39, 4.50s/it, loss=0.628, lr=0.000242]\nSteps: 66%|██████▌ | 331/500 [25:22<12:39, 4.50s/it, loss=1.03, lr=0.00024] \nSteps: 66%|██████▋ | 332/500 [25:24<10:23, 3.71s/it, loss=1.03, lr=0.00024]\nSteps: 66%|██████▋ | 332/500 [25:24<10:23, 3.71s/it, loss=0.624, lr=0.000237]\nSteps: 67%|██████▋ | 333/500 [25:26<08:48, 3.16s/it, loss=0.624, lr=0.000237]\nSteps: 67%|██████▋ | 333/500 [25:26<08:48, 3.16s/it, loss=0.602, lr=0.000235]\nSteps: 67%|██████▋ | 334/500 [25:34<12:25, 4.49s/it, loss=0.602, lr=0.000235]\nSteps: 67%|██████▋ | 334/500 [25:34<12:25, 4.49s/it, loss=0.792, lr=0.000233]\nSteps: 67%|██████▋ | 335/500 [25:36<10:12, 3.71s/it, loss=0.792, lr=0.000233]\nSteps: 67%|██████▋ | 335/500 [25:36<10:12, 3.71s/it, loss=0.55, lr=0.000231] \nSteps: 67%|██████▋ | 336/500 [25:37<08:38, 3.16s/it, loss=0.55, lr=0.000231]\nSteps: 67%|██████▋ | 336/500 [25:37<08:38, 3.16s/it, loss=0.265, lr=0.000229]\nSteps: 67%|██████▋ | 337/500 [25:45<12:14, 4.51s/it, loss=0.265, lr=0.000229]\nSteps: 67%|██████▋ | 337/500 [25:45<12:14, 4.51s/it, loss=0.538, lr=0.000227]\nSteps: 68%|██████▊ | 338/500 [25:47<10:02, 3.72s/it, loss=0.538, lr=0.000227]\nSteps: 68%|██████▊ | 338/500 [25:47<10:02, 3.72s/it, loss=0.478, lr=0.000225]\nSteps: 68%|██████▊ | 339/500 [25:49<08:30, 3.17s/it, loss=0.478, lr=0.000225]\nSteps: 68%|██████▊ | 339/500 [25:49<08:30, 3.17s/it, loss=0.347, lr=0.000223]\nSteps: 68%|██████▊ | 340/500 [25:57<12:03, 4.52s/it, loss=0.347, lr=0.000223]\nSteps: 68%|██████▊ | 340/500 [25:57<12:03, 4.52s/it, loss=0.502, lr=0.000221]\nSteps: 68%|██████▊ | 341/500 [25:58<09:52, 3.73s/it, loss=0.502, lr=0.000221]\nSteps: 68%|██████▊ | 341/500 [25:58<09:52, 3.73s/it, loss=0.999, lr=0.000219]\nSteps: 68%|██████▊ | 342/500 [26:00<08:21, 3.17s/it, loss=0.999, lr=0.000219]\nSteps: 68%|██████▊ | 342/500 [26:00<08:21, 3.17s/it, loss=0.995, lr=0.000217]\nSteps: 69%|██████▊ | 343/500 [26:08<11:49, 4.52s/it, loss=0.995, lr=0.000217]\nSteps: 69%|██████▊ | 343/500 [26:08<11:49, 4.52s/it, loss=0.577, lr=0.000215]\nSteps: 69%|██████▉ | 344/500 [26:10<09:41, 3.73s/it, loss=0.577, lr=0.000215]\nSteps: 69%|██████▉ | 344/500 [26:10<09:41, 3.73s/it, loss=0.421, lr=0.000213]\nSteps: 69%|██████▉ | 345/500 [26:12<08:12, 3.18s/it, loss=0.421, lr=0.000213]\nSteps: 69%|██████▉ | 345/500 [26:12<08:12, 3.18s/it, loss=0.325, lr=0.00021] \nSteps: 69%|██████▉ | 346/500 [26:19<11:36, 4.52s/it, loss=0.325, lr=0.00021]\nSteps: 69%|██████▉ | 346/500 [26:19<11:36, 4.52s/it, loss=0.504, lr=0.000208]\nSteps: 69%|██████▉ | 347/500 [26:21<09:30, 3.73s/it, loss=0.504, lr=0.000208]\nSteps: 69%|██████▉ | 347/500 [26:21<09:30, 3.73s/it, loss=0.544, lr=0.000206]\nSteps: 70%|██████▉ | 348/500 [26:23<08:02, 3.18s/it, loss=0.544, lr=0.000206]\nSteps: 70%|██████▉ | 348/500 [26:23<08:02, 3.18s/it, loss=0.278, lr=0.000204]\nSteps: 70%|██████▉ | 349/500 [26:31<11:33, 4.59s/it, loss=0.278, lr=0.000204]\nSteps: 70%|██████▉ | 349/500 [26:31<11:33, 4.59s/it, loss=0.871, lr=0.000202]\nSteps: 70%|███████ | 350/500 [26:33<09:26, 3.78s/it, loss=0.871, lr=0.000202]\nSteps: 70%|███████ | 350/500 [26:33<09:26, 3.78s/it, loss=0.423, lr=0.0002] \nSteps: 70%|███████ | 351/500 [26:35<07:58, 3.21s/it, loss=0.423, lr=0.0002]\nSteps: 70%|███████ | 351/500 [26:35<07:58, 3.21s/it, loss=0.278, lr=0.000198]\nSteps: 70%|███████ | 352/500 [26:43<11:15, 4.57s/it, loss=0.278, lr=0.000198]\nSteps: 70%|███████ | 352/500 [26:43<11:15, 4.57s/it, loss=0.651, lr=0.000196]\nSteps: 71%|███████ | 353/500 [26:44<09:13, 3.76s/it, loss=0.651, lr=0.000196]\nSteps: 71%|███████ | 353/500 [26:44<09:13, 3.76s/it, loss=1.03, lr=0.000194] \nSteps: 71%|███████ | 354/500 [26:46<07:47, 3.20s/it, loss=1.03, lr=0.000194]\nSteps: 71%|███████ | 354/500 [26:46<07:47, 3.20s/it, loss=0.269, lr=0.000192]\nSteps: 71%|███████ | 355/500 [26:54<10:56, 4.53s/it, loss=0.269, lr=0.000192]\nSteps: 71%|███████ | 355/500 [26:54<10:56, 4.53s/it, loss=0.53, lr=0.00019] \nSteps: 71%|███████ | 356/500 [26:56<08:57, 3.73s/it, loss=0.53, lr=0.00019]\nSteps: 71%|███████ | 356/500 [26:56<08:57, 3.73s/it, loss=0.422, lr=0.000187]\nSteps: 71%|███████▏ | 357/500 [26:58<07:34, 3.18s/it, loss=0.422, lr=0.000187]\nSteps: 71%|███████▏ | 357/500 [26:58<07:34, 3.18s/it, loss=0.36, lr=0.000185] \nSteps: 72%|███████▏ | 358/500 [27:05<10:45, 4.54s/it, loss=0.36, lr=0.000185]\nSteps: 72%|███████▏ | 358/500 [27:05<10:45, 4.54s/it, loss=0.574, lr=0.000183]\nSteps: 72%|███████▏ | 359/500 [27:07<08:47, 3.74s/it, loss=0.574, lr=0.000183]\nSteps: 72%|███████▏ | 359/500 [27:07<08:47, 3.74s/it, loss=0.838, lr=0.000181]\nSteps: 72%|███████▏ | 360/500 [27:09<07:25, 3.19s/it, loss=0.838, lr=0.000181]\nSteps: 72%|███████▏ | 360/500 [27:09<07:25, 3.19s/it, loss=0.261, lr=0.000179]\nSteps: 72%|███████▏ | 361/500 [27:17<10:29, 4.53s/it, loss=0.261, lr=0.000179]\nSteps: 72%|███████▏ | 361/500 [27:17<10:29, 4.53s/it, loss=1.13, lr=0.000177] \nSteps: 72%|███████▏ | 362/500 [27:19<08:35, 3.73s/it, loss=1.13, lr=0.000177]\nSteps: 72%|███████▏ | 362/500 [27:19<08:35, 3.73s/it, loss=0.874, lr=0.000175]\nSteps: 73%|███████▎ | 363/500 [27:21<07:15, 3.18s/it, loss=0.874, lr=0.000175]\nSteps: 73%|███████▎ | 363/500 [27:21<07:15, 3.18s/it, loss=0.338, lr=0.000173]\nSteps: 73%|███████▎ | 364/500 [27:28<10:15, 4.53s/it, loss=0.338, lr=0.000173]\nSteps: 73%|███████▎ | 364/500 [27:28<10:15, 4.53s/it, loss=1.06, lr=0.000171] \nSteps: 73%|███████▎ | 365/500 [27:30<08:24, 3.73s/it, loss=1.06, lr=0.000171]\nSteps: 73%|███████▎ | 365/500 [27:30<08:24, 3.73s/it, loss=0.452, lr=0.000169]\nSteps: 73%|███████▎ | 366/500 [27:32<07:05, 3.18s/it, loss=0.452, lr=0.000169]\nSteps: 73%|███████▎ | 366/500 [27:32<07:05, 3.18s/it, loss=0.264, lr=0.000167]\nSteps: 73%|███████▎ | 367/500 [27:40<10:05, 4.56s/it, loss=0.264, lr=0.000167]\nSteps: 73%|███████▎ | 367/500 [27:40<10:05, 4.56s/it, loss=0.643, lr=0.000165]\nSteps: 74%|███████▎ | 368/500 [27:42<08:15, 3.75s/it, loss=0.643, lr=0.000165]\nSteps: 74%|███████▎ | 368/500 [27:42<08:15, 3.75s/it, loss=0.488, lr=0.000163]\nSteps: 74%|███████▍ | 369/500 [27:44<06:57, 3.19s/it, loss=0.488, lr=0.000163]\nSteps: 74%|███████▍ | 369/500 [27:44<06:57, 3.19s/it, loss=0.272, lr=0.00016] \nSteps: 74%|███████▍ | 370/500 [27:51<09:56, 4.59s/it, loss=0.272, lr=0.00016]\nSteps: 74%|███████▍ | 370/500 [27:51<09:56, 4.59s/it, loss=0.5, lr=0.000158] \nSteps: 74%|███████▍ | 371/500 [27:53<08:07, 3.78s/it, loss=0.5, lr=0.000158]\nSteps: 74%|███████▍ | 371/500 [27:53<08:07, 3.78s/it, loss=0.458, lr=0.000156]\nSteps: 74%|███████▍ | 372/500 [27:55<06:50, 3.21s/it, loss=0.458, lr=0.000156]\nSteps: 74%|███████▍ | 372/500 [27:55<06:50, 3.21s/it, loss=0.399, lr=0.000154]\nSteps: 75%|███████▍ | 373/500 [28:03<09:35, 4.53s/it, loss=0.399, lr=0.000154]\nSteps: 75%|███████▍ | 373/500 [28:03<09:35, 4.53s/it, loss=0.773, lr=0.000152]\nSteps: 75%|███████▍ | 374/500 [28:05<07:50, 3.74s/it, loss=0.773, lr=0.000152]\nSteps: 75%|███████▍ | 374/500 [28:05<07:50, 3.74s/it, loss=0.631, lr=0.00015] \nSteps: 75%|███████▌ | 375/500 [28:07<06:37, 3.18s/it, loss=0.631, lr=0.00015]\nSteps: 75%|███████▌ | 375/500 [28:07<06:37, 3.18s/it, loss=0.662, lr=0.000148]\nSteps: 75%|███████▌ | 376/500 [28:14<09:21, 4.53s/it, loss=0.662, lr=0.000148]\nSteps: 75%|███████▌ | 376/500 [28:14<09:21, 4.53s/it, loss=0.506, lr=0.000146]\nSteps: 75%|███████▌ | 377/500 [28:16<07:39, 3.73s/it, loss=0.506, lr=0.000146]\nSteps: 75%|███████▌ | 377/500 [28:16<07:39, 3.73s/it, loss=0.417, lr=0.000144]\nSteps: 76%|███████▌ | 378/500 [28:18<06:27, 3.18s/it, loss=0.417, lr=0.000144]\nSteps: 76%|███████▌ | 378/500 [28:18<06:27, 3.18s/it, loss=0.884, lr=0.000142]\nSteps: 76%|███████▌ | 379/500 [28:26<09:08, 4.54s/it, loss=0.884, lr=0.000142]\nSteps: 76%|███████▌ | 379/500 [28:26<09:08, 4.54s/it, loss=0.514, lr=0.00014] \nSteps: 76%|███████▌ | 380/500 [28:28<07:28, 3.74s/it, loss=0.514, lr=0.00014]\nSteps: 76%|███████▌ | 380/500 [28:28<07:28, 3.74s/it, loss=0.419, lr=0.000138]\nSteps: 76%|███████▌ | 381/500 [28:29<06:18, 3.18s/it, loss=0.419, lr=0.000138]\nSteps: 76%|███████▌ | 381/500 [28:30<06:18, 3.18s/it, loss=0.354, lr=0.000136]\nSteps: 76%|███████▋ | 382/500 [28:37<08:53, 4.52s/it, loss=0.354, lr=0.000136]\nSteps: 76%|███████▋ | 382/500 [28:37<08:53, 4.52s/it, loss=1.03, lr=0.000134] \nSteps: 77%|███████▋ | 383/500 [28:39<07:16, 3.73s/it, loss=1.03, lr=0.000134]\nSteps: 77%|███████▋ | 383/500 [28:39<07:16, 3.73s/it, loss=0.412, lr=0.000132]\nSteps: 77%|███████▋ | 384/500 [28:41<06:08, 3.17s/it, loss=0.412, lr=0.000132]\nSteps: 77%|███████▋ | 384/500 [28:41<06:08, 3.17s/it, loss=0.284, lr=0.00013] \nSteps: 77%|███████▋ | 385/500 [28:49<08:40, 4.53s/it, loss=0.284, lr=0.00013]\nSteps: 77%|███████▋ | 385/500 [28:49<08:40, 4.53s/it, loss=0.513, lr=0.000128]\nSteps: 77%|███████▋ | 386/500 [28:50<07:05, 3.73s/it, loss=0.513, lr=0.000128]\nSteps: 77%|███████▋ | 386/500 [28:50<07:05, 3.73s/it, loss=0.705, lr=0.000126]\nSteps: 77%|███████▋ | 387/500 [28:52<05:59, 3.18s/it, loss=0.705, lr=0.000126]\nSteps: 77%|███████▋ | 387/500 [28:52<05:59, 3.18s/it, loss=0.31, lr=0.000124] \nSteps: 78%|███████▊ | 388/500 [29:00<08:26, 4.53s/it, loss=0.31, lr=0.000124]\nSteps: 78%|███████▊ | 388/500 [29:00<08:26, 4.53s/it, loss=0.572, lr=0.000122]\nSteps: 78%|███████▊ | 389/500 [29:02<06:54, 3.73s/it, loss=0.572, lr=0.000122]\nSteps: 78%|███████▊ | 389/500 [29:02<06:54, 3.73s/it, loss=0.433, lr=0.000121]\nSteps: 78%|███████▊ | 390/500 [29:04<05:49, 3.18s/it, loss=0.433, lr=0.000121]\nSteps: 78%|███████▊ | 390/500 [29:04<05:49, 3.18s/it, loss=0.263, lr=0.000119]\nSteps: 78%|███████▊ | 391/500 [29:12<08:16, 4.55s/it, loss=0.263, lr=0.000119]\nSteps: 78%|███████▊ | 391/500 [29:12<08:16, 4.55s/it, loss=0.898, lr=0.000117]\nSteps: 78%|███████▊ | 392/500 [29:13<06:45, 3.75s/it, loss=0.898, lr=0.000117]\nSteps: 78%|███████▊ | 392/500 [29:13<06:45, 3.75s/it, loss=0.825, lr=0.000115]\nSteps: 79%|███████▊ | 393/500 [29:15<05:41, 3.19s/it, loss=0.825, lr=0.000115]\nSteps: 79%|███████▊ | 393/500 [29:15<05:41, 3.19s/it, loss=1.07, lr=0.000113] \nSteps: 79%|███████▉ | 394/500 [29:23<08:03, 4.56s/it, loss=1.07, lr=0.000113]\nSteps: 79%|███████▉ | 394/500 [29:23<08:03, 4.56s/it, loss=0.497, lr=0.000111]\nSteps: 79%|███████▉ | 395/500 [29:25<06:34, 3.76s/it, loss=0.497, lr=0.000111]\nSteps: 79%|███████▉ | 395/500 [29:25<06:34, 3.76s/it, loss=0.42, lr=0.000109] \nSteps: 79%|███████▉ | 396/500 [29:27<05:32, 3.20s/it, loss=0.42, lr=0.000109]\nSteps: 79%|███████▉ | 396/500 [29:27<05:32, 3.20s/it, loss=0.336, lr=0.000107]\nSteps: 79%|███████▉ | 397/500 [29:34<07:41, 4.48s/it, loss=0.336, lr=0.000107]\nSteps: 79%|███████▉ | 397/500 [29:34<07:41, 4.48s/it, loss=0.591, lr=0.000105]\nSteps: 80%|███████▉ | 398/500 [29:36<06:17, 3.70s/it, loss=0.591, lr=0.000105]\nSteps: 80%|███████▉ | 398/500 [29:36<06:17, 3.70s/it, loss=0.437, lr=0.000104]\nSteps: 80%|███████▉ | 399/500 [29:38<05:18, 3.16s/it, loss=0.437, lr=0.000104]\nSteps: 80%|███████▉ | 399/500 [29:38<05:18, 3.16s/it, loss=0.59, lr=0.000102] \nSteps: 80%|████████ | 400/500 [29:46<07:31, 4.51s/it, loss=0.59, lr=0.000102]\nSteps: 80%|████████ | 400/500 [29:46<07:31, 4.51s/it, loss=0.799, lr=0.0001] \nSteps: 80%|████████ | 401/500 [29:48<06:08, 3.72s/it, loss=0.799, lr=0.0001]\nSteps: 80%|████████ | 401/500 [29:48<06:08, 3.72s/it, loss=1.03, lr=9.82e-5]\nSteps: 80%|████████ | 402/500 [29:50<05:10, 3.17s/it, loss=1.03, lr=9.82e-5]\nSteps: 80%|████████ | 402/500 [29:50<05:10, 3.17s/it, loss=0.283, lr=9.64e-5]\nSteps: 81%|████████ | 403/500 [29:57<07:21, 4.55s/it, loss=0.283, lr=9.64e-5]\nSteps: 81%|████████ | 403/500 [29:57<07:21, 4.55s/it, loss=1.2, lr=9.46e-5] \nSteps: 81%|████████ | 404/500 [29:59<05:59, 3.75s/it, loss=1.2, lr=9.46e-5]\nSteps: 81%|████████ | 404/500 [29:59<05:59, 3.75s/it, loss=0.414, lr=9.28e-5]\nSteps: 81%|████████ | 405/500 [30:01<05:02, 3.19s/it, loss=0.414, lr=9.28e-5]\nSteps: 81%|████████ | 405/500 [30:01<05:02, 3.19s/it, loss=0.562, lr=9.11e-5]\nSteps: 81%|████████ | 406/500 [30:09<07:06, 4.54s/it, loss=0.562, lr=9.11e-5]\nSteps: 81%|████████ | 406/500 [30:09<07:06, 4.54s/it, loss=0.673, lr=8.93e-5]\nSteps: 81%|████████▏ | 407/500 [30:11<05:47, 3.74s/it, loss=0.673, lr=8.93e-5]\nSteps: 81%|████████▏ | 407/500 [30:11<05:47, 3.74s/it, loss=0.474, lr=8.76e-5]\nSteps: 82%|████████▏ | 408/500 [30:13<04:52, 3.18s/it, loss=0.474, lr=8.76e-5]\nSteps: 82%|████████▏ | 408/500 [30:13<04:52, 3.18s/it, loss=0.376, lr=8.59e-5]\nSteps: 82%|████████▏ | 409/500 [30:20<06:46, 4.47s/it, loss=0.376, lr=8.59e-5]\nSteps: 82%|████████▏ | 409/500 [30:20<06:46, 4.47s/it, loss=0.502, lr=8.41e-5]\nSteps: 82%|████████▏ | 410/500 [30:22<05:32, 3.69s/it, loss=0.502, lr=8.41e-5]\nSteps: 82%|████████▏ | 410/500 [30:22<05:32, 3.69s/it, loss=0.54, lr=8.24e-5] \nSteps: 82%|████████▏ | 411/500 [30:24<04:40, 3.15s/it, loss=0.54, lr=8.24e-5]\nSteps: 82%|████████▏ | 411/500 [30:24<04:40, 3.15s/it, loss=0.569, lr=8.08e-5]\nSteps: 82%|████████▏ | 412/500 [30:31<06:35, 4.49s/it, loss=0.569, lr=8.08e-5]\nSteps: 82%|████████▏ | 412/500 [30:31<06:35, 4.49s/it, loss=0.659, lr=7.91e-5]\nSteps: 83%|████████▎ | 413/500 [30:33<05:22, 3.71s/it, loss=0.659, lr=7.91e-5]\nSteps: 83%|████████▎ | 413/500 [30:33<05:22, 3.71s/it, loss=0.976, lr=7.74e-5]\nSteps: 83%|████████▎ | 414/500 [30:35<04:31, 3.16s/it, loss=0.976, lr=7.74e-5]\nSteps: 83%|████████▎ | 414/500 [30:35<04:31, 3.16s/it, loss=0.258, lr=7.58e-5]\nSteps: 83%|████████▎ | 415/500 [30:43<06:21, 4.49s/it, loss=0.258, lr=7.58e-5]\nSteps: 83%|████████▎ | 415/500 [30:43<06:21, 4.49s/it, loss=0.508, lr=7.41e-5]\nSteps: 83%|████████▎ | 416/500 [30:45<05:11, 3.71s/it, loss=0.508, lr=7.41e-5]\nSteps: 83%|████████▎ | 416/500 [30:45<05:11, 3.71s/it, loss=0.791, lr=7.25e-5]\nSteps: 83%|████████▎ | 417/500 [30:47<04:22, 3.16s/it, loss=0.791, lr=7.25e-5]\nSteps: 83%|████████▎ | 417/500 [30:47<04:22, 3.16s/it, loss=0.898, lr=7.09e-5]\nSteps: 84%|████████▎ | 418/500 [30:54<06:10, 4.52s/it, loss=0.898, lr=7.09e-5]\nSteps: 84%|████████▎ | 418/500 [30:54<06:10, 4.52s/it, loss=0.492, lr=6.93e-5]\nSteps: 84%|████████▍ | 419/500 [30:56<05:01, 3.73s/it, loss=0.492, lr=6.93e-5]\nSteps: 84%|████████▍ | 419/500 [30:56<05:01, 3.73s/it, loss=0.432, lr=6.77e-5]\nSteps: 84%|████████▍ | 420/500 [30:58<04:13, 3.17s/it, loss=0.432, lr=6.77e-5]\nSteps: 84%|████████▍ | 420/500 [30:58<04:13, 3.17s/it, loss=0.651, lr=6.62e-5]\nSteps: 84%|████████▍ | 421/500 [31:06<05:57, 4.52s/it, loss=0.651, lr=6.62e-5]\nSteps: 84%|████████▍ | 421/500 [31:06<05:57, 4.52s/it, loss=0.569, lr=6.46e-5]\nSteps: 84%|████████▍ | 422/500 [31:08<04:51, 3.73s/it, loss=0.569, lr=6.46e-5]\nSteps: 84%|████████▍ | 422/500 [31:08<04:51, 3.73s/it, loss=0.95, lr=6.31e-5] \nSteps: 85%|████████▍ | 423/500 [31:09<04:04, 3.18s/it, loss=0.95, lr=6.31e-5]\nSteps: 85%|████████▍ | 423/500 [31:09<04:04, 3.18s/it, loss=0.262, lr=6.16e-5]\nSteps: 85%|████████▍ | 424/500 [31:17<05:44, 4.54s/it, loss=0.262, lr=6.16e-5]\nSteps: 85%|████████▍ | 424/500 [31:17<05:44, 4.54s/it, loss=0.5, lr=6.01e-5] \nSteps: 85%|████████▌ | 425/500 [31:19<04:40, 3.74s/it, loss=0.5, lr=6.01e-5]\nSteps: 85%|████████▌ | 425/500 [31:19<04:40, 3.74s/it, loss=0.715, lr=5.86e-5]\nSteps: 85%|████████▌ | 426/500 [31:21<03:55, 3.18s/it, loss=0.715, lr=5.86e-5]\nSteps: 85%|████████▌ | 426/500 [31:21<03:55, 3.18s/it, loss=0.514, lr=5.71e-5]\nSteps: 85%|████████▌ | 427/500 [31:29<05:31, 4.53s/it, loss=0.514, lr=5.71e-5]\nSteps: 85%|████████▌ | 427/500 [31:29<05:31, 4.53s/it, loss=0.494, lr=5.56e-5]\nSteps: 86%|████████▌ | 428/500 [31:30<04:29, 3.74s/it, loss=0.494, lr=5.56e-5]\nSteps: 86%|████████▌ | 428/500 [31:30<04:29, 3.74s/it, loss=0.679, lr=5.42e-5]\nSteps: 86%|████████▌ | 429/500 [31:32<03:45, 3.18s/it, loss=0.679, lr=5.42e-5]\nSteps: 86%|████████▌ | 429/500 [31:32<03:45, 3.18s/it, loss=0.998, lr=5.28e-5]\nSteps: 86%|████████▌ | 430/500 [31:40<05:17, 4.53s/it, loss=0.998, lr=5.28e-5]\nSteps: 86%|████████▌ | 430/500 [31:40<05:17, 4.53s/it, loss=0.495, lr=5.14e-5]\nSteps: 86%|████████▌ | 431/500 [31:42<04:17, 3.73s/it, loss=0.495, lr=5.14e-5]\nSteps: 86%|████████▌ | 431/500 [31:42<04:17, 3.73s/it, loss=0.771, lr=5e-5] \nSteps: 86%|████████▋ | 432/500 [31:44<03:36, 3.18s/it, loss=0.771, lr=5e-5]\nSteps: 86%|████████▋ | 432/500 [31:44<03:36, 3.18s/it, loss=0.876, lr=4.86e-5]\nSteps: 87%|████████▋ | 433/500 [31:51<05:02, 4.52s/it, loss=0.876, lr=4.86e-5]\nSteps: 87%|████████▋ | 433/500 [31:51<05:02, 4.52s/it, loss=0.559, lr=4.72e-5]\nSteps: 87%|████████▋ | 434/500 [31:53<04:05, 3.73s/it, loss=0.559, lr=4.72e-5]\nSteps: 87%|████████▋ | 434/500 [31:53<04:05, 3.73s/it, loss=0.625, lr=4.59e-5]\nSteps: 87%|████████▋ | 435/500 [31:55<03:26, 3.17s/it, loss=0.625, lr=4.59e-5]\nSteps: 87%|████████▋ | 435/500 [31:55<03:26, 3.17s/it, loss=0.262, lr=4.46e-5]\nSteps: 87%|████████▋ | 436/500 [32:03<04:48, 4.51s/it, loss=0.262, lr=4.46e-5]\nSteps: 87%|████████▋ | 436/500 [32:03<04:48, 4.51s/it, loss=0.577, lr=4.33e-5]\nSteps: 87%|████████▋ | 437/500 [32:05<03:54, 3.72s/it, loss=0.577, lr=4.33e-5]\nSteps: 87%|████████▋ | 437/500 [32:05<03:54, 3.72s/it, loss=0.416, lr=4.2e-5] \nSteps: 88%|████████▊ | 438/500 [32:07<03:16, 3.17s/it, loss=0.416, lr=4.2e-5]\nSteps: 88%|████████▊ | 438/500 [32:07<03:16, 3.17s/it, loss=0.516, lr=4.07e-5]\nSteps: 88%|████████▊ | 439/500 [32:14<04:35, 4.51s/it, loss=0.516, lr=4.07e-5]\nSteps: 88%|████████▊ | 439/500 [32:14<04:35, 4.51s/it, loss=0.498, lr=3.94e-5]\nSteps: 88%|████████▊ | 440/500 [32:16<03:43, 3.72s/it, loss=0.498, lr=3.94e-5]\nSteps: 88%|████████▊ | 440/500 [32:16<03:43, 3.72s/it, loss=0.76, lr=3.82e-5] \nSteps: 88%|████████▊ | 441/500 [32:18<03:07, 3.17s/it, loss=0.76, lr=3.82e-5]\nSteps: 88%|████████▊ | 441/500 [32:18<03:07, 3.17s/it, loss=0.486, lr=3.7e-5]\nSteps: 88%|████████▊ | 442/500 [32:26<04:21, 4.51s/it, loss=0.486, lr=3.7e-5]\nSteps: 88%|████████▊ | 442/500 [32:26<04:21, 4.51s/it, loss=0.9, lr=3.58e-5] \nSteps: 89%|████████▊ | 443/500 [32:27<03:32, 3.72s/it, loss=0.9, lr=3.58e-5]\nSteps: 89%|████████▊ | 443/500 [32:27<03:32, 3.72s/it, loss=0.658, lr=3.46e-5]\nSteps: 89%|████████▉ | 444/500 [32:29<02:57, 3.17s/it, loss=0.658, lr=3.46e-5]\nSteps: 89%|████████▉ | 444/500 [32:29<02:57, 3.17s/it, loss=0.726, lr=3.34e-5]\nSteps: 89%|████████▉ | 445/500 [32:37<04:08, 4.52s/it, loss=0.726, lr=3.34e-5]\nSteps: 89%|████████▉ | 445/500 [32:37<04:08, 4.52s/it, loss=0.512, lr=3.23e-5]\nSteps: 89%|████████▉ | 446/500 [32:39<03:21, 3.73s/it, loss=0.512, lr=3.23e-5]\nSteps: 89%|████████▉ | 446/500 [32:39<03:21, 3.73s/it, loss=0.468, lr=3.11e-5]\nSteps: 89%|████████▉ | 447/500 [32:41<02:48, 3.17s/it, loss=0.468, lr=3.11e-5]\nSteps: 89%|████████▉ | 447/500 [32:41<02:48, 3.17s/it, loss=0.318, lr=3e-5] \nSteps: 90%|████████▉ | 448/500 [32:48<03:55, 4.52s/it, loss=0.318, lr=3e-5]\nSteps: 90%|████████▉ | 448/500 [32:48<03:55, 4.52s/it, loss=0.527, lr=2.89e-5]\nSteps: 90%|████████▉ | 449/500 [32:50<03:10, 3.73s/it, loss=0.527, lr=2.89e-5]\nSteps: 90%|████████▉ | 449/500 [32:50<03:10, 3.73s/it, loss=0.433, lr=2.79e-5]\nSteps: 90%|█████████ | 450/500 [32:52<02:38, 3.18s/it, loss=0.433, lr=2.79e-5]\nSteps: 90%|█████████ | 450/500 [32:52<02:38, 3.18s/it, loss=0.859, lr=2.68e-5]\nSteps: 90%|█████████ | 451/500 [33:00<03:42, 4.54s/it, loss=0.859, lr=2.68e-5]\nSteps: 90%|█████████ | 451/500 [33:00<03:42, 4.54s/it, loss=0.539, lr=2.58e-5]\nSteps: 90%|█████████ | 452/500 [33:02<02:59, 3.74s/it, loss=0.539, lr=2.58e-5]\nSteps: 90%|█████████ | 452/500 [33:02<02:59, 3.74s/it, loss=0.422, lr=2.47e-5]\nSteps: 91%|█████████ | 453/500 [33:04<02:29, 3.19s/it, loss=0.422, lr=2.47e-5]\nSteps: 91%|█████████ | 453/500 [33:04<02:29, 3.19s/it, loss=0.258, lr=2.37e-5]\nSteps: 91%|█████████ | 454/500 [33:12<03:30, 4.57s/it, loss=0.258, lr=2.37e-5]\nSteps: 91%|█████████ | 454/500 [33:12<03:30, 4.57s/it, loss=0.517, lr=2.28e-5]\nSteps: 91%|█████████ | 455/500 [33:13<02:49, 3.76s/it, loss=0.517, lr=2.28e-5]\nSteps: 91%|█████████ | 455/500 [33:13<02:49, 3.76s/it, loss=1.03, lr=2.18e-5] \nSteps: 91%|█████████ | 456/500 [33:15<02:20, 3.20s/it, loss=1.03, lr=2.18e-5]\nSteps: 91%|█████████ | 456/500 [33:15<02:20, 3.20s/it, loss=0.708, lr=2.09e-5]\nSteps: 91%|█████████▏| 457/500 [33:23<03:14, 4.52s/it, loss=0.708, lr=2.09e-5]\nSteps: 91%|█████████▏| 457/500 [33:23<03:14, 4.52s/it, loss=0.763, lr=1.99e-5]\nSteps: 92%|█████████▏| 458/500 [33:25<02:36, 3.73s/it, loss=0.763, lr=1.99e-5]\nSteps: 92%|█████████▏| 458/500 [33:25<02:36, 3.73s/it, loss=1.03, lr=1.9e-5] \nSteps: 92%|█████████▏| 459/500 [33:27<02:10, 3.18s/it, loss=1.03, lr=1.9e-5]\nSteps: 92%|█████████▏| 459/500 [33:27<02:10, 3.18s/it, loss=0.278, lr=1.82e-5]\nSteps: 92%|█████████▏| 460/500 [33:34<03:01, 4.54s/it, loss=0.278, lr=1.82e-5]\nSteps: 92%|█████████▏| 460/500 [33:34<03:01, 4.54s/it, loss=0.628, lr=1.73e-5]\nSteps: 92%|█████████▏| 461/500 [33:36<02:25, 3.74s/it, loss=0.628, lr=1.73e-5]\nSteps: 92%|█████████▏| 461/500 [33:36<02:25, 3.74s/it, loss=0.912, lr=1.64e-5]\nSteps: 92%|█████████▏| 462/500 [33:38<02:00, 3.18s/it, loss=0.912, lr=1.64e-5]\nSteps: 92%|█████████▏| 462/500 [33:38<02:00, 3.18s/it, loss=0.257, lr=1.56e-5]\nSteps: 93%|█████████▎| 463/500 [33:46<02:48, 4.55s/it, loss=0.257, lr=1.56e-5]\nSteps: 93%|█████████▎| 463/500 [33:46<02:48, 4.55s/it, loss=1.24, lr=1.48e-5] \nSteps: 93%|█████████▎| 464/500 [33:48<02:14, 3.75s/it, loss=1.24, lr=1.48e-5]\nSteps: 93%|█████████▎| 464/500 [33:48<02:14, 3.75s/it, loss=0.45, lr=1.4e-5] \nSteps: 93%|█████████▎| 465/500 [33:50<01:51, 3.19s/it, loss=0.45, lr=1.4e-5]\nSteps: 93%|█████████▎| 465/500 [33:50<01:51, 3.19s/it, loss=0.263, lr=1.33e-5]\nSteps: 93%|█████████▎| 466/500 [33:57<02:33, 4.53s/it, loss=0.263, lr=1.33e-5]\nSteps: 93%|█████████▎| 466/500 [33:57<02:33, 4.53s/it, loss=0.56, lr=1.25e-5] \nSteps: 93%|█████████▎| 467/500 [33:59<02:03, 3.73s/it, loss=0.56, lr=1.25e-5]\nSteps: 93%|█████████▎| 467/500 [33:59<02:03, 3.73s/it, loss=0.629, lr=1.18e-5]\nSteps: 94%|█████████▎| 468/500 [34:01<01:41, 3.18s/it, loss=0.629, lr=1.18e-5]\nSteps: 94%|█████████▎| 468/500 [34:01<01:41, 3.18s/it, loss=0.263, lr=1.11e-5]\nSteps: 94%|█████████▍| 469/500 [34:09<02:20, 4.52s/it, loss=0.263, lr=1.11e-5]\nSteps: 94%|█████████▍| 469/500 [34:09<02:20, 4.52s/it, loss=0.918, lr=1.04e-5]\nSteps: 94%|█████████▍| 470/500 [34:11<01:51, 3.73s/it, loss=0.918, lr=1.04e-5]\nSteps: 94%|█████████▍| 470/500 [34:11<01:51, 3.73s/it, loss=0.991, lr=9.79e-6]\nSteps: 94%|█████████▍| 471/500 [34:12<01:32, 3.18s/it, loss=0.991, lr=9.79e-6]\nSteps: 94%|█████████▍| 471/500 [34:12<01:32, 3.18s/it, loss=0.33, lr=9.15e-6] \nSteps: 94%|█████████▍| 472/500 [34:20<02:06, 4.53s/it, loss=0.33, lr=9.15e-6]\nSteps: 94%|█████████▍| 472/500 [34:20<02:06, 4.53s/it, loss=0.536, lr=8.54e-6]\nSteps: 95%|█████████▍| 473/500 [34:22<01:40, 3.73s/it, loss=0.536, lr=8.54e-6]\nSteps: 95%|█████████▍| 473/500 [34:22<01:40, 3.73s/it, loss=0.487, lr=7.94e-6]\nSteps: 95%|█████████▍| 474/500 [34:24<01:22, 3.18s/it, loss=0.487, lr=7.94e-6]\nSteps: 95%|█████████▍| 474/500 [34:24<01:22, 3.18s/it, loss=0.262, lr=7.37e-6]\nSteps: 95%|█████████▌| 475/500 [34:32<01:53, 4.54s/it, loss=0.262, lr=7.37e-6]\nSteps: 95%|█████████▌| 475/500 [34:32<01:53, 4.54s/it, loss=0.487, lr=6.81e-6]\nSteps: 95%|█████████▌| 476/500 [34:34<01:29, 3.74s/it, loss=0.487, lr=6.81e-6]\nSteps: 95%|█████████▌| 476/500 [34:34<01:29, 3.74s/it, loss=0.797, lr=6.28e-6]\nSteps: 95%|█████████▌| 477/500 [34:35<01:13, 3.18s/it, loss=0.797, lr=6.28e-6]\nSteps: 95%|█████████▌| 477/500 [34:35<01:13, 3.18s/it, loss=0.99, lr=5.77e-6] \nSteps: 96%|█████████▌| 478/500 [34:43<01:38, 4.47s/it, loss=0.99, lr=5.77e-6]\nSteps: 96%|█████████▌| 478/500 [34:43<01:38, 4.47s/it, loss=0.612, lr=5.28e-6]\nSteps: 96%|█████████▌| 479/500 [34:45<01:17, 3.70s/it, loss=0.612, lr=5.28e-6]\nSteps: 96%|█████████▌| 479/500 [34:45<01:17, 3.70s/it, loss=0.422, lr=4.82e-6]\nSteps: 96%|█████████▌| 480/500 [34:47<01:03, 3.15s/it, loss=0.422, lr=4.82e-6]\nSteps: 96%|█████████▌| 480/500 [34:47<01:03, 3.15s/it, loss=0.384, lr=4.37e-6]\nSteps: 96%|█████████▌| 481/500 [34:54<01:25, 4.48s/it, loss=0.384, lr=4.37e-6]\nSteps: 96%|█████████▌| 481/500 [34:54<01:25, 4.48s/it, loss=1.03, lr=3.95e-6] \nSteps: 96%|█████████▋| 482/500 [34:56<01:06, 3.70s/it, loss=1.03, lr=3.95e-6]\nSteps: 96%|█████████▋| 482/500 [34:56<01:06, 3.70s/it, loss=0.76, lr=3.54e-6]\nSteps: 97%|█████████▋| 483/500 [34:58<00:53, 3.16s/it, loss=0.76, lr=3.54e-6]\nSteps: 97%|█████████▋| 483/500 [34:58<00:53, 3.16s/it, loss=0.366, lr=3.16e-6]\nSteps: 97%|█████████▋| 484/500 [35:06<01:12, 4.52s/it, loss=0.366, lr=3.16e-6]\nSteps: 97%|█████████▋| 484/500 [35:06<01:12, 4.52s/it, loss=0.491, lr=2.8e-6] \nSteps: 97%|█████████▋| 485/500 [35:08<00:55, 3.73s/it, loss=0.491, lr=2.8e-6]\nSteps: 97%|█████████▋| 485/500 [35:08<00:55, 3.73s/it, loss=0.633, lr=2.46e-6]\nSteps: 97%|█████████▋| 486/500 [35:09<00:44, 3.18s/it, loss=0.633, lr=2.46e-6]\nSteps: 97%|█████████▋| 486/500 [35:09<00:44, 3.18s/it, loss=1.02, lr=2.15e-6] \nSteps: 97%|█████████▋| 487/500 [35:17<00:58, 4.48s/it, loss=1.02, lr=2.15e-6]\nSteps: 97%|█████████▋| 487/500 [35:17<00:58, 4.48s/it, loss=0.521, lr=1.85e-6]\nSteps: 98%|█████████▊| 488/500 [35:19<00:44, 3.70s/it, loss=0.521, lr=1.85e-6]\nSteps: 98%|█████████▊| 488/500 [35:19<00:44, 3.70s/it, loss=0.488, lr=1.58e-6]\nSteps: 98%|█████████▊| 489/500 [35:21<00:34, 3.16s/it, loss=0.488, lr=1.58e-6]\nSteps: 98%|█████████▊| 489/500 [35:21<00:34, 3.16s/it, loss=0.295, lr=1.33e-6]\nSteps: 98%|█████████▊| 490/500 [35:28<00:44, 4.48s/it, loss=0.295, lr=1.33e-6]\nSteps: 98%|█████████▊| 490/500 [35:28<00:44, 4.48s/it, loss=0.598, lr=1.1e-6] \nSteps: 98%|█████████▊| 491/500 [35:30<00:33, 3.70s/it, loss=0.598, lr=1.1e-6]\nSteps: 98%|█████████▊| 491/500 [35:30<00:33, 3.70s/it, loss=0.925, lr=8.88e-7]\nSteps: 98%|█████████▊| 492/500 [35:32<00:25, 3.16s/it, loss=0.925, lr=8.88e-7]\nSteps: 98%|█████████▊| 492/500 [35:32<00:25, 3.16s/it, loss=0.259, lr=7.01e-7]\nSteps: 99%|█████████▊| 493/500 [35:40<00:31, 4.53s/it, loss=0.259, lr=7.01e-7]\nSteps: 99%|█████████▊| 493/500 [35:40<00:31, 4.53s/it, loss=0.579, lr=5.37e-7]\nSteps: 99%|█████████▉| 494/500 [35:42<00:22, 3.74s/it, loss=0.579, lr=5.37e-7]\nSteps: 99%|█████████▉| 494/500 [35:42<00:22, 3.74s/it, loss=0.505, lr=3.95e-7]\nSteps: 99%|█████████▉| 495/500 [35:44<00:15, 3.18s/it, loss=0.505, lr=3.95e-7]\nSteps: 99%|█████████▉| 495/500 [35:44<00:15, 3.18s/it, loss=0.429, lr=2.74e-7]\nSteps: 99%|█████████▉| 496/500 [35:51<00:18, 4.50s/it, loss=0.429, lr=2.74e-7]\nSteps: 99%|█████████▉| 496/500 [35:51<00:18, 4.50s/it, loss=0.827, lr=1.75e-7]\nSteps: 99%|█████████▉| 497/500 [35:53<00:11, 3.72s/it, loss=0.827, lr=1.75e-7]\nSteps: 99%|█████████▉| 497/500 [35:53<00:11, 3.72s/it, loss=0.435, lr=9.87e-8]\nSteps: 100%|█████████▉| 498/500 [35:55<00:06, 3.17s/it, loss=0.435, lr=9.87e-8]\nSteps: 100%|█████████▉| 498/500 [35:55<00:06, 3.17s/it, loss=0.473, lr=4.39e-8]\nSteps: 100%|█████████▉| 499/500 [36:03<00:04, 4.53s/it, loss=0.473, lr=4.39e-8]\nSteps: 100%|█████████▉| 499/500 [36:03<00:04, 4.53s/it, loss=1.04, lr=1.1e-8] \nSteps: 100%|██████████| 500/500 [36:05<00:00, 3.73s/it, loss=1.04, lr=1.1e-8]\nSteps: 100%|██████████| 500/500 [36:05<00:00, 3.73s/it, loss=0.847, lr=0] \nSteps: 100%|██████████| 500/500 [36:09<00:00, 4.34s/it, loss=0.847, lr=0]\n---Tar up output directory---\nmochi-lora/\nmochi-lora/pytorch_lora_weights.safetensors\nUploading to Hugging Face: lucataco/mochi-lora-melty\nHF Repo URL: https://huggingface.co/lucataco/mochi-lora-melty\npytorch_lora_weights.safetensors: 0%| | 0.00/76.1M [00:00<?, ?B/s]\npytorch_lora_weights.safetensors: 11%|█ | 8.52M/76.1M [00:00<00:00, 85.2MB/s]\npytorch_lora_weights.safetensors: 22%|██▏ | 17.0M/76.1M [00:00<00:01, 51.7MB/s]\npytorch_lora_weights.safetensors: 42%|████▏ | 32.0M/76.1M [00:00<00:00, 55.8MB/s]\npytorch_lora_weights.safetensors: 63%|██████▎ | 48.0M/76.1M [00:00<00:00, 56.2MB/s]\npytorch_lora_weights.safetensors: 84%|████████▍ | 64.0M/76.1M [00:01<00:00, 61.6MB/s]\npytorch_lora_weights.safetensors: 100%|██████████| 76.1M/76.1M [00:01<00:00, 54.3MB/s]\nSuccessfully uploaded model to https://huggingface.co/lucataco/mochi-lora-melty", "metrics": { "predict_time": 2194.219378981, "total_time": 2423.855873 }, "output": { "weights": "https://replicate.delivery/xezq/V7qtfPfCvAqSdEv2yHhornGlbFH3IQzPzhfrFe6a1SB1I7oPB/trained_model.tar" }, "started_at": "2024-12-12T18:56:59.564494Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/jnmeawnxn1rma0ckqgvr6t5q28", "cancel": "https://api.replicate.com/v1/predictions/jnmeawnxn1rma0ckqgvr6t5q28/cancel" }, "version": "9f7c62841ad839d7091201100d03dd778e2a78ca702fa65addd69ecbd511f64e" }
Generated inCleaning up previous runs Extracted 6 files from zip to videos_input ---Starting to Trim input videos--- Processing: videos_input/cat.mov videos_input/cat.mov as target resolution 480x848 is larger than input 668x650. So, upsampling the video. Copied videos_input/cat.txt to videos_prepared/cat.txt Moviepy - Building video videos_prepared/cat.mp4. Moviepy - Writing video videos_prepared/cat.mp4 0%| | 0/3 [00:00<?, ?it/s] 0%| | 0/3 [00:00<?, ?it/s] 0%| | 0/3 [00:00<?, ?it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 65%|██████▌ | 26/40 [00:00<00:00, 254.70it/s, now=None] 0%| | 0/3 [00:00<?, ?it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/cat.mp4 0%| | 0/3 [00:00<?, ?it/s] Processing: videos_input/glove.mov Copied videos_input/glove.txt to videos_prepared/glove.txt Moviepy - Building video videos_prepared/glove.mp4. Moviepy - Writing video videos_prepared/glove.mp4 33%|███▎ | 1/3 [00:00<00:00, 2.34it/s] 33%|███▎ | 1/3 [00:00<00:00, 2.34it/s] 33%|███▎ | 1/3 [00:00<00:00, 2.34it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 15%|█▌ | 6/40 [00:00<00:00, 58.08it/s, now=None] t: 30%|███ | 12/40 [00:00<00:00, 53.57it/s, now=None] t: 45%|████▌ | 18/40 [00:00<00:00, 52.35it/s, now=None] t: 60%|██████ | 24/40 [00:00<00:00, 51.47it/s, now=None] t: 75%|███████▌ | 30/40 [00:00<00:00, 50.92it/s, now=None] t: 90%|█████████ | 36/40 [00:00<00:00, 50.87it/s, now=None] 33%|███▎ | 1/3 [00:01<00:00, 2.34it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/glove.mp4 33%|███▎ | 1/3 [00:01<00:00, 2.34it/s] Processing: videos_input/straws.mov videos_input/straws.mov as target resolution 480x848 is larger than input 668x514. So, upsampling the video. Copied videos_input/straws.txt to videos_prepared/straws.txt Moviepy - Building video videos_prepared/straws.mp4. Moviepy - Writing video videos_prepared/straws.mp4 67%|██████▋ | 2/3 [00:01<00:00, 1.17it/s] 67%|██████▋ | 2/3 [00:01<00:00, 1.17it/s] 67%|██████▋ | 2/3 [00:01<00:00, 1.17it/s] t: 0%| | 0/40 [00:00<?, ?it/s, now=None] t: 72%|███████▎ | 29/40 [00:00<00:00, 286.01it/s, now=None] 67%|██████▋ | 2/3 [00:01<00:00, 1.17it/s] Moviepy - Done ! Moviepy - video ready videos_prepared/straws.mp4 67%|██████▋ | 2/3 [00:01<00:00, 1.17it/s] 100%|██████████| 3/3 [00:01<00:00, 1.59it/s] 100%|██████████| 3/3 [00:01<00:00, 1.54it/s] ---Starting to Embed videos--- Loading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s] Loading checkpoint shards: 50%|█████ | 1/2 [00:00<00:00, 1.84it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:01<00:00, 1.88it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:01<00:00, 1.87it/s] Loading pipeline components...: 0%| | 0/3 [00:00<?, ?it/s] Loading pipeline components...: 100%|██████████| 3/3 [00:00<00:00, 670.55it/s] Processing videos_prepared/cat.mp4 Trimmed video from 40 to first 37 frames 0it [00:00, ?it/s] Processing videos_prepared/glove.mp4 Trimmed video from 40 to first 37 frames 1it [00:01, 1.64s/it] Processing videos_prepared/straws.mp4 Trimmed video from 40 to first 37 frames 2it [00:02, 1.30s/it] 3it [00:03, 1.15s/it] 3it [00:03, 1.23s/it] ---Starting training--- Found 3 training videos in videos_prepared Loaded 3/3 valid file pairs. ===== Memory before training ===== memory_allocated=18.903 GB max_memory_allocated=18.903 GB max_memory_reserved=28.078 GB ***** Running training ***** Num trainable parameters = 19005440 Num examples = 3 Num batches each epoch = 3 Num epochs = 167 Instantaneous batch size per device = 1 Total train batch size (w. parallel, distributed & accumulation) = 1 Total optimization steps = 500 Steps: 0%| | 0/500 [00:00<?, ?it/s]W1212 18:59:29.908000 138667870809600 torch/fx/experimental/symbolic_shapes.py:4449] [0/0] xindex is not in var_ranges, defaulting to unknown range. W1212 18:59:29.922000 138667870809600 torch/fx/experimental/symbolic_shapes.py:4449] [0/0] xindex is not in var_ranges, defaulting to unknown range. W1212 18:59:30.058000 138667870809600 torch/fx/experimental/symbolic_shapes.py:4449] [0/0] xindex is not in var_ranges, defaulting to unknown range. Steps: 0%| | 1/500 [04:22<36:20:40, 262.21s/it] Steps: 0%| | 1/500 [04:22<36:20:40, 262.21s/it, loss=0.758, lr=2e-6] Steps: 0%| | 2/500 [04:24<15:05:07, 109.05s/it, loss=0.758, lr=2e-6] Steps: 0%| | 2/500 [04:24<15:05:07, 109.05s/it, loss=0.492, lr=4e-6] Steps: 1%| | 3/500 [04:25<8:17:55, 60.11s/it, loss=0.492, lr=4e-6] Steps: 1%| | 3/500 [04:25<8:17:55, 60.11s/it, loss=0.291, lr=6e-6] Steps: 1%| | 4/500 [04:34<5:27:59, 39.68s/it, loss=0.291, lr=6e-6] Steps: 1%| | 4/500 [04:34<5:27:59, 39.68s/it, loss=0.581, lr=8e-6] Steps: 1%| | 5/500 [04:36<3:34:51, 26.04s/it, loss=0.581, lr=8e-6] Steps: 1%| | 5/500 [04:36<3:34:51, 26.04s/it, loss=0.675, lr=1e-5] Steps: 1%| | 6/500 [04:38<2:26:44, 17.82s/it, loss=0.675, lr=1e-5] Steps: 1%| | 6/500 [04:38<2:26:44, 17.82s/it, loss=0.387, lr=1.2e-5] Steps: 1%|▏ | 7/500 [04:46<2:02:39, 14.93s/it, loss=0.387, lr=1.2e-5] Steps: 1%|▏ | 7/500 [04:46<2:02:39, 14.93s/it, loss=1.16, lr=1.4e-5] Steps: 2%|▏ | 8/500 [04:48<1:28:19, 10.77s/it, loss=1.16, lr=1.4e-5] Steps: 2%|▏ | 8/500 [04:48<1:28:19, 10.77s/it, loss=0.585, lr=1.6e-5] Steps: 2%|▏ | 9/500 [04:50<1:05:23, 7.99s/it, loss=0.585, lr=1.6e-5] Steps: 2%|▏ | 9/500 [04:50<1:05:23, 7.99s/it, loss=0.294, lr=1.8e-5] Steps: 2%|▏ | 10/500 [04:59<1:06:54, 8.19s/it, loss=0.294, lr=1.8e-5] Steps: 2%|▏ | 10/500 [04:59<1:06:54, 8.19s/it, loss=0.598, lr=2e-5] Steps: 2%|▏ | 11/500 [05:01<51:00, 6.26s/it, loss=0.598, lr=2e-5] Steps: 2%|▏ | 11/500 [05:01<51:00, 6.26s/it, loss=0.949, lr=2.2e-5] Steps: 2%|▏ | 12/500 [05:03<40:03, 4.93s/it, loss=0.949, lr=2.2e-5] Steps: 2%|▏ | 12/500 [05:03<40:03, 4.93s/it, loss=0.543, lr=2.4e-5] Steps: 3%|▎ | 13/500 [05:10<46:42, 5.75s/it, loss=0.543, lr=2.4e-5] Steps: 3%|▎ | 13/500 [05:10<46:42, 5.75s/it, loss=0.574, lr=2.6e-5] Steps: 3%|▎ | 14/500 [05:12<37:08, 4.58s/it, loss=0.574, lr=2.6e-5] Steps: 3%|▎ | 14/500 [05:12<37:08, 4.58s/it, loss=0.671, lr=2.8e-5] Steps: 3%|▎ | 15/500 [05:14<30:27, 3.77s/it, loss=0.671, lr=2.8e-5] Steps: 3%|▎ | 15/500 [05:14<30:27, 3.77s/it, loss=0.666, lr=3e-5] Steps: 3%|▎ | 16/500 [05:22<39:55, 4.95s/it, loss=0.666, lr=3e-5] Steps: 3%|▎ | 16/500 [05:22<39:55, 4.95s/it, loss=1.08, lr=3.2e-5] Steps: 3%|▎ | 17/500 [05:24<32:23, 4.02s/it, loss=1.08, lr=3.2e-5] Steps: 3%|▎ | 17/500 [05:24<32:23, 4.02s/it, loss=1.05, lr=3.4e-5] Steps: 4%|▎ | 18/500 [05:25<27:08, 3.38s/it, loss=1.05, lr=3.4e-5] Steps: 4%|▎ | 18/500 [05:25<27:08, 3.38s/it, loss=0.292, lr=3.6e-5] Steps: 4%|▍ | 19/500 [05:33<37:18, 4.65s/it, loss=0.292, lr=3.6e-5] Steps: 4%|▍ | 19/500 [05:33<37:18, 4.65s/it, loss=0.574, lr=3.8e-5] Steps: 4%|▍ | 20/500 [05:35<30:34, 3.82s/it, loss=0.574, lr=3.8e-5] Steps: 4%|▍ | 20/500 [05:35<30:34, 3.82s/it, loss=0.575, lr=4e-5] Steps: 4%|▍ | 21/500 [05:37<25:51, 3.24s/it, loss=0.575, lr=4e-5] Steps: 4%|▍ | 21/500 [05:37<25:51, 3.24s/it, loss=0.855, lr=4.2e-5] Steps: 4%|▍ | 22/500 [05:45<36:30, 4.58s/it, loss=0.855, lr=4.2e-5] Steps: 4%|▍ | 22/500 [05:45<36:30, 4.58s/it, loss=0.625, lr=4.4e-5] Steps: 5%|▍ | 23/500 [05:46<29:59, 3.77s/it, loss=0.625, lr=4.4e-5] Steps: 5%|▍ | 23/500 [05:46<29:59, 3.77s/it, loss=0.851, lr=4.6e-5] Steps: 5%|▍ | 24/500 [05:48<25:25, 3.20s/it, loss=0.851, lr=4.6e-5] Steps: 5%|▍ | 24/500 [05:48<25:25, 3.20s/it, loss=0.878, lr=4.8e-5] Steps: 5%|▌ | 25/500 [05:56<35:52, 4.53s/it, loss=0.878, lr=4.8e-5] Steps: 5%|▌ | 25/500 [05:56<35:52, 4.53s/it, loss=0.626, lr=5e-5] Steps: 5%|▌ | 26/500 [05:58<29:30, 3.73s/it, loss=0.626, lr=5e-5] Steps: 5%|▌ | 26/500 [05:58<29:30, 3.73s/it, loss=0.46, lr=5.2e-5] Steps: 5%|▌ | 27/500 [06:00<25:04, 3.18s/it, loss=0.46, lr=5.2e-5] Steps: 5%|▌ | 27/500 [06:00<25:04, 3.18s/it, loss=0.653, lr=5.4e-5] Steps: 6%|▌ | 28/500 [06:07<35:41, 4.54s/it, loss=0.653, lr=5.4e-5] Steps: 6%|▌ | 28/500 [06:07<35:41, 4.54s/it, loss=0.579, lr=5.6e-5] Steps: 6%|▌ | 29/500 [06:09<29:20, 3.74s/it, loss=0.579, lr=5.6e-5] Steps: 6%|▌ | 29/500 [06:09<29:20, 3.74s/it, loss=0.698, lr=5.8e-5] Steps: 6%|▌ | 30/500 [06:11<24:54, 3.18s/it, loss=0.698, lr=5.8e-5] Steps: 6%|▌ | 30/500 [06:11<24:54, 3.18s/it, loss=0.557, lr=6e-5] Steps: 6%|▌ | 31/500 [06:19<35:33, 4.55s/it, loss=0.557, lr=6e-5] Steps: 6%|▌ | 31/500 [06:19<35:33, 4.55s/it, loss=1.19, lr=6.2e-5] Steps: 6%|▋ | 32/500 [06:21<29:13, 3.75s/it, loss=1.19, lr=6.2e-5] Steps: 6%|▋ | 32/500 [06:21<29:13, 3.75s/it, loss=0.738, lr=6.4e-5] Steps: 7%|▋ | 33/500 [06:23<24:48, 3.19s/it, loss=0.738, lr=6.4e-5] Steps: 7%|▋ | 33/500 [06:23<24:48, 3.19s/it, loss=0.294, lr=6.6e-5] Steps: 7%|▋ | 34/500 [06:30<35:21, 4.55s/it, loss=0.294, lr=6.6e-5] Steps: 7%|▋ | 34/500 [06:30<35:21, 4.55s/it, loss=0.585, lr=6.8e-5] Steps: 7%|▋ | 35/500 [06:32<29:03, 3.75s/it, loss=0.585, lr=6.8e-5] Steps: 7%|▋ | 35/500 [06:32<29:03, 3.75s/it, loss=0.594, lr=7e-5] Steps: 7%|▋ | 36/500 [06:34<24:40, 3.19s/it, loss=0.594, lr=7e-5] Steps: 7%|▋ | 36/500 [06:34<24:40, 3.19s/it, loss=0.303, lr=7.2e-5] Steps: 7%|▋ | 37/500 [06:42<34:55, 4.52s/it, loss=0.303, lr=7.2e-5] Steps: 7%|▋ | 37/500 [06:42<34:55, 4.52s/it, loss=0.561, lr=7.4e-5] Steps: 8%|▊ | 38/500 [06:44<28:43, 3.73s/it, loss=0.561, lr=7.4e-5] Steps: 8%|▊ | 38/500 [06:44<28:43, 3.73s/it, loss=0.658, lr=7.6e-5] Steps: 8%|▊ | 39/500 [06:46<24:23, 3.18s/it, loss=0.658, lr=7.6e-5] Steps: 8%|▊ | 39/500 [06:46<24:23, 3.18s/it, loss=0.299, lr=7.8e-5] Steps: 8%|▊ | 40/500 [06:53<34:23, 4.49s/it, loss=0.299, lr=7.8e-5] Steps: 8%|▊ | 40/500 [06:53<34:23, 4.49s/it, loss=0.773, lr=8e-5] Steps: 8%|▊ | 41/500 [06:55<28:19, 3.70s/it, loss=0.773, lr=8e-5] Steps: 8%|▊ | 41/500 [06:55<28:19, 3.70s/it, loss=0.454, lr=8.2e-5] Steps: 8%|▊ | 42/500 [06:57<24:05, 3.16s/it, loss=0.454, lr=8.2e-5] Steps: 8%|▊ | 42/500 [06:57<24:05, 3.16s/it, loss=0.344, lr=8.4e-5] Steps: 9%|▊ | 43/500 [07:05<34:21, 4.51s/it, loss=0.344, lr=8.4e-5] Steps: 9%|▊ | 43/500 [07:05<34:21, 4.51s/it, loss=0.608, lr=8.6e-5] Steps: 9%|▉ | 44/500 [07:06<28:17, 3.72s/it, loss=0.608, lr=8.6e-5] Steps: 9%|▉ | 44/500 [07:06<28:17, 3.72s/it, loss=0.461, lr=8.8e-5] Steps: 9%|▉ | 45/500 [07:08<24:01, 3.17s/it, loss=0.461, lr=8.8e-5] Steps: 9%|▉ | 45/500 [07:08<24:01, 3.17s/it, loss=0.291, lr=9e-5] Steps: 9%|▉ | 46/500 [07:16<34:17, 4.53s/it, loss=0.291, lr=9e-5] Steps: 9%|▉ | 46/500 [07:16<34:17, 4.53s/it, loss=0.671, lr=9.2e-5] Steps: 9%|▉ | 47/500 [07:18<28:11, 3.73s/it, loss=0.671, lr=9.2e-5] Steps: 9%|▉ | 47/500 [07:18<28:11, 3.73s/it, loss=0.6, lr=9.4e-5] Steps: 10%|▉ | 48/500 [07:20<23:56, 3.18s/it, loss=0.6, lr=9.4e-5] Steps: 10%|▉ | 48/500 [07:20<23:56, 3.18s/it, loss=0.285, lr=9.6e-5] Steps: 10%|▉ | 49/500 [07:27<34:01, 4.53s/it, loss=0.285, lr=9.6e-5] Steps: 10%|▉ | 49/500 [07:27<34:01, 4.53s/it, loss=1.07, lr=9.8e-5] Steps: 10%|█ | 50/500 [07:29<27:59, 3.73s/it, loss=1.07, lr=9.8e-5] Steps: 10%|█ | 50/500 [07:29<27:59, 3.73s/it, loss=0.633, lr=0.0001] Steps: 10%|█ | 51/500 [07:31<23:46, 3.18s/it, loss=0.633, lr=0.0001] Steps: 10%|█ | 51/500 [07:31<23:46, 3.18s/it, loss=0.29, lr=0.000102] Steps: 10%|█ | 52/500 [07:39<33:35, 4.50s/it, loss=0.29, lr=0.000102] Steps: 10%|█ | 52/500 [07:39<33:35, 4.50s/it, loss=0.555, lr=0.000104] Steps: 11%|█ | 53/500 [07:41<27:39, 3.71s/it, loss=0.555, lr=0.000104] Steps: 11%|█ | 53/500 [07:41<27:39, 3.71s/it, loss=0.457, lr=0.000106] Steps: 11%|█ | 54/500 [07:43<23:30, 3.16s/it, loss=0.457, lr=0.000106] Steps: 11%|█ | 54/500 [07:43<23:30, 3.16s/it, loss=0.365, lr=0.000108] Steps: 11%|█ | 55/500 [07:50<33:32, 4.52s/it, loss=0.365, lr=0.000108] Steps: 11%|█ | 55/500 [07:50<33:32, 4.52s/it, loss=0.618, lr=0.00011] Steps: 11%|█ | 56/500 [07:52<27:35, 3.73s/it, loss=0.618, lr=0.00011] Steps: 11%|█ | 56/500 [07:52<27:35, 3.73s/it, loss=0.464, lr=0.000112] Steps: 11%|█▏ | 57/500 [07:54<23:25, 3.17s/it, loss=0.464, lr=0.000112] Steps: 11%|█▏ | 57/500 [07:54<23:25, 3.17s/it, loss=0.283, lr=0.000114] Steps: 12%|█▏ | 58/500 [08:02<33:09, 4.50s/it, loss=0.283, lr=0.000114] Steps: 12%|█▏ | 58/500 [08:02<33:09, 4.50s/it, loss=0.575, lr=0.000116] Steps: 12%|█▏ | 59/500 [08:03<27:18, 3.72s/it, loss=0.575, lr=0.000116] Steps: 12%|█▏ | 59/500 [08:04<27:18, 3.72s/it, loss=0.539, lr=0.000118] Steps: 12%|█▏ | 60/500 [08:05<23:12, 3.16s/it, loss=0.539, lr=0.000118] Steps: 12%|█▏ | 60/500 [08:05<23:12, 3.16s/it, loss=0.595, lr=0.00012] Steps: 12%|█▏ | 61/500 [08:13<33:02, 4.52s/it, loss=0.595, lr=0.00012] Steps: 12%|█▏ | 61/500 [08:13<33:02, 4.52s/it, loss=1.18, lr=0.000122] Steps: 12%|█▏ | 62/500 [08:15<27:11, 3.73s/it, loss=1.18, lr=0.000122] Steps: 12%|█▏ | 62/500 [08:15<27:11, 3.73s/it, loss=0.447, lr=0.000124] Steps: 13%|█▎ | 63/500 [08:17<23:06, 3.17s/it, loss=0.447, lr=0.000124] Steps: 13%|█▎ | 63/500 [08:17<23:06, 3.17s/it, loss=0.352, lr=0.000126] Steps: 13%|█▎ | 64/500 [08:24<32:43, 4.50s/it, loss=0.352, lr=0.000126] Steps: 13%|█▎ | 64/500 [08:24<32:43, 4.50s/it, loss=0.631, lr=0.000128] Steps: 13%|█▎ | 65/500 [08:26<26:56, 3.72s/it, loss=0.631, lr=0.000128] Steps: 13%|█▎ | 65/500 [08:26<26:56, 3.72s/it, loss=0.571, lr=0.00013] Steps: 13%|█▎ | 66/500 [08:28<22:54, 3.17s/it, loss=0.571, lr=0.00013] Steps: 13%|█▎ | 66/500 [08:28<22:54, 3.17s/it, loss=0.544, lr=0.000132] Steps: 13%|█▎ | 67/500 [08:36<32:43, 4.53s/it, loss=0.544, lr=0.000132] Steps: 13%|█▎ | 67/500 [08:36<32:43, 4.53s/it, loss=0.549, lr=0.000134] Steps: 14%|█▎ | 68/500 [08:38<26:55, 3.74s/it, loss=0.549, lr=0.000134] Steps: 14%|█▎ | 68/500 [08:38<26:55, 3.74s/it, loss=0.488, lr=0.000136] Steps: 14%|█▍ | 69/500 [08:40<22:51, 3.18s/it, loss=0.488, lr=0.000136] Steps: 14%|█▍ | 69/500 [08:40<22:51, 3.18s/it, loss=0.983, lr=0.000138] Steps: 14%|█▍ | 70/500 [08:47<32:09, 4.49s/it, loss=0.983, lr=0.000138] Steps: 14%|█▍ | 70/500 [08:47<32:09, 4.49s/it, loss=0.544, lr=0.00014] Steps: 14%|█▍ | 71/500 [08:49<26:28, 3.70s/it, loss=0.544, lr=0.00014] Steps: 14%|█▍ | 71/500 [08:49<26:28, 3.70s/it, loss=0.832, lr=0.000142] Steps: 14%|█▍ | 72/500 [08:51<22:31, 3.16s/it, loss=0.832, lr=0.000142] Steps: 14%|█▍ | 72/500 [08:51<22:31, 3.16s/it, loss=0.288, lr=0.000144] Steps: 15%|█▍ | 73/500 [08:59<32:10, 4.52s/it, loss=0.288, lr=0.000144] Steps: 15%|█▍ | 73/500 [08:59<32:10, 4.52s/it, loss=0.844, lr=0.000146] Steps: 15%|█▍ | 74/500 [09:01<26:29, 3.73s/it, loss=0.844, lr=0.000146] Steps: 15%|█▍ | 74/500 [09:01<26:29, 3.73s/it, loss=0.457, lr=0.000148] Steps: 15%|█▌ | 75/500 [09:02<22:29, 3.18s/it, loss=0.457, lr=0.000148] Steps: 15%|█▌ | 75/500 [09:02<22:29, 3.18s/it, loss=0.39, lr=0.00015] Steps: 15%|█▌ | 76/500 [09:10<31:42, 4.49s/it, loss=0.39, lr=0.00015] Steps: 15%|█▌ | 76/500 [09:10<31:42, 4.49s/it, loss=0.668, lr=0.000152] Steps: 15%|█▌ | 77/500 [09:12<26:07, 3.70s/it, loss=0.668, lr=0.000152] Steps: 15%|█▌ | 77/500 [09:12<26:07, 3.70s/it, loss=1, lr=0.000154] Steps: 16%|█▌ | 78/500 [09:14<22:12, 3.16s/it, loss=1, lr=0.000154] Steps: 16%|█▌ | 78/500 [09:14<22:12, 3.16s/it, loss=0.336, lr=0.000156] Steps: 16%|█▌ | 79/500 [09:21<31:36, 4.50s/it, loss=0.336, lr=0.000156] Steps: 16%|█▌ | 79/500 [09:21<31:36, 4.50s/it, loss=0.569, lr=0.000158] Steps: 16%|█▌ | 80/500 [09:23<26:01, 3.72s/it, loss=0.569, lr=0.000158] Steps: 16%|█▌ | 80/500 [09:23<26:01, 3.72s/it, loss=0.543, lr=0.00016] Steps: 16%|█▌ | 81/500 [09:25<22:08, 3.17s/it, loss=0.543, lr=0.00016] Steps: 16%|█▌ | 81/500 [09:25<22:08, 3.17s/it, loss=0.703, lr=0.000162] Steps: 16%|█▋ | 82/500 [09:33<31:23, 4.51s/it, loss=0.703, lr=0.000162] Steps: 16%|█▋ | 82/500 [09:33<31:23, 4.51s/it, loss=0.54, lr=0.000164] Steps: 17%|█▋ | 83/500 [09:35<25:50, 3.72s/it, loss=0.54, lr=0.000164] Steps: 17%|█▋ | 83/500 [09:35<25:50, 3.72s/it, loss=0.485, lr=0.000166] Steps: 17%|█▋ | 84/500 [09:37<21:58, 3.17s/it, loss=0.485, lr=0.000166] Steps: 17%|█▋ | 84/500 [09:37<21:58, 3.17s/it, loss=0.348, lr=0.000168] Steps: 17%|█▋ | 85/500 [09:44<31:12, 4.51s/it, loss=0.348, lr=0.000168] Steps: 17%|█▋ | 85/500 [09:44<31:12, 4.51s/it, loss=0.587, lr=0.00017] Steps: 17%|█▋ | 86/500 [09:46<25:41, 3.72s/it, loss=0.587, lr=0.00017] Steps: 17%|█▋ | 86/500 [09:46<25:41, 3.72s/it, loss=0.626, lr=0.000172] Steps: 17%|█▋ | 87/500 [09:48<21:50, 3.17s/it, loss=0.626, lr=0.000172] Steps: 17%|█▋ | 87/500 [09:48<21:50, 3.17s/it, loss=0.827, lr=0.000174] Steps: 18%|█▊ | 88/500 [09:56<31:01, 4.52s/it, loss=0.827, lr=0.000174] Steps: 18%|█▊ | 88/500 [09:56<31:01, 4.52s/it, loss=0.877, lr=0.000176] Steps: 18%|█▊ | 89/500 [09:57<25:31, 3.73s/it, loss=0.877, lr=0.000176] Steps: 18%|█▊ | 89/500 [09:58<25:31, 3.73s/it, loss=0.614, lr=0.000178] Steps: 18%|█▊ | 90/500 [09:59<21:41, 3.17s/it, loss=0.614, lr=0.000178] Steps: 18%|█▊ | 90/500 [09:59<21:41, 3.17s/it, loss=0.416, lr=0.00018] Steps: 18%|█▊ | 91/500 [10:07<30:52, 4.53s/it, loss=0.416, lr=0.00018] Steps: 18%|█▊ | 91/500 [10:07<30:52, 4.53s/it, loss=0.654, lr=0.000182] Steps: 18%|█▊ | 92/500 [10:09<25:23, 3.73s/it, loss=0.654, lr=0.000182] Steps: 18%|█▊ | 92/500 [10:09<25:23, 3.73s/it, loss=0.689, lr=0.000184] Steps: 19%|█▊ | 93/500 [10:11<21:33, 3.18s/it, loss=0.689, lr=0.000184] Steps: 19%|█▊ | 93/500 [10:11<21:33, 3.18s/it, loss=0.295, lr=0.000186] Steps: 19%|█▉ | 94/500 [10:19<30:42, 4.54s/it, loss=0.295, lr=0.000186] Steps: 19%|█▉ | 94/500 [10:19<30:42, 4.54s/it, loss=0.975, lr=0.000188] Steps: 19%|█▉ | 95/500 [10:20<25:15, 3.74s/it, loss=0.975, lr=0.000188] Steps: 19%|█▉ | 95/500 [10:20<25:15, 3.74s/it, loss=0.585, lr=0.00019] Steps: 19%|█▉ | 96/500 [10:22<21:26, 3.18s/it, loss=0.585, lr=0.00019] Steps: 19%|█▉ | 96/500 [10:22<21:26, 3.18s/it, loss=1.01, lr=0.000192] Steps: 19%|█▉ | 97/500 [10:30<30:34, 4.55s/it, loss=1.01, lr=0.000192] Steps: 19%|█▉ | 97/500 [10:30<30:34, 4.55s/it, loss=0.605, lr=0.000194] Steps: 20%|█▉ | 98/500 [10:32<25:08, 3.75s/it, loss=0.605, lr=0.000194] Steps: 20%|█▉ | 98/500 [10:32<25:08, 3.75s/it, loss=0.681, lr=0.000196] Steps: 20%|█▉ | 99/500 [10:34<21:18, 3.19s/it, loss=0.681, lr=0.000196] Steps: 20%|█▉ | 99/500 [10:34<21:18, 3.19s/it, loss=0.431, lr=0.000198] Steps: 20%|██ | 100/500 [10:42<30:25, 4.56s/it, loss=0.431, lr=0.000198] Steps: 20%|██ | 100/500 [10:42<30:25, 4.56s/it, loss=0.549, lr=0.0002] Steps: 20%|██ | 101/500 [10:43<24:59, 3.76s/it, loss=0.549, lr=0.0002] Steps: 20%|██ | 101/500 [10:43<24:59, 3.76s/it, loss=0.541, lr=0.000202] Steps: 20%|██ | 102/500 [10:45<21:12, 3.20s/it, loss=0.541, lr=0.000202] Steps: 20%|██ | 102/500 [10:45<21:12, 3.20s/it, loss=0.434, lr=0.000204] Steps: 21%|██ | 103/500 [10:53<30:07, 4.55s/it, loss=0.434, lr=0.000204] Steps: 21%|██ | 103/500 [10:53<30:07, 4.55s/it, loss=0.813, lr=0.000206] Steps: 21%|██ | 104/500 [10:55<24:46, 3.75s/it, loss=0.813, lr=0.000206] Steps: 21%|██ | 104/500 [10:55<24:46, 3.75s/it, loss=0.48, lr=0.000208] Steps: 21%|██ | 105/500 [10:57<21:00, 3.19s/it, loss=0.48, lr=0.000208] Steps: 21%|██ | 105/500 [10:57<21:00, 3.19s/it, loss=0.568, lr=0.00021] Steps: 21%|██ | 106/500 [11:04<29:45, 4.53s/it, loss=0.568, lr=0.00021] Steps: 21%|██ | 106/500 [11:05<29:45, 4.53s/it, loss=0.535, lr=0.000212] Steps: 21%|██▏ | 107/500 [11:06<24:28, 3.74s/it, loss=0.535, lr=0.000212] Steps: 21%|██▏ | 107/500 [11:06<24:28, 3.74s/it, loss=1.08, lr=0.000214] Steps: 22%|██▏ | 108/500 [11:08<20:46, 3.18s/it, loss=1.08, lr=0.000214] Steps: 22%|██▏ | 108/500 [11:08<20:46, 3.18s/it, loss=0.41, lr=0.000216] Steps: 22%|██▏ | 109/500 [11:16<29:24, 4.51s/it, loss=0.41, lr=0.000216] Steps: 22%|██▏ | 109/500 [11:16<29:24, 4.51s/it, loss=0.535, lr=0.000218] Steps: 22%|██▏ | 110/500 [11:18<24:12, 3.72s/it, loss=0.535, lr=0.000218] Steps: 22%|██▏ | 110/500 [11:18<24:12, 3.72s/it, loss=0.543, lr=0.00022] Steps: 22%|██▏ | 111/500 [11:20<20:34, 3.17s/it, loss=0.543, lr=0.00022] Steps: 22%|██▏ | 111/500 [11:20<20:34, 3.17s/it, loss=0.978, lr=0.000222] Steps: 22%|██▏ | 112/500 [11:27<29:19, 4.53s/it, loss=0.978, lr=0.000222] Steps: 22%|██▏ | 112/500 [11:27<29:19, 4.53s/it, loss=0.569, lr=0.000224] Steps: 23%|██▎ | 113/500 [11:29<24:07, 3.74s/it, loss=0.569, lr=0.000224] Steps: 23%|██▎ | 113/500 [11:29<24:07, 3.74s/it, loss=0.497, lr=0.000226] Steps: 23%|██▎ | 114/500 [11:31<20:28, 3.18s/it, loss=0.497, lr=0.000226] Steps: 23%|██▎ | 114/500 [11:31<20:28, 3.18s/it, loss=0.59, lr=0.000228] Steps: 23%|██▎ | 115/500 [11:39<29:06, 4.54s/it, loss=0.59, lr=0.000228] Steps: 23%|██▎ | 115/500 [11:39<29:06, 4.54s/it, loss=0.532, lr=0.00023] Steps: 23%|██▎ | 116/500 [11:41<23:56, 3.74s/it, loss=0.532, lr=0.00023] Steps: 23%|██▎ | 116/500 [11:41<23:56, 3.74s/it, loss=0.543, lr=0.000232] Steps: 23%|██▎ | 117/500 [11:43<20:18, 3.18s/it, loss=0.543, lr=0.000232] Steps: 23%|██▎ | 117/500 [11:43<20:18, 3.18s/it, loss=0.781, lr=0.000234] Steps: 24%|██▎ | 118/500 [11:50<28:49, 4.53s/it, loss=0.781, lr=0.000234] Steps: 24%|██▎ | 118/500 [11:50<28:49, 4.53s/it, loss=0.533, lr=0.000236] Steps: 24%|██▍ | 119/500 [11:52<23:42, 3.73s/it, loss=0.533, lr=0.000236] Steps: 24%|██▍ | 119/500 [11:52<23:42, 3.73s/it, loss=0.45, lr=0.000238] Steps: 24%|██▍ | 120/500 [11:54<20:07, 3.18s/it, loss=0.45, lr=0.000238] Steps: 24%|██▍ | 120/500 [11:54<20:07, 3.18s/it, loss=0.285, lr=0.00024] Steps: 24%|██▍ | 121/500 [12:02<28:36, 4.53s/it, loss=0.285, lr=0.00024] Steps: 24%|██▍ | 121/500 [12:02<28:36, 4.53s/it, loss=0.544, lr=0.000242] Steps: 24%|██▍ | 122/500 [12:04<23:31, 3.74s/it, loss=0.544, lr=0.000242] Steps: 24%|██▍ | 122/500 [12:04<23:31, 3.74s/it, loss=1.03, lr=0.000244] Steps: 25%|██▍ | 123/500 [12:05<19:59, 3.18s/it, loss=1.03, lr=0.000244] Steps: 25%|██▍ | 123/500 [12:05<19:59, 3.18s/it, loss=0.448, lr=0.000246] Steps: 25%|██▍ | 124/500 [12:13<28:26, 4.54s/it, loss=0.448, lr=0.000246] Steps: 25%|██▍ | 124/500 [12:13<28:26, 4.54s/it, loss=0.662, lr=0.000248] Steps: 25%|██▌ | 125/500 [12:15<23:22, 3.74s/it, loss=0.662, lr=0.000248] Steps: 25%|██▌ | 125/500 [12:15<23:22, 3.74s/it, loss=0.6, lr=0.00025] Steps: 25%|██▌ | 126/500 [12:17<19:51, 3.19s/it, loss=0.6, lr=0.00025] Steps: 25%|██▌ | 126/500 [12:17<19:51, 3.19s/it, loss=0.291, lr=0.000252] Steps: 25%|██▌ | 127/500 [12:25<28:19, 4.56s/it, loss=0.291, lr=0.000252] Steps: 25%|██▌ | 127/500 [12:25<28:19, 4.56s/it, loss=0.999, lr=0.000254] Steps: 26%|██▌ | 128/500 [12:27<23:17, 3.76s/it, loss=0.999, lr=0.000254] Steps: 26%|██▌ | 128/500 [12:27<23:17, 3.76s/it, loss=0.446, lr=0.000256] Steps: 26%|██▌ | 129/500 [12:28<19:45, 3.19s/it, loss=0.446, lr=0.000256] Steps: 26%|██▌ | 129/500 [12:28<19:45, 3.19s/it, loss=0.297, lr=0.000258] Steps: 26%|██▌ | 130/500 [12:36<27:53, 4.52s/it, loss=0.297, lr=0.000258] Steps: 26%|██▌ | 130/500 [12:36<27:53, 4.52s/it, loss=0.8, lr=0.00026] Steps: 26%|██▌ | 131/500 [12:38<22:57, 3.73s/it, loss=0.8, lr=0.00026] Steps: 26%|██▌ | 131/500 [12:38<22:57, 3.73s/it, loss=0.899, lr=0.000262] Steps: 26%|██▋ | 132/500 [12:40<19:29, 3.18s/it, loss=0.899, lr=0.000262] Steps: 26%|██▋ | 132/500 [12:40<19:29, 3.18s/it, loss=0.369, lr=0.000264] Steps: 27%|██▋ | 133/500 [12:48<27:38, 4.52s/it, loss=0.369, lr=0.000264] Steps: 27%|██▋ | 133/500 [12:48<27:38, 4.52s/it, loss=0.545, lr=0.000266] Steps: 27%|██▋ | 134/500 [12:49<22:45, 3.73s/it, loss=0.545, lr=0.000266] Steps: 27%|██▋ | 134/500 [12:49<22:45, 3.73s/it, loss=0.45, lr=0.000268] Steps: 27%|██▋ | 135/500 [12:51<19:19, 3.18s/it, loss=0.45, lr=0.000268] Steps: 27%|██▋ | 135/500 [12:51<19:19, 3.18s/it, loss=0.634, lr=0.00027] Steps: 27%|██▋ | 136/500 [12:59<27:35, 4.55s/it, loss=0.634, lr=0.00027] Steps: 27%|██▋ | 136/500 [12:59<27:35, 4.55s/it, loss=0.848, lr=0.000272] Steps: 27%|██▋ | 137/500 [13:01<22:40, 3.75s/it, loss=0.848, lr=0.000272] Steps: 27%|██▋ | 137/500 [13:01<22:40, 3.75s/it, loss=0.462, lr=0.000274] Steps: 28%|██▊ | 138/500 [13:03<19:14, 3.19s/it, loss=0.462, lr=0.000274] Steps: 28%|██▊ | 138/500 [13:03<19:14, 3.19s/it, loss=0.277, lr=0.000276] Steps: 28%|██▊ | 139/500 [13:10<26:59, 4.49s/it, loss=0.277, lr=0.000276] Steps: 28%|██▊ | 139/500 [13:10<26:59, 4.49s/it, loss=0.874, lr=0.000278] Steps: 28%|██▊ | 140/500 [13:12<22:13, 3.71s/it, loss=0.874, lr=0.000278] Steps: 28%|██▊ | 140/500 [13:12<22:13, 3.71s/it, loss=0.527, lr=0.00028] Steps: 28%|██▊ | 141/500 [13:14<18:53, 3.16s/it, loss=0.527, lr=0.00028] Steps: 28%|██▊ | 141/500 [13:14<18:53, 3.16s/it, loss=1, lr=0.000282] Steps: 28%|██▊ | 142/500 [13:22<26:42, 4.48s/it, loss=1, lr=0.000282] Steps: 28%|██▊ | 142/500 [13:22<26:42, 4.48s/it, loss=0.889, lr=0.000284] Steps: 29%|██▊ | 143/500 [13:24<22:00, 3.70s/it, loss=0.889, lr=0.000284] Steps: 29%|██▊ | 143/500 [13:24<22:00, 3.70s/it, loss=0.439, lr=0.000286] Steps: 29%|██▉ | 144/500 [13:25<18:43, 3.15s/it, loss=0.439, lr=0.000286] Steps: 29%|██▉ | 144/500 [13:25<18:43, 3.15s/it, loss=0.725, lr=0.000288] Steps: 29%|██▉ | 145/500 [13:33<26:32, 4.49s/it, loss=0.725, lr=0.000288] Steps: 29%|██▉ | 145/500 [13:33<26:32, 4.49s/it, loss=0.523, lr=0.00029] Steps: 29%|██▉ | 146/500 [13:35<21:51, 3.71s/it, loss=0.523, lr=0.00029] Steps: 29%|██▉ | 146/500 [13:35<21:51, 3.71s/it, loss=0.829, lr=0.000292] Steps: 29%|██▉ | 147/500 [13:37<18:35, 3.16s/it, loss=0.829, lr=0.000292] Steps: 29%|██▉ | 147/500 [13:37<18:35, 3.16s/it, loss=0.412, lr=0.000294] Steps: 30%|██▉ | 148/500 [13:44<26:23, 4.50s/it, loss=0.412, lr=0.000294] Steps: 30%|██▉ | 148/500 [13:44<26:23, 4.50s/it, loss=0.812, lr=0.000296] Steps: 30%|██▉ | 149/500 [13:46<21:43, 3.71s/it, loss=0.812, lr=0.000296] Steps: 30%|██▉ | 149/500 [13:46<21:43, 3.71s/it, loss=0.48, lr=0.000298] Steps: 30%|███ | 150/500 [13:48<18:27, 3.17s/it, loss=0.48, lr=0.000298] Steps: 30%|███ | 150/500 [13:48<18:27, 3.17s/it, loss=0.815, lr=0.0003] Steps: 30%|███ | 151/500 [13:56<26:07, 4.49s/it, loss=0.815, lr=0.0003] Steps: 30%|███ | 151/500 [13:56<26:07, 4.49s/it, loss=0.841, lr=0.000302] Steps: 30%|███ | 152/500 [13:58<21:30, 3.71s/it, loss=0.841, lr=0.000302] Steps: 30%|███ | 152/500 [13:58<21:30, 3.71s/it, loss=0.652, lr=0.000304] Steps: 31%|███ | 153/500 [14:00<18:16, 3.16s/it, loss=0.652, lr=0.000304] Steps: 31%|███ | 153/500 [14:00<18:16, 3.16s/it, loss=0.356, lr=0.000306] Steps: 31%|███ | 154/500 [14:07<26:10, 4.54s/it, loss=0.356, lr=0.000306] Steps: 31%|███ | 154/500 [14:07<26:10, 4.54s/it, loss=0.994, lr=0.000308] Steps: 31%|███ | 155/500 [14:09<21:31, 3.74s/it, loss=0.994, lr=0.000308] Steps: 31%|███ | 155/500 [14:09<21:31, 3.74s/it, loss=0.458, lr=0.00031] Steps: 31%|███ | 156/500 [14:11<18:15, 3.18s/it, loss=0.458, lr=0.00031] Steps: 31%|███ | 156/500 [14:11<18:15, 3.18s/it, loss=1.01, lr=0.000312] Steps: 31%|███▏ | 157/500 [14:19<25:50, 4.52s/it, loss=1.01, lr=0.000312] Steps: 31%|███▏ | 157/500 [14:19<25:50, 4.52s/it, loss=0.67, lr=0.000314] Steps: 32%|███▏ | 158/500 [14:21<21:15, 3.73s/it, loss=0.67, lr=0.000314] Steps: 32%|███▏ | 158/500 [14:21<21:15, 3.73s/it, loss=0.468, lr=0.000316] Steps: 32%|███▏ | 159/500 [14:22<18:02, 3.17s/it, loss=0.468, lr=0.000316] Steps: 32%|███▏ | 159/500 [14:22<18:02, 3.17s/it, loss=0.272, lr=0.000318] Steps: 32%|███▏ | 160/500 [14:30<25:56, 4.58s/it, loss=0.272, lr=0.000318] Steps: 32%|███▏ | 160/500 [14:30<25:56, 4.58s/it, loss=0.746, lr=0.00032] Steps: 32%|███▏ | 161/500 [14:32<21:18, 3.77s/it, loss=0.746, lr=0.00032] Steps: 32%|███▏ | 161/500 [14:32<21:18, 3.77s/it, loss=0.688, lr=0.000322] Steps: 32%|███▏ | 162/500 [14:34<18:03, 3.20s/it, loss=0.688, lr=0.000322] Steps: 32%|███▏ | 162/500 [14:34<18:03, 3.20s/it, loss=0.411, lr=0.000324] Steps: 33%|███▎ | 163/500 [14:42<25:19, 4.51s/it, loss=0.411, lr=0.000324] Steps: 33%|███▎ | 163/500 [14:42<25:19, 4.51s/it, loss=0.568, lr=0.000326] Steps: 33%|███▎ | 164/500 [14:43<20:50, 3.72s/it, loss=0.568, lr=0.000326] Steps: 33%|███▎ | 164/500 [14:43<20:50, 3.72s/it, loss=0.46, lr=0.000328] Steps: 33%|███▎ | 165/500 [14:45<17:41, 3.17s/it, loss=0.46, lr=0.000328] Steps: 33%|███▎ | 165/500 [14:45<17:41, 3.17s/it, loss=0.553, lr=0.00033] Steps: 33%|███▎ | 166/500 [14:53<25:05, 4.51s/it, loss=0.553, lr=0.00033] Steps: 33%|███▎ | 166/500 [14:53<25:05, 4.51s/it, loss=0.581, lr=0.000332] Steps: 33%|███▎ | 167/500 [14:55<20:38, 3.72s/it, loss=0.581, lr=0.000332] Steps: 33%|███▎ | 167/500 [14:55<20:38, 3.72s/it, loss=0.502, lr=0.000334] Steps: 34%|███▎ | 168/500 [14:57<17:32, 3.17s/it, loss=0.502, lr=0.000334] Steps: 34%|███▎ | 168/500 [14:57<17:32, 3.17s/it, loss=0.337, lr=0.000336] Steps: 34%|███▍ | 169/500 [15:04<24:49, 4.50s/it, loss=0.337, lr=0.000336] Steps: 34%|███▍ | 169/500 [15:04<24:49, 4.50s/it, loss=0.648, lr=0.000338] Steps: 34%|███▍ | 170/500 [15:06<20:25, 3.72s/it, loss=0.648, lr=0.000338] Steps: 34%|███▍ | 170/500 [15:06<20:25, 3.72s/it, loss=0.443, lr=0.00034] Steps: 34%|███▍ | 171/500 [15:08<17:21, 3.17s/it, loss=0.443, lr=0.00034] Steps: 34%|███▍ | 171/500 [15:08<17:21, 3.17s/it, loss=0.39, lr=0.000342] Steps: 34%|███▍ | 172/500 [15:16<24:38, 4.51s/it, loss=0.39, lr=0.000342] Steps: 34%|███▍ | 172/500 [15:16<24:38, 4.51s/it, loss=0.984, lr=0.000344] Steps: 35%|███▍ | 173/500 [15:18<20:16, 3.72s/it, loss=0.984, lr=0.000344] Steps: 35%|███▍ | 173/500 [15:18<20:16, 3.72s/it, loss=0.758, lr=0.000346] Steps: 35%|███▍ | 174/500 [15:20<17:13, 3.17s/it, loss=0.758, lr=0.000346] Steps: 35%|███▍ | 174/500 [15:20<17:13, 3.17s/it, loss=0.343, lr=0.000348] Steps: 35%|███▌ | 175/500 [15:27<24:27, 4.52s/it, loss=0.343, lr=0.000348] Steps: 35%|███▌ | 175/500 [15:27<24:27, 4.52s/it, loss=0.686, lr=0.00035] Steps: 35%|███▌ | 176/500 [15:29<20:06, 3.73s/it, loss=0.686, lr=0.00035] Steps: 35%|███▌ | 176/500 [15:29<20:06, 3.73s/it, loss=1, lr=0.000352] Steps: 35%|███▌ | 177/500 [15:31<17:04, 3.17s/it, loss=1, lr=0.000352] Steps: 35%|███▌ | 177/500 [15:31<17:04, 3.17s/it, loss=0.293, lr=0.000354] Steps: 36%|███▌ | 178/500 [15:39<24:18, 4.53s/it, loss=0.293, lr=0.000354] Steps: 36%|███▌ | 178/500 [15:39<24:18, 4.53s/it, loss=0.58, lr=0.000356] Steps: 36%|███▌ | 179/500 [15:41<19:59, 3.74s/it, loss=0.58, lr=0.000356] Steps: 36%|███▌ | 179/500 [15:41<19:59, 3.74s/it, loss=0.978, lr=0.000358] Steps: 36%|███▌ | 180/500 [15:42<16:58, 3.18s/it, loss=0.978, lr=0.000358] Steps: 36%|███▌ | 180/500 [15:42<16:58, 3.18s/it, loss=0.312, lr=0.00036] Steps: 36%|███▌ | 181/500 [15:50<24:04, 4.53s/it, loss=0.312, lr=0.00036] Steps: 36%|███▌ | 181/500 [15:50<24:04, 4.53s/it, loss=0.996, lr=0.000362] Steps: 36%|███▋ | 182/500 [15:52<19:47, 3.73s/it, loss=0.996, lr=0.000362] Steps: 36%|███▋ | 182/500 [15:52<19:47, 3.73s/it, loss=0.481, lr=0.000364] Steps: 37%|███▋ | 183/500 [15:54<16:47, 3.18s/it, loss=0.481, lr=0.000364] Steps: 37%|███▋ | 183/500 [15:54<16:47, 3.18s/it, loss=0.923, lr=0.000366] Steps: 37%|███▋ | 184/500 [16:02<23:48, 4.52s/it, loss=0.923, lr=0.000366] Steps: 37%|███▋ | 184/500 [16:02<23:48, 4.52s/it, loss=0.949, lr=0.000368] Steps: 37%|███▋ | 185/500 [16:03<19:35, 3.73s/it, loss=0.949, lr=0.000368] Steps: 37%|███▋ | 185/500 [16:03<19:35, 3.73s/it, loss=0.996, lr=0.00037] Steps: 37%|███▋ | 186/500 [16:05<16:37, 3.18s/it, loss=0.996, lr=0.00037] Steps: 37%|███▋ | 186/500 [16:05<16:37, 3.18s/it, loss=1.01, lr=0.000372] Steps: 37%|███▋ | 187/500 [16:13<23:41, 4.54s/it, loss=1.01, lr=0.000372] Steps: 37%|███▋ | 187/500 [16:13<23:41, 4.54s/it, loss=0.628, lr=0.000374] Steps: 38%|███▊ | 188/500 [16:15<19:28, 3.74s/it, loss=0.628, lr=0.000374] Steps: 38%|███▊ | 188/500 [16:15<19:28, 3.74s/it, loss=0.465, lr=0.000376] Steps: 38%|███▊ | 189/500 [16:17<16:30, 3.19s/it, loss=0.465, lr=0.000376] Steps: 38%|███▊ | 189/500 [16:17<16:30, 3.19s/it, loss=0.51, lr=0.000378] Steps: 38%|███▊ | 190/500 [16:24<23:25, 4.53s/it, loss=0.51, lr=0.000378] Steps: 38%|███▊ | 190/500 [16:24<23:25, 4.53s/it, loss=0.547, lr=0.00038] Steps: 38%|███▊ | 191/500 [16:26<19:16, 3.74s/it, loss=0.547, lr=0.00038] Steps: 38%|███▊ | 191/500 [16:26<19:16, 3.74s/it, loss=0.441, lr=0.000382] Steps: 38%|███▊ | 192/500 [16:28<16:20, 3.18s/it, loss=0.441, lr=0.000382] Steps: 38%|███▊ | 192/500 [16:28<16:20, 3.18s/it, loss=0.73, lr=0.000384] Steps: 39%|███▊ | 193/500 [16:36<22:57, 4.49s/it, loss=0.73, lr=0.000384] Steps: 39%|███▊ | 193/500 [16:36<22:57, 4.49s/it, loss=0.541, lr=0.000386] Steps: 39%|███▉ | 194/500 [16:38<18:53, 3.70s/it, loss=0.541, lr=0.000386] Steps: 39%|███▉ | 194/500 [16:38<18:53, 3.70s/it, loss=0.805, lr=0.000388] Steps: 39%|███▉ | 195/500 [16:40<16:02, 3.16s/it, loss=0.805, lr=0.000388] Steps: 39%|███▉ | 195/500 [16:40<16:02, 3.16s/it, loss=0.501, lr=0.00039] Steps: 39%|███▉ | 196/500 [16:47<22:53, 4.52s/it, loss=0.501, lr=0.00039] Steps: 39%|███▉ | 196/500 [16:47<22:53, 4.52s/it, loss=0.607, lr=0.000392] Steps: 39%|███▉ | 197/500 [16:49<18:49, 3.73s/it, loss=0.607, lr=0.000392] Steps: 39%|███▉ | 197/500 [16:49<18:49, 3.73s/it, loss=1.03, lr=0.000394] Steps: 40%|███▉ | 198/500 [16:51<15:58, 3.17s/it, loss=1.03, lr=0.000394] Steps: 40%|███▉ | 198/500 [16:51<15:58, 3.17s/it, loss=0.45, lr=0.000396] Steps: 40%|███▉ | 199/500 [16:59<22:29, 4.48s/it, loss=0.45, lr=0.000396] Steps: 40%|███▉ | 199/500 [16:59<22:29, 4.48s/it, loss=0.579, lr=0.000398] Steps: 40%|████ | 200/500 [17:00<18:30, 3.70s/it, loss=0.579, lr=0.000398] Steps: 40%|████ | 200/500 [17:00<18:30, 3.70s/it, loss=0.451, lr=0.0004] Steps: 40%|████ | 201/500 [17:02<15:43, 3.16s/it, loss=0.451, lr=0.0004] Steps: 40%|████ | 201/500 [17:02<15:43, 3.16s/it, loss=0.468, lr=0.0004] Steps: 40%|████ | 202/500 [17:10<22:23, 4.51s/it, loss=0.468, lr=0.0004] Steps: 40%|████ | 202/500 [17:10<22:23, 4.51s/it, loss=1.06, lr=0.0004] Steps: 41%|████ | 203/500 [17:12<18:25, 3.72s/it, loss=1.06, lr=0.0004] Steps: 41%|████ | 203/500 [17:12<18:25, 3.72s/it, loss=0.462, lr=0.0004] Steps: 41%|████ | 204/500 [17:14<15:38, 3.17s/it, loss=0.462, lr=0.0004] Steps: 41%|████ | 204/500 [17:14<15:38, 3.17s/it, loss=0.405, lr=0.0004] Steps: 41%|████ | 205/500 [17:21<22:15, 4.53s/it, loss=0.405, lr=0.0004] Steps: 41%|████ | 205/500 [17:21<22:15, 4.53s/it, loss=0.921, lr=0.0004] Steps: 41%|████ | 206/500 [17:23<18:17, 3.73s/it, loss=0.921, lr=0.0004] Steps: 41%|████ | 206/500 [17:23<18:17, 3.73s/it, loss=0.462, lr=0.0004] Steps: 41%|████▏ | 207/500 [17:25<15:31, 3.18s/it, loss=0.462, lr=0.0004] Steps: 41%|████▏ | 207/500 [17:25<15:31, 3.18s/it, loss=0.799, lr=0.000399] Steps: 42%|████▏ | 208/500 [17:33<22:01, 4.53s/it, loss=0.799, lr=0.000399] Steps: 42%|████▏ | 208/500 [17:33<22:01, 4.53s/it, loss=1.22, lr=0.000399] Steps: 42%|████▏ | 209/500 [17:35<18:05, 3.73s/it, loss=1.22, lr=0.000399] Steps: 42%|████▏ | 209/500 [17:35<18:05, 3.73s/it, loss=0.448, lr=0.000399] Steps: 42%|████▏ | 210/500 [17:37<15:21, 3.18s/it, loss=0.448, lr=0.000399] Steps: 42%|████▏ | 210/500 [17:37<15:21, 3.18s/it, loss=0.317, lr=0.000399] Steps: 42%|████▏ | 211/500 [17:44<21:46, 4.52s/it, loss=0.317, lr=0.000399] Steps: 42%|████▏ | 211/500 [17:44<21:46, 4.52s/it, loss=0.61, lr=0.000399] Steps: 42%|████▏ | 212/500 [17:46<17:53, 3.73s/it, loss=0.61, lr=0.000399] Steps: 42%|████▏ | 212/500 [17:46<17:53, 3.73s/it, loss=0.751, lr=0.000398] Steps: 43%|████▎ | 213/500 [17:48<15:11, 3.18s/it, loss=0.751, lr=0.000398] Steps: 43%|████▎ | 213/500 [17:48<15:11, 3.18s/it, loss=0.31, lr=0.000398] Steps: 43%|████▎ | 214/500 [17:56<21:43, 4.56s/it, loss=0.31, lr=0.000398] Steps: 43%|████▎ | 214/500 [17:56<21:43, 4.56s/it, loss=0.919, lr=0.000398] Steps: 43%|████▎ | 215/500 [17:58<17:50, 3.76s/it, loss=0.919, lr=0.000398] Steps: 43%|████▎ | 215/500 [17:58<17:50, 3.76s/it, loss=0.454, lr=0.000398] Steps: 43%|████▎ | 216/500 [18:00<15:07, 3.19s/it, loss=0.454, lr=0.000398] Steps: 43%|████▎ | 216/500 [18:00<15:07, 3.19s/it, loss=0.395, lr=0.000397] Steps: 43%|████▎ | 217/500 [18:07<21:27, 4.55s/it, loss=0.395, lr=0.000397] Steps: 43%|████▎ | 217/500 [18:07<21:27, 4.55s/it, loss=0.579, lr=0.000397] Steps: 44%|████▎ | 218/500 [18:09<17:37, 3.75s/it, loss=0.579, lr=0.000397] Steps: 44%|████▎ | 218/500 [18:09<17:37, 3.75s/it, loss=0.641, lr=0.000396] Steps: 44%|████▍ | 219/500 [18:11<14:56, 3.19s/it, loss=0.641, lr=0.000396] Steps: 44%|████▍ | 219/500 [18:11<14:56, 3.19s/it, loss=0.817, lr=0.000396] Steps: 44%|████▍ | 220/500 [18:19<21:07, 4.53s/it, loss=0.817, lr=0.000396] Steps: 44%|████▍ | 220/500 [18:19<21:07, 4.53s/it, loss=0.93, lr=0.000396] Steps: 44%|████▍ | 221/500 [18:21<17:21, 3.73s/it, loss=0.93, lr=0.000396] Steps: 44%|████▍ | 221/500 [18:21<17:21, 3.73s/it, loss=0.556, lr=0.000395] Steps: 44%|████▍ | 222/500 [18:22<14:43, 3.18s/it, loss=0.556, lr=0.000395] Steps: 44%|████▍ | 222/500 [18:22<14:43, 3.18s/it, loss=0.666, lr=0.000395] Steps: 45%|████▍ | 223/500 [18:30<20:56, 4.53s/it, loss=0.666, lr=0.000395] Steps: 45%|████▍ | 223/500 [18:30<20:56, 4.53s/it, loss=0.991, lr=0.000394] Steps: 45%|████▍ | 224/500 [18:32<17:11, 3.74s/it, loss=0.991, lr=0.000394] Steps: 45%|████▍ | 224/500 [18:32<17:11, 3.74s/it, loss=0.529, lr=0.000394] Steps: 45%|████▌ | 225/500 [18:34<14:34, 3.18s/it, loss=0.529, lr=0.000394] Steps: 45%|████▌ | 225/500 [18:34<14:34, 3.18s/it, loss=0.307, lr=0.000393] Steps: 45%|████▌ | 226/500 [18:42<20:45, 4.55s/it, loss=0.307, lr=0.000393] Steps: 45%|████▌ | 226/500 [18:42<20:45, 4.55s/it, loss=0.905, lr=0.000393] Steps: 45%|████▌ | 227/500 [18:44<17:03, 3.75s/it, loss=0.905, lr=0.000393] Steps: 45%|████▌ | 227/500 [18:44<17:03, 3.75s/it, loss=1.03, lr=0.000392] Steps: 46%|████▌ | 228/500 [18:45<14:27, 3.19s/it, loss=1.03, lr=0.000392] Steps: 46%|████▌ | 228/500 [18:45<14:27, 3.19s/it, loss=0.808, lr=0.000391] Steps: 46%|████▌ | 229/500 [18:53<20:31, 4.55s/it, loss=0.808, lr=0.000391] Steps: 46%|████▌ | 229/500 [18:53<20:31, 4.55s/it, loss=0.572, lr=0.000391] Steps: 46%|████▌ | 230/500 [18:55<16:51, 3.75s/it, loss=0.572, lr=0.000391] Steps: 46%|████▌ | 230/500 [18:55<16:51, 3.75s/it, loss=0.452, lr=0.00039] Steps: 46%|████▌ | 231/500 [18:57<14:17, 3.19s/it, loss=0.452, lr=0.00039] Steps: 46%|████▌ | 231/500 [18:57<14:17, 3.19s/it, loss=0.638, lr=0.00039] Steps: 46%|████▋ | 232/500 [19:05<20:16, 4.54s/it, loss=0.638, lr=0.00039] Steps: 46%|████▋ | 232/500 [19:05<20:16, 4.54s/it, loss=0.549, lr=0.000389] Steps: 47%|████▋ | 233/500 [19:06<16:39, 3.74s/it, loss=0.549, lr=0.000389] Steps: 47%|████▋ | 233/500 [19:06<16:39, 3.74s/it, loss=0.448, lr=0.000388] Steps: 47%|████▋ | 234/500 [19:08<14:07, 3.19s/it, loss=0.448, lr=0.000388] Steps: 47%|████▋ | 234/500 [19:08<14:07, 3.19s/it, loss=0.284, lr=0.000387] Steps: 47%|████▋ | 235/500 [19:16<20:07, 4.56s/it, loss=0.284, lr=0.000387] Steps: 47%|████▋ | 235/500 [19:16<20:07, 4.56s/it, loss=0.538, lr=0.000387] Steps: 47%|████▋ | 236/500 [19:18<16:31, 3.75s/it, loss=0.538, lr=0.000387] Steps: 47%|████▋ | 236/500 [19:18<16:31, 3.75s/it, loss=0.485, lr=0.000386] Steps: 47%|████▋ | 237/500 [19:20<13:59, 3.19s/it, loss=0.485, lr=0.000386] Steps: 47%|████▋ | 237/500 [19:20<13:59, 3.19s/it, loss=0.848, lr=0.000385] Steps: 48%|████▊ | 238/500 [19:28<19:54, 4.56s/it, loss=0.848, lr=0.000385] Steps: 48%|████▊ | 238/500 [19:28<19:54, 4.56s/it, loss=1.03, lr=0.000384] Steps: 48%|████▊ | 239/500 [19:30<16:20, 3.76s/it, loss=1.03, lr=0.000384] Steps: 48%|████▊ | 239/500 [19:30<16:20, 3.76s/it, loss=0.848, lr=0.000384] Steps: 48%|████▊ | 240/500 [19:31<13:50, 3.19s/it, loss=0.848, lr=0.000384] Steps: 48%|████▊ | 240/500 [19:31<13:50, 3.19s/it, loss=0.973, lr=0.000383] Steps: 48%|████▊ | 241/500 [19:39<19:31, 4.52s/it, loss=0.973, lr=0.000383] Steps: 48%|████▊ | 241/500 [19:39<19:31, 4.52s/it, loss=0.888, lr=0.000382] Steps: 48%|████▊ | 242/500 [19:41<16:02, 3.73s/it, loss=0.888, lr=0.000382] Steps: 48%|████▊ | 242/500 [19:41<16:02, 3.73s/it, loss=0.717, lr=0.000381] Steps: 49%|████▊ | 243/500 [19:43<13:36, 3.18s/it, loss=0.717, lr=0.000381] Steps: 49%|████▊ | 243/500 [19:43<13:36, 3.18s/it, loss=0.284, lr=0.00038] Steps: 49%|████▉ | 244/500 [19:50<19:09, 4.49s/it, loss=0.284, lr=0.00038] Steps: 49%|████▉ | 244/500 [19:50<19:09, 4.49s/it, loss=0.639, lr=0.000379] Steps: 49%|████▉ | 245/500 [19:52<15:44, 3.71s/it, loss=0.639, lr=0.000379] Steps: 49%|████▉ | 245/500 [19:52<15:44, 3.71s/it, loss=0.885, lr=0.000378] Steps: 49%|████▉ | 246/500 [19:54<13:22, 3.16s/it, loss=0.885, lr=0.000378] Steps: 49%|████▉ | 246/500 [19:54<13:22, 3.16s/it, loss=0.32, lr=0.000377] Steps: 49%|████▉ | 247/500 [20:02<18:58, 4.50s/it, loss=0.32, lr=0.000377] Steps: 49%|████▉ | 247/500 [20:02<18:58, 4.50s/it, loss=0.567, lr=0.000376] Steps: 50%|████▉ | 248/500 [20:04<15:36, 3.72s/it, loss=0.567, lr=0.000376] Steps: 50%|████▉ | 248/500 [20:04<15:36, 3.72s/it, loss=0.506, lr=0.000375] Steps: 50%|████▉ | 249/500 [20:05<13:14, 3.17s/it, loss=0.506, lr=0.000375] Steps: 50%|████▉ | 249/500 [20:06<13:14, 3.17s/it, loss=0.478, lr=0.000374] Steps: 50%|█████ | 250/500 [20:13<18:52, 4.53s/it, loss=0.478, lr=0.000374] Steps: 50%|█████ | 250/500 [20:13<18:52, 4.53s/it, loss=0.97, lr=0.000373] Steps: 50%|█████ | 251/500 [20:15<15:30, 3.74s/it, loss=0.97, lr=0.000373] Steps: 50%|█████ | 251/500 [20:15<15:30, 3.74s/it, loss=0.446, lr=0.000372] Steps: 50%|█████ | 252/500 [20:17<13:08, 3.18s/it, loss=0.446, lr=0.000372] Steps: 50%|█████ | 252/500 [20:17<13:08, 3.18s/it, loss=0.295, lr=0.000371] Steps: 51%|█████ | 253/500 [20:25<18:39, 4.53s/it, loss=0.295, lr=0.000371] Steps: 51%|█████ | 253/500 [20:25<18:39, 4.53s/it, loss=0.732, lr=0.00037] Steps: 51%|█████ | 254/500 [20:27<15:19, 3.74s/it, loss=0.732, lr=0.00037] Steps: 51%|█████ | 254/500 [20:27<15:19, 3.74s/it, loss=0.527, lr=0.000369] Steps: 51%|█████ | 255/500 [20:28<12:59, 3.18s/it, loss=0.527, lr=0.000369] Steps: 51%|█████ | 255/500 [20:28<12:59, 3.18s/it, loss=0.658, lr=0.000368] Steps: 51%|█████ | 256/500 [20:36<18:23, 4.52s/it, loss=0.658, lr=0.000368] Steps: 51%|█████ | 256/500 [20:36<18:23, 4.52s/it, loss=0.673, lr=0.000367] Steps: 51%|█████▏ | 257/500 [20:38<15:05, 3.73s/it, loss=0.673, lr=0.000367] Steps: 51%|█████▏ | 257/500 [20:38<15:05, 3.73s/it, loss=0.984, lr=0.000365] Steps: 52%|█████▏ | 258/500 [20:40<12:48, 3.18s/it, loss=0.984, lr=0.000365] Steps: 52%|█████▏ | 258/500 [20:40<12:48, 3.18s/it, loss=0.774, lr=0.000364] Steps: 52%|█████▏ | 259/500 [20:47<18:07, 4.51s/it, loss=0.774, lr=0.000364] Steps: 52%|█████▏ | 259/500 [20:47<18:07, 4.51s/it, loss=0.542, lr=0.000363] Steps: 52%|█████▏ | 260/500 [20:49<14:53, 3.72s/it, loss=0.542, lr=0.000363] Steps: 52%|█████▏ | 260/500 [20:49<14:53, 3.72s/it, loss=1.01, lr=0.000362] Steps: 52%|█████▏ | 261/500 [20:51<12:37, 3.17s/it, loss=1.01, lr=0.000362] Steps: 52%|█████▏ | 261/500 [20:51<12:37, 3.17s/it, loss=0.554, lr=0.000361] Steps: 52%|█████▏ | 262/500 [20:59<17:55, 4.52s/it, loss=0.554, lr=0.000361] Steps: 52%|█████▏ | 262/500 [20:59<17:55, 4.52s/it, loss=0.589, lr=0.000359] Steps: 53%|█████▎ | 263/500 [21:01<14:43, 3.73s/it, loss=0.589, lr=0.000359] Steps: 53%|█████▎ | 263/500 [21:01<14:43, 3.73s/it, loss=0.699, lr=0.000358] Steps: 53%|█████▎ | 264/500 [21:03<12:29, 3.18s/it, loss=0.699, lr=0.000358] Steps: 53%|█████▎ | 264/500 [21:03<12:29, 3.18s/it, loss=0.283, lr=0.000357] Steps: 53%|█████▎ | 265/500 [21:10<17:44, 4.53s/it, loss=0.283, lr=0.000357] Steps: 53%|█████▎ | 265/500 [21:10<17:44, 4.53s/it, loss=0.661, lr=0.000355] Steps: 53%|█████▎ | 266/500 [21:12<14:34, 3.74s/it, loss=0.661, lr=0.000355] Steps: 53%|█████▎ | 266/500 [21:12<14:34, 3.74s/it, loss=0.492, lr=0.000354] Steps: 53%|█████▎ | 267/500 [21:14<12:20, 3.18s/it, loss=0.492, lr=0.000354] Steps: 53%|█████▎ | 267/500 [21:14<12:20, 3.18s/it, loss=0.592, lr=0.000353] Steps: 54%|█████▎ | 268/500 [21:22<17:23, 4.50s/it, loss=0.592, lr=0.000353] Steps: 54%|█████▎ | 268/500 [21:22<17:23, 4.50s/it, loss=0.575, lr=0.000351] Steps: 54%|█████▍ | 269/500 [21:24<14:17, 3.71s/it, loss=0.575, lr=0.000351] Steps: 54%|█████▍ | 269/500 [21:24<14:17, 3.71s/it, loss=0.493, lr=0.00035] Steps: 54%|█████▍ | 270/500 [21:25<12:07, 3.16s/it, loss=0.493, lr=0.00035] Steps: 54%|█████▍ | 270/500 [21:25<12:07, 3.16s/it, loss=0.783, lr=0.000349] Steps: 54%|█████▍ | 271/500 [21:33<17:13, 4.51s/it, loss=0.783, lr=0.000349] Steps: 54%|█████▍ | 271/500 [21:33<17:13, 4.51s/it, loss=0.56, lr=0.000347] Steps: 54%|█████▍ | 272/500 [21:35<14:08, 3.72s/it, loss=0.56, lr=0.000347] Steps: 54%|█████▍ | 272/500 [21:35<14:08, 3.72s/it, loss=0.823, lr=0.000346] Steps: 55%|█████▍ | 273/500 [21:37<11:59, 3.17s/it, loss=0.823, lr=0.000346] Steps: 55%|█████▍ | 273/500 [21:37<11:59, 3.17s/it, loss=0.536, lr=0.000344] Steps: 55%|█████▍ | 274/500 [21:44<16:59, 4.51s/it, loss=0.536, lr=0.000344] Steps: 55%|█████▍ | 274/500 [21:45<16:59, 4.51s/it, loss=0.589, lr=0.000343] Steps: 55%|█████▌ | 275/500 [21:46<13:57, 3.72s/it, loss=0.589, lr=0.000343] Steps: 55%|█████▌ | 275/500 [21:46<13:57, 3.72s/it, loss=0.437, lr=0.000341] Steps: 55%|█████▌ | 276/500 [21:48<11:50, 3.17s/it, loss=0.437, lr=0.000341] Steps: 55%|█████▌ | 276/500 [21:48<11:50, 3.17s/it, loss=0.693, lr=0.00034] Steps: 55%|█████▌ | 277/500 [21:56<16:53, 4.55s/it, loss=0.693, lr=0.00034] Steps: 55%|█████▌ | 277/500 [21:56<16:53, 4.55s/it, loss=0.601, lr=0.000338] Steps: 56%|█████▌ | 278/500 [21:58<13:51, 3.75s/it, loss=0.601, lr=0.000338] Steps: 56%|█████▌ | 278/500 [21:58<13:51, 3.75s/it, loss=0.609, lr=0.000337] Steps: 56%|█████▌ | 279/500 [22:00<11:44, 3.19s/it, loss=0.609, lr=0.000337] Steps: 56%|█████▌ | 279/500 [22:00<11:44, 3.19s/it, loss=0.619, lr=0.000335] Steps: 56%|█████▌ | 280/500 [22:07<16:36, 4.53s/it, loss=0.619, lr=0.000335] Steps: 56%|█████▌ | 280/500 [22:07<16:36, 4.53s/it, loss=0.653, lr=0.000334] Steps: 56%|█████▌ | 281/500 [22:09<13:37, 3.73s/it, loss=0.653, lr=0.000334] Steps: 56%|█████▌ | 281/500 [22:09<13:37, 3.73s/it, loss=0.946, lr=0.000332] Steps: 56%|█████▋ | 282/500 [22:11<11:32, 3.18s/it, loss=0.946, lr=0.000332] Steps: 56%|█████▋ | 282/500 [22:11<11:32, 3.18s/it, loss=0.272, lr=0.000331] Steps: 57%|█████▋ | 283/500 [22:19<16:26, 4.55s/it, loss=0.272, lr=0.000331] Steps: 57%|█████▋ | 283/500 [22:19<16:26, 4.55s/it, loss=0.567, lr=0.000329] Steps: 57%|█████▋ | 284/500 [22:21<13:29, 3.75s/it, loss=0.567, lr=0.000329] Steps: 57%|█████▋ | 284/500 [22:21<13:29, 3.75s/it, loss=0.45, lr=0.000327] Steps: 57%|█████▋ | 285/500 [22:23<11:25, 3.19s/it, loss=0.45, lr=0.000327] Steps: 57%|█████▋ | 285/500 [22:23<11:25, 3.19s/it, loss=0.299, lr=0.000326] Steps: 57%|█████▋ | 286/500 [22:30<16:08, 4.53s/it, loss=0.299, lr=0.000326] Steps: 57%|█████▋ | 286/500 [22:30<16:08, 4.53s/it, loss=0.687, lr=0.000324] Steps: 57%|█████▋ | 287/500 [22:32<13:15, 3.73s/it, loss=0.687, lr=0.000324] Steps: 57%|█████▋ | 287/500 [22:32<13:15, 3.73s/it, loss=0.457, lr=0.000323] Steps: 58%|█████▊ | 288/500 [22:34<11:13, 3.18s/it, loss=0.457, lr=0.000323] Steps: 58%|█████▊ | 288/500 [22:34<11:13, 3.18s/it, loss=0.333, lr=0.000321] Steps: 58%|█████▊ | 289/500 [22:42<15:55, 4.53s/it, loss=0.333, lr=0.000321] Steps: 58%|█████▊ | 289/500 [22:42<15:55, 4.53s/it, loss=0.936, lr=0.000319] Steps: 58%|█████▊ | 290/500 [22:44<13:04, 3.74s/it, loss=0.936, lr=0.000319] Steps: 58%|█████▊ | 290/500 [22:44<13:04, 3.74s/it, loss=0.6, lr=0.000318] Steps: 58%|█████▊ | 291/500 [22:46<11:04, 3.18s/it, loss=0.6, lr=0.000318] Steps: 58%|█████▊ | 291/500 [22:46<11:04, 3.18s/it, loss=0.434, lr=0.000316] Steps: 58%|█████▊ | 292/500 [22:53<15:44, 4.54s/it, loss=0.434, lr=0.000316] Steps: 58%|█████▊ | 292/500 [22:53<15:44, 4.54s/it, loss=0.562, lr=0.000314] Steps: 59%|█████▊ | 293/500 [22:55<12:54, 3.74s/it, loss=0.562, lr=0.000314] Steps: 59%|█████▊ | 293/500 [22:55<12:54, 3.74s/it, loss=0.462, lr=0.000312] Steps: 59%|█████▉ | 294/500 [22:57<10:56, 3.19s/it, loss=0.462, lr=0.000312] Steps: 59%|█████▉ | 294/500 [22:57<10:56, 3.19s/it, loss=0.667, lr=0.000311] Steps: 59%|█████▉ | 295/500 [23:05<15:33, 4.55s/it, loss=0.667, lr=0.000311] Steps: 59%|█████▉ | 295/500 [23:05<15:33, 4.55s/it, loss=0.514, lr=0.000309] Steps: 59%|█████▉ | 296/500 [23:07<12:45, 3.75s/it, loss=0.514, lr=0.000309] Steps: 59%|█████▉ | 296/500 [23:07<12:45, 3.75s/it, loss=1.03, lr=0.000307] Steps: 59%|█████▉ | 297/500 [23:09<10:47, 3.19s/it, loss=1.03, lr=0.000307] Steps: 59%|█████▉ | 297/500 [23:09<10:47, 3.19s/it, loss=0.667, lr=0.000305] Steps: 60%|█████▉ | 298/500 [23:16<15:13, 4.52s/it, loss=0.667, lr=0.000305] Steps: 60%|█████▉ | 298/500 [23:16<15:13, 4.52s/it, loss=0.792, lr=0.000304] Steps: 60%|█████▉ | 299/500 [23:18<12:29, 3.73s/it, loss=0.792, lr=0.000304] Steps: 60%|█████▉ | 299/500 [23:18<12:29, 3.73s/it, loss=0.582, lr=0.000302] Steps: 60%|██████ | 300/500 [23:20<10:35, 3.18s/it, loss=0.582, lr=0.000302] Steps: 60%|██████ | 300/500 [23:20<10:35, 3.18s/it, loss=0.754, lr=0.0003] Steps: 60%|██████ | 301/500 [23:28<15:09, 4.57s/it, loss=0.754, lr=0.0003] Steps: 60%|██████ | 301/500 [23:28<15:09, 4.57s/it, loss=0.508, lr=0.000298] Steps: 60%|██████ | 302/500 [23:30<12:25, 3.76s/it, loss=0.508, lr=0.000298] Steps: 60%|██████ | 302/500 [23:30<12:25, 3.76s/it, loss=0.472, lr=0.000296] Steps: 61%|██████ | 303/500 [23:32<10:30, 3.20s/it, loss=0.472, lr=0.000296] Steps: 61%|██████ | 303/500 [23:32<10:30, 3.20s/it, loss=0.28, lr=0.000295] Steps: 61%|██████ | 304/500 [23:39<14:54, 4.57s/it, loss=0.28, lr=0.000295] Steps: 61%|██████ | 304/500 [23:39<14:54, 4.57s/it, loss=0.529, lr=0.000293] Steps: 61%|██████ | 305/500 [23:41<12:13, 3.76s/it, loss=0.529, lr=0.000293] Steps: 61%|██████ | 305/500 [23:41<12:13, 3.76s/it, loss=0.757, lr=0.000291] Steps: 61%|██████ | 306/500 [23:43<10:20, 3.20s/it, loss=0.757, lr=0.000291] Steps: 61%|██████ | 306/500 [23:43<10:20, 3.20s/it, loss=0.314, lr=0.000289] Steps: 61%|██████▏ | 307/500 [23:51<14:47, 4.60s/it, loss=0.314, lr=0.000289] Steps: 61%|██████▏ | 307/500 [23:51<14:47, 4.60s/it, loss=0.878, lr=0.000287] Steps: 62%|██████▏ | 308/500 [23:53<12:06, 3.78s/it, loss=0.878, lr=0.000287] Steps: 62%|██████▏ | 308/500 [23:53<12:06, 3.78s/it, loss=0.701, lr=0.000285] Steps: 62%|██████▏ | 309/500 [23:55<10:13, 3.21s/it, loss=0.701, lr=0.000285] Steps: 62%|██████▏ | 309/500 [23:55<10:13, 3.21s/it, loss=0.363, lr=0.000283] Steps: 62%|██████▏ | 310/500 [24:02<14:25, 4.56s/it, loss=0.363, lr=0.000283] Steps: 62%|██████▏ | 310/500 [24:02<14:25, 4.56s/it, loss=1.04, lr=0.000281] Steps: 62%|██████▏ | 311/500 [24:04<11:49, 3.75s/it, loss=1.04, lr=0.000281] Steps: 62%|██████▏ | 311/500 [24:04<11:49, 3.75s/it, loss=0.431, lr=0.000279] Steps: 62%|██████▏ | 312/500 [24:06<10:00, 3.19s/it, loss=0.431, lr=0.000279] Steps: 62%|██████▏ | 312/500 [24:06<10:00, 3.19s/it, loss=0.693, lr=0.000278] Steps: 63%|██████▎ | 313/500 [24:14<14:06, 4.52s/it, loss=0.693, lr=0.000278] Steps: 63%|██████▎ | 313/500 [24:14<14:06, 4.52s/it, loss=0.872, lr=0.000276] Steps: 63%|██████▎ | 314/500 [24:16<11:34, 3.73s/it, loss=0.872, lr=0.000276] Steps: 63%|██████▎ | 314/500 [24:16<11:34, 3.73s/it, loss=0.551, lr=0.000274] Steps: 63%|██████▎ | 315/500 [24:18<09:47, 3.18s/it, loss=0.551, lr=0.000274] Steps: 63%|██████▎ | 315/500 [24:18<09:47, 3.18s/it, loss=0.497, lr=0.000272] Steps: 63%|██████▎ | 316/500 [24:25<13:52, 4.52s/it, loss=0.497, lr=0.000272] Steps: 63%|██████▎ | 316/500 [24:25<13:52, 4.52s/it, loss=0.579, lr=0.00027] Steps: 63%|██████▎ | 317/500 [24:27<11:22, 3.73s/it, loss=0.579, lr=0.00027] Steps: 63%|██████▎ | 317/500 [24:27<11:22, 3.73s/it, loss=0.668, lr=0.000268] Steps: 64%|██████▎ | 318/500 [24:29<09:38, 3.18s/it, loss=0.668, lr=0.000268] Steps: 64%|██████▎ | 318/500 [24:29<09:38, 3.18s/it, loss=0.846, lr=0.000266] Steps: 64%|██████▍ | 319/500 [24:37<13:41, 4.54s/it, loss=0.846, lr=0.000266] Steps: 64%|██████▍ | 319/500 [24:37<13:41, 4.54s/it, loss=1.03, lr=0.000264] Steps: 64%|██████▍ | 320/500 [24:39<11:13, 3.74s/it, loss=1.03, lr=0.000264] Steps: 64%|██████▍ | 320/500 [24:39<11:13, 3.74s/it, loss=1.01, lr=0.000262] Steps: 64%|██████▍ | 321/500 [24:40<09:30, 3.19s/it, loss=1.01, lr=0.000262] Steps: 64%|██████▍ | 321/500 [24:40<09:30, 3.19s/it, loss=0.334, lr=0.00026] Steps: 64%|██████▍ | 322/500 [24:48<13:27, 4.53s/it, loss=0.334, lr=0.00026] Steps: 64%|██████▍ | 322/500 [24:48<13:27, 4.53s/it, loss=0.601, lr=0.000258] Steps: 65%|██████▍ | 323/500 [24:50<11:01, 3.74s/it, loss=0.601, lr=0.000258] Steps: 65%|██████▍ | 323/500 [24:50<11:01, 3.74s/it, loss=1.01, lr=0.000256] Steps: 65%|██████▍ | 324/500 [24:52<09:20, 3.18s/it, loss=1.01, lr=0.000256] Steps: 65%|██████▍ | 324/500 [24:52<09:20, 3.18s/it, loss=0.628, lr=0.000254] Steps: 65%|██████▌ | 325/500 [25:00<13:11, 4.52s/it, loss=0.628, lr=0.000254] Steps: 65%|██████▌ | 325/500 [25:00<13:11, 4.52s/it, loss=0.52, lr=0.000252] Steps: 65%|██████▌ | 326/500 [25:01<10:49, 3.73s/it, loss=0.52, lr=0.000252] Steps: 65%|██████▌ | 326/500 [25:01<10:49, 3.73s/it, loss=0.422, lr=0.00025] Steps: 65%|██████▌ | 327/500 [25:03<09:09, 3.18s/it, loss=0.422, lr=0.00025] Steps: 65%|██████▌ | 327/500 [25:03<09:09, 3.18s/it, loss=0.941, lr=0.000248] Steps: 66%|██████▌ | 328/500 [25:11<12:56, 4.51s/it, loss=0.941, lr=0.000248] Steps: 66%|██████▌ | 328/500 [25:11<12:56, 4.51s/it, loss=1.02, lr=0.000246] Steps: 66%|██████▌ | 329/500 [25:13<10:36, 3.73s/it, loss=1.02, lr=0.000246] Steps: 66%|██████▌ | 329/500 [25:13<10:36, 3.73s/it, loss=0.62, lr=0.000244] Steps: 66%|██████▌ | 330/500 [25:15<08:59, 3.17s/it, loss=0.62, lr=0.000244] Steps: 66%|██████▌ | 330/500 [25:15<08:59, 3.17s/it, loss=0.628, lr=0.000242] Steps: 66%|██████▌ | 331/500 [25:22<12:39, 4.50s/it, loss=0.628, lr=0.000242] Steps: 66%|██████▌ | 331/500 [25:22<12:39, 4.50s/it, loss=1.03, lr=0.00024] Steps: 66%|██████▋ | 332/500 [25:24<10:23, 3.71s/it, loss=1.03, lr=0.00024] Steps: 66%|██████▋ | 332/500 [25:24<10:23, 3.71s/it, loss=0.624, lr=0.000237] Steps: 67%|██████▋ | 333/500 [25:26<08:48, 3.16s/it, loss=0.624, lr=0.000237] Steps: 67%|██████▋ | 333/500 [25:26<08:48, 3.16s/it, loss=0.602, lr=0.000235] Steps: 67%|██████▋ | 334/500 [25:34<12:25, 4.49s/it, loss=0.602, lr=0.000235] Steps: 67%|██████▋ | 334/500 [25:34<12:25, 4.49s/it, loss=0.792, lr=0.000233] Steps: 67%|██████▋ | 335/500 [25:36<10:12, 3.71s/it, loss=0.792, lr=0.000233] Steps: 67%|██████▋ | 335/500 [25:36<10:12, 3.71s/it, loss=0.55, lr=0.000231] Steps: 67%|██████▋ | 336/500 [25:37<08:38, 3.16s/it, loss=0.55, lr=0.000231] Steps: 67%|██████▋ | 336/500 [25:37<08:38, 3.16s/it, loss=0.265, lr=0.000229] Steps: 67%|██████▋ | 337/500 [25:45<12:14, 4.51s/it, loss=0.265, lr=0.000229] Steps: 67%|██████▋ | 337/500 [25:45<12:14, 4.51s/it, loss=0.538, lr=0.000227] Steps: 68%|██████▊ | 338/500 [25:47<10:02, 3.72s/it, loss=0.538, lr=0.000227] Steps: 68%|██████▊ | 338/500 [25:47<10:02, 3.72s/it, loss=0.478, lr=0.000225] Steps: 68%|██████▊ | 339/500 [25:49<08:30, 3.17s/it, loss=0.478, lr=0.000225] Steps: 68%|██████▊ | 339/500 [25:49<08:30, 3.17s/it, loss=0.347, lr=0.000223] Steps: 68%|██████▊ | 340/500 [25:57<12:03, 4.52s/it, loss=0.347, lr=0.000223] Steps: 68%|██████▊ | 340/500 [25:57<12:03, 4.52s/it, loss=0.502, lr=0.000221] Steps: 68%|██████▊ | 341/500 [25:58<09:52, 3.73s/it, loss=0.502, lr=0.000221] Steps: 68%|██████▊ | 341/500 [25:58<09:52, 3.73s/it, loss=0.999, lr=0.000219] Steps: 68%|██████▊ | 342/500 [26:00<08:21, 3.17s/it, loss=0.999, lr=0.000219] Steps: 68%|██████▊ | 342/500 [26:00<08:21, 3.17s/it, loss=0.995, lr=0.000217] Steps: 69%|██████▊ | 343/500 [26:08<11:49, 4.52s/it, loss=0.995, lr=0.000217] Steps: 69%|██████▊ | 343/500 [26:08<11:49, 4.52s/it, loss=0.577, lr=0.000215] Steps: 69%|██████▉ | 344/500 [26:10<09:41, 3.73s/it, loss=0.577, lr=0.000215] Steps: 69%|██████▉ | 344/500 [26:10<09:41, 3.73s/it, loss=0.421, lr=0.000213] Steps: 69%|██████▉ | 345/500 [26:12<08:12, 3.18s/it, loss=0.421, lr=0.000213] Steps: 69%|██████▉ | 345/500 [26:12<08:12, 3.18s/it, loss=0.325, lr=0.00021] Steps: 69%|██████▉ | 346/500 [26:19<11:36, 4.52s/it, loss=0.325, lr=0.00021] Steps: 69%|██████▉ | 346/500 [26:19<11:36, 4.52s/it, loss=0.504, lr=0.000208] Steps: 69%|██████▉ | 347/500 [26:21<09:30, 3.73s/it, loss=0.504, lr=0.000208] Steps: 69%|██████▉ | 347/500 [26:21<09:30, 3.73s/it, loss=0.544, lr=0.000206] Steps: 70%|██████▉ | 348/500 [26:23<08:02, 3.18s/it, loss=0.544, lr=0.000206] Steps: 70%|██████▉ | 348/500 [26:23<08:02, 3.18s/it, loss=0.278, lr=0.000204] Steps: 70%|██████▉ | 349/500 [26:31<11:33, 4.59s/it, loss=0.278, lr=0.000204] Steps: 70%|██████▉ | 349/500 [26:31<11:33, 4.59s/it, loss=0.871, lr=0.000202] Steps: 70%|███████ | 350/500 [26:33<09:26, 3.78s/it, loss=0.871, lr=0.000202] Steps: 70%|███████ | 350/500 [26:33<09:26, 3.78s/it, loss=0.423, lr=0.0002] Steps: 70%|███████ | 351/500 [26:35<07:58, 3.21s/it, loss=0.423, lr=0.0002] Steps: 70%|███████ | 351/500 [26:35<07:58, 3.21s/it, loss=0.278, lr=0.000198] Steps: 70%|███████ | 352/500 [26:43<11:15, 4.57s/it, loss=0.278, lr=0.000198] Steps: 70%|███████ | 352/500 [26:43<11:15, 4.57s/it, loss=0.651, lr=0.000196] Steps: 71%|███████ | 353/500 [26:44<09:13, 3.76s/it, loss=0.651, lr=0.000196] Steps: 71%|███████ | 353/500 [26:44<09:13, 3.76s/it, loss=1.03, lr=0.000194] Steps: 71%|███████ | 354/500 [26:46<07:47, 3.20s/it, loss=1.03, lr=0.000194] Steps: 71%|███████ | 354/500 [26:46<07:47, 3.20s/it, loss=0.269, lr=0.000192] Steps: 71%|███████ | 355/500 [26:54<10:56, 4.53s/it, loss=0.269, lr=0.000192] Steps: 71%|███████ | 355/500 [26:54<10:56, 4.53s/it, loss=0.53, lr=0.00019] Steps: 71%|███████ | 356/500 [26:56<08:57, 3.73s/it, loss=0.53, lr=0.00019] Steps: 71%|███████ | 356/500 [26:56<08:57, 3.73s/it, loss=0.422, lr=0.000187] Steps: 71%|███████▏ | 357/500 [26:58<07:34, 3.18s/it, loss=0.422, lr=0.000187] Steps: 71%|███████▏ | 357/500 [26:58<07:34, 3.18s/it, loss=0.36, lr=0.000185] Steps: 72%|███████▏ | 358/500 [27:05<10:45, 4.54s/it, loss=0.36, lr=0.000185] Steps: 72%|███████▏ | 358/500 [27:05<10:45, 4.54s/it, loss=0.574, lr=0.000183] Steps: 72%|███████▏ | 359/500 [27:07<08:47, 3.74s/it, loss=0.574, lr=0.000183] Steps: 72%|███████▏ | 359/500 [27:07<08:47, 3.74s/it, loss=0.838, lr=0.000181] Steps: 72%|███████▏ | 360/500 [27:09<07:25, 3.19s/it, loss=0.838, lr=0.000181] Steps: 72%|███████▏ | 360/500 [27:09<07:25, 3.19s/it, loss=0.261, lr=0.000179] Steps: 72%|███████▏ | 361/500 [27:17<10:29, 4.53s/it, loss=0.261, lr=0.000179] Steps: 72%|███████▏ | 361/500 [27:17<10:29, 4.53s/it, loss=1.13, lr=0.000177] Steps: 72%|███████▏ | 362/500 [27:19<08:35, 3.73s/it, loss=1.13, lr=0.000177] Steps: 72%|███████▏ | 362/500 [27:19<08:35, 3.73s/it, loss=0.874, lr=0.000175] Steps: 73%|███████▎ | 363/500 [27:21<07:15, 3.18s/it, loss=0.874, lr=0.000175] Steps: 73%|███████▎ | 363/500 [27:21<07:15, 3.18s/it, loss=0.338, lr=0.000173] Steps: 73%|███████▎ | 364/500 [27:28<10:15, 4.53s/it, loss=0.338, lr=0.000173] Steps: 73%|███████▎ | 364/500 [27:28<10:15, 4.53s/it, loss=1.06, lr=0.000171] Steps: 73%|███████▎ | 365/500 [27:30<08:24, 3.73s/it, loss=1.06, lr=0.000171] Steps: 73%|███████▎ | 365/500 [27:30<08:24, 3.73s/it, loss=0.452, lr=0.000169] Steps: 73%|███████▎ | 366/500 [27:32<07:05, 3.18s/it, loss=0.452, lr=0.000169] Steps: 73%|███████▎ | 366/500 [27:32<07:05, 3.18s/it, loss=0.264, lr=0.000167] Steps: 73%|███████▎ | 367/500 [27:40<10:05, 4.56s/it, loss=0.264, lr=0.000167] Steps: 73%|███████▎ | 367/500 [27:40<10:05, 4.56s/it, loss=0.643, lr=0.000165] Steps: 74%|███████▎ | 368/500 [27:42<08:15, 3.75s/it, loss=0.643, lr=0.000165] Steps: 74%|███████▎ | 368/500 [27:42<08:15, 3.75s/it, loss=0.488, lr=0.000163] Steps: 74%|███████▍ | 369/500 [27:44<06:57, 3.19s/it, loss=0.488, lr=0.000163] Steps: 74%|███████▍ | 369/500 [27:44<06:57, 3.19s/it, loss=0.272, lr=0.00016] Steps: 74%|███████▍ | 370/500 [27:51<09:56, 4.59s/it, loss=0.272, lr=0.00016] Steps: 74%|███████▍ | 370/500 [27:51<09:56, 4.59s/it, loss=0.5, lr=0.000158] Steps: 74%|███████▍ | 371/500 [27:53<08:07, 3.78s/it, loss=0.5, lr=0.000158] Steps: 74%|███████▍ | 371/500 [27:53<08:07, 3.78s/it, loss=0.458, lr=0.000156] Steps: 74%|███████▍ | 372/500 [27:55<06:50, 3.21s/it, loss=0.458, lr=0.000156] Steps: 74%|███████▍ | 372/500 [27:55<06:50, 3.21s/it, loss=0.399, lr=0.000154] Steps: 75%|███████▍ | 373/500 [28:03<09:35, 4.53s/it, loss=0.399, lr=0.000154] Steps: 75%|███████▍ | 373/500 [28:03<09:35, 4.53s/it, loss=0.773, lr=0.000152] Steps: 75%|███████▍ | 374/500 [28:05<07:50, 3.74s/it, loss=0.773, lr=0.000152] Steps: 75%|███████▍ | 374/500 [28:05<07:50, 3.74s/it, loss=0.631, lr=0.00015] Steps: 75%|███████▌ | 375/500 [28:07<06:37, 3.18s/it, loss=0.631, lr=0.00015] Steps: 75%|███████▌ | 375/500 [28:07<06:37, 3.18s/it, loss=0.662, lr=0.000148] Steps: 75%|███████▌ | 376/500 [28:14<09:21, 4.53s/it, loss=0.662, lr=0.000148] Steps: 75%|███████▌ | 376/500 [28:14<09:21, 4.53s/it, loss=0.506, lr=0.000146] Steps: 75%|███████▌ | 377/500 [28:16<07:39, 3.73s/it, loss=0.506, lr=0.000146] Steps: 75%|███████▌ | 377/500 [28:16<07:39, 3.73s/it, loss=0.417, lr=0.000144] Steps: 76%|███████▌ | 378/500 [28:18<06:27, 3.18s/it, loss=0.417, lr=0.000144] Steps: 76%|███████▌ | 378/500 [28:18<06:27, 3.18s/it, loss=0.884, lr=0.000142] Steps: 76%|███████▌ | 379/500 [28:26<09:08, 4.54s/it, loss=0.884, lr=0.000142] Steps: 76%|███████▌ | 379/500 [28:26<09:08, 4.54s/it, loss=0.514, lr=0.00014] Steps: 76%|███████▌ | 380/500 [28:28<07:28, 3.74s/it, loss=0.514, lr=0.00014] Steps: 76%|███████▌ | 380/500 [28:28<07:28, 3.74s/it, loss=0.419, lr=0.000138] Steps: 76%|███████▌ | 381/500 [28:29<06:18, 3.18s/it, loss=0.419, lr=0.000138] Steps: 76%|███████▌ | 381/500 [28:30<06:18, 3.18s/it, loss=0.354, lr=0.000136] Steps: 76%|███████▋ | 382/500 [28:37<08:53, 4.52s/it, loss=0.354, lr=0.000136] Steps: 76%|███████▋ | 382/500 [28:37<08:53, 4.52s/it, loss=1.03, lr=0.000134] Steps: 77%|███████▋ | 383/500 [28:39<07:16, 3.73s/it, loss=1.03, lr=0.000134] Steps: 77%|███████▋ | 383/500 [28:39<07:16, 3.73s/it, loss=0.412, lr=0.000132] Steps: 77%|███████▋ | 384/500 [28:41<06:08, 3.17s/it, loss=0.412, lr=0.000132] Steps: 77%|███████▋ | 384/500 [28:41<06:08, 3.17s/it, loss=0.284, lr=0.00013] Steps: 77%|███████▋ | 385/500 [28:49<08:40, 4.53s/it, loss=0.284, lr=0.00013] Steps: 77%|███████▋ | 385/500 [28:49<08:40, 4.53s/it, loss=0.513, lr=0.000128] Steps: 77%|███████▋ | 386/500 [28:50<07:05, 3.73s/it, loss=0.513, lr=0.000128] Steps: 77%|███████▋ | 386/500 [28:50<07:05, 3.73s/it, loss=0.705, lr=0.000126] Steps: 77%|███████▋ | 387/500 [28:52<05:59, 3.18s/it, loss=0.705, lr=0.000126] Steps: 77%|███████▋ | 387/500 [28:52<05:59, 3.18s/it, loss=0.31, lr=0.000124] Steps: 78%|███████▊ | 388/500 [29:00<08:26, 4.53s/it, loss=0.31, lr=0.000124] Steps: 78%|███████▊ | 388/500 [29:00<08:26, 4.53s/it, loss=0.572, lr=0.000122] Steps: 78%|███████▊ | 389/500 [29:02<06:54, 3.73s/it, loss=0.572, lr=0.000122] Steps: 78%|███████▊ | 389/500 [29:02<06:54, 3.73s/it, loss=0.433, lr=0.000121] Steps: 78%|███████▊ | 390/500 [29:04<05:49, 3.18s/it, loss=0.433, lr=0.000121] Steps: 78%|███████▊ | 390/500 [29:04<05:49, 3.18s/it, loss=0.263, lr=0.000119] Steps: 78%|███████▊ | 391/500 [29:12<08:16, 4.55s/it, loss=0.263, lr=0.000119] Steps: 78%|███████▊ | 391/500 [29:12<08:16, 4.55s/it, loss=0.898, lr=0.000117] Steps: 78%|███████▊ | 392/500 [29:13<06:45, 3.75s/it, loss=0.898, lr=0.000117] Steps: 78%|███████▊ | 392/500 [29:13<06:45, 3.75s/it, loss=0.825, lr=0.000115] Steps: 79%|███████▊ | 393/500 [29:15<05:41, 3.19s/it, loss=0.825, lr=0.000115] Steps: 79%|███████▊ | 393/500 [29:15<05:41, 3.19s/it, loss=1.07, lr=0.000113] Steps: 79%|███████▉ | 394/500 [29:23<08:03, 4.56s/it, loss=1.07, lr=0.000113] Steps: 79%|███████▉ | 394/500 [29:23<08:03, 4.56s/it, loss=0.497, lr=0.000111] Steps: 79%|███████▉ | 395/500 [29:25<06:34, 3.76s/it, loss=0.497, lr=0.000111] Steps: 79%|███████▉ | 395/500 [29:25<06:34, 3.76s/it, loss=0.42, lr=0.000109] Steps: 79%|███████▉ | 396/500 [29:27<05:32, 3.20s/it, loss=0.42, lr=0.000109] Steps: 79%|███████▉ | 396/500 [29:27<05:32, 3.20s/it, loss=0.336, lr=0.000107] Steps: 79%|███████▉ | 397/500 [29:34<07:41, 4.48s/it, loss=0.336, lr=0.000107] Steps: 79%|███████▉ | 397/500 [29:34<07:41, 4.48s/it, loss=0.591, lr=0.000105] Steps: 80%|███████▉ | 398/500 [29:36<06:17, 3.70s/it, loss=0.591, lr=0.000105] Steps: 80%|███████▉ | 398/500 [29:36<06:17, 3.70s/it, loss=0.437, lr=0.000104] Steps: 80%|███████▉ | 399/500 [29:38<05:18, 3.16s/it, loss=0.437, lr=0.000104] Steps: 80%|███████▉ | 399/500 [29:38<05:18, 3.16s/it, loss=0.59, lr=0.000102] Steps: 80%|████████ | 400/500 [29:46<07:31, 4.51s/it, loss=0.59, lr=0.000102] Steps: 80%|████████ | 400/500 [29:46<07:31, 4.51s/it, loss=0.799, lr=0.0001] Steps: 80%|████████ | 401/500 [29:48<06:08, 3.72s/it, loss=0.799, lr=0.0001] Steps: 80%|████████ | 401/500 [29:48<06:08, 3.72s/it, loss=1.03, lr=9.82e-5] Steps: 80%|████████ | 402/500 [29:50<05:10, 3.17s/it, loss=1.03, lr=9.82e-5] Steps: 80%|████████ | 402/500 [29:50<05:10, 3.17s/it, loss=0.283, lr=9.64e-5] Steps: 81%|████████ | 403/500 [29:57<07:21, 4.55s/it, loss=0.283, lr=9.64e-5] Steps: 81%|████████ | 403/500 [29:57<07:21, 4.55s/it, loss=1.2, lr=9.46e-5] Steps: 81%|████████ | 404/500 [29:59<05:59, 3.75s/it, loss=1.2, lr=9.46e-5] Steps: 81%|████████ | 404/500 [29:59<05:59, 3.75s/it, loss=0.414, lr=9.28e-5] Steps: 81%|████████ | 405/500 [30:01<05:02, 3.19s/it, loss=0.414, lr=9.28e-5] Steps: 81%|████████ | 405/500 [30:01<05:02, 3.19s/it, loss=0.562, lr=9.11e-5] Steps: 81%|████████ | 406/500 [30:09<07:06, 4.54s/it, loss=0.562, lr=9.11e-5] Steps: 81%|████████ | 406/500 [30:09<07:06, 4.54s/it, loss=0.673, lr=8.93e-5] Steps: 81%|████████▏ | 407/500 [30:11<05:47, 3.74s/it, loss=0.673, lr=8.93e-5] Steps: 81%|████████▏ | 407/500 [30:11<05:47, 3.74s/it, loss=0.474, lr=8.76e-5] Steps: 82%|████████▏ | 408/500 [30:13<04:52, 3.18s/it, loss=0.474, lr=8.76e-5] Steps: 82%|████████▏ | 408/500 [30:13<04:52, 3.18s/it, loss=0.376, lr=8.59e-5] Steps: 82%|████████▏ | 409/500 [30:20<06:46, 4.47s/it, loss=0.376, lr=8.59e-5] Steps: 82%|████████▏ | 409/500 [30:20<06:46, 4.47s/it, loss=0.502, lr=8.41e-5] Steps: 82%|████████▏ | 410/500 [30:22<05:32, 3.69s/it, loss=0.502, lr=8.41e-5] Steps: 82%|████████▏ | 410/500 [30:22<05:32, 3.69s/it, loss=0.54, lr=8.24e-5] Steps: 82%|████████▏ | 411/500 [30:24<04:40, 3.15s/it, loss=0.54, lr=8.24e-5] Steps: 82%|████████▏ | 411/500 [30:24<04:40, 3.15s/it, loss=0.569, lr=8.08e-5] Steps: 82%|████████▏ | 412/500 [30:31<06:35, 4.49s/it, loss=0.569, lr=8.08e-5] Steps: 82%|████████▏ | 412/500 [30:31<06:35, 4.49s/it, loss=0.659, lr=7.91e-5] Steps: 83%|████████▎ | 413/500 [30:33<05:22, 3.71s/it, loss=0.659, lr=7.91e-5] Steps: 83%|████████▎ | 413/500 [30:33<05:22, 3.71s/it, loss=0.976, lr=7.74e-5] Steps: 83%|████████▎ | 414/500 [30:35<04:31, 3.16s/it, loss=0.976, lr=7.74e-5] Steps: 83%|████████▎ | 414/500 [30:35<04:31, 3.16s/it, loss=0.258, lr=7.58e-5] Steps: 83%|████████▎ | 415/500 [30:43<06:21, 4.49s/it, loss=0.258, lr=7.58e-5] Steps: 83%|████████▎ | 415/500 [30:43<06:21, 4.49s/it, loss=0.508, lr=7.41e-5] Steps: 83%|████████▎ | 416/500 [30:45<05:11, 3.71s/it, loss=0.508, lr=7.41e-5] Steps: 83%|████████▎ | 416/500 [30:45<05:11, 3.71s/it, loss=0.791, lr=7.25e-5] Steps: 83%|████████▎ | 417/500 [30:47<04:22, 3.16s/it, loss=0.791, lr=7.25e-5] Steps: 83%|████████▎ | 417/500 [30:47<04:22, 3.16s/it, loss=0.898, lr=7.09e-5] Steps: 84%|████████▎ | 418/500 [30:54<06:10, 4.52s/it, loss=0.898, lr=7.09e-5] Steps: 84%|████████▎ | 418/500 [30:54<06:10, 4.52s/it, loss=0.492, lr=6.93e-5] Steps: 84%|████████▍ | 419/500 [30:56<05:01, 3.73s/it, loss=0.492, lr=6.93e-5] Steps: 84%|████████▍ | 419/500 [30:56<05:01, 3.73s/it, loss=0.432, lr=6.77e-5] Steps: 84%|████████▍ | 420/500 [30:58<04:13, 3.17s/it, loss=0.432, lr=6.77e-5] Steps: 84%|████████▍ | 420/500 [30:58<04:13, 3.17s/it, loss=0.651, lr=6.62e-5] Steps: 84%|████████▍ | 421/500 [31:06<05:57, 4.52s/it, loss=0.651, lr=6.62e-5] Steps: 84%|████████▍ | 421/500 [31:06<05:57, 4.52s/it, loss=0.569, lr=6.46e-5] Steps: 84%|████████▍ | 422/500 [31:08<04:51, 3.73s/it, loss=0.569, lr=6.46e-5] Steps: 84%|████████▍ | 422/500 [31:08<04:51, 3.73s/it, loss=0.95, lr=6.31e-5] Steps: 85%|████████▍ | 423/500 [31:09<04:04, 3.18s/it, loss=0.95, lr=6.31e-5] Steps: 85%|████████▍ | 423/500 [31:09<04:04, 3.18s/it, loss=0.262, lr=6.16e-5] Steps: 85%|████████▍ | 424/500 [31:17<05:44, 4.54s/it, loss=0.262, lr=6.16e-5] Steps: 85%|████████▍ | 424/500 [31:17<05:44, 4.54s/it, loss=0.5, lr=6.01e-5] Steps: 85%|████████▌ | 425/500 [31:19<04:40, 3.74s/it, loss=0.5, lr=6.01e-5] Steps: 85%|████████▌ | 425/500 [31:19<04:40, 3.74s/it, loss=0.715, lr=5.86e-5] Steps: 85%|████████▌ | 426/500 [31:21<03:55, 3.18s/it, loss=0.715, lr=5.86e-5] Steps: 85%|████████▌ | 426/500 [31:21<03:55, 3.18s/it, loss=0.514, lr=5.71e-5] Steps: 85%|████████▌ | 427/500 [31:29<05:31, 4.53s/it, loss=0.514, lr=5.71e-5] Steps: 85%|████████▌ | 427/500 [31:29<05:31, 4.53s/it, loss=0.494, lr=5.56e-5] Steps: 86%|████████▌ | 428/500 [31:30<04:29, 3.74s/it, loss=0.494, lr=5.56e-5] Steps: 86%|████████▌ | 428/500 [31:30<04:29, 3.74s/it, loss=0.679, lr=5.42e-5] Steps: 86%|████████▌ | 429/500 [31:32<03:45, 3.18s/it, loss=0.679, lr=5.42e-5] Steps: 86%|████████▌ | 429/500 [31:32<03:45, 3.18s/it, loss=0.998, lr=5.28e-5] Steps: 86%|████████▌ | 430/500 [31:40<05:17, 4.53s/it, loss=0.998, lr=5.28e-5] Steps: 86%|████████▌ | 430/500 [31:40<05:17, 4.53s/it, loss=0.495, lr=5.14e-5] Steps: 86%|████████▌ | 431/500 [31:42<04:17, 3.73s/it, loss=0.495, lr=5.14e-5] Steps: 86%|████████▌ | 431/500 [31:42<04:17, 3.73s/it, loss=0.771, lr=5e-5] Steps: 86%|████████▋ | 432/500 [31:44<03:36, 3.18s/it, loss=0.771, lr=5e-5] Steps: 86%|████████▋ | 432/500 [31:44<03:36, 3.18s/it, loss=0.876, lr=4.86e-5] Steps: 87%|████████▋ | 433/500 [31:51<05:02, 4.52s/it, loss=0.876, lr=4.86e-5] Steps: 87%|████████▋ | 433/500 [31:51<05:02, 4.52s/it, loss=0.559, lr=4.72e-5] Steps: 87%|████████▋ | 434/500 [31:53<04:05, 3.73s/it, loss=0.559, lr=4.72e-5] Steps: 87%|████████▋ | 434/500 [31:53<04:05, 3.73s/it, loss=0.625, lr=4.59e-5] Steps: 87%|████████▋ | 435/500 [31:55<03:26, 3.17s/it, loss=0.625, lr=4.59e-5] Steps: 87%|████████▋ | 435/500 [31:55<03:26, 3.17s/it, loss=0.262, lr=4.46e-5] Steps: 87%|████████▋ | 436/500 [32:03<04:48, 4.51s/it, loss=0.262, lr=4.46e-5] Steps: 87%|████████▋ | 436/500 [32:03<04:48, 4.51s/it, loss=0.577, lr=4.33e-5] Steps: 87%|████████▋ | 437/500 [32:05<03:54, 3.72s/it, loss=0.577, lr=4.33e-5] Steps: 87%|████████▋ | 437/500 [32:05<03:54, 3.72s/it, loss=0.416, lr=4.2e-5] Steps: 88%|████████▊ | 438/500 [32:07<03:16, 3.17s/it, loss=0.416, lr=4.2e-5] Steps: 88%|████████▊ | 438/500 [32:07<03:16, 3.17s/it, loss=0.516, lr=4.07e-5] Steps: 88%|████████▊ | 439/500 [32:14<04:35, 4.51s/it, loss=0.516, lr=4.07e-5] Steps: 88%|████████▊ | 439/500 [32:14<04:35, 4.51s/it, loss=0.498, lr=3.94e-5] Steps: 88%|████████▊ | 440/500 [32:16<03:43, 3.72s/it, loss=0.498, lr=3.94e-5] Steps: 88%|████████▊ | 440/500 [32:16<03:43, 3.72s/it, loss=0.76, lr=3.82e-5] Steps: 88%|████████▊ | 441/500 [32:18<03:07, 3.17s/it, loss=0.76, lr=3.82e-5] Steps: 88%|████████▊ | 441/500 [32:18<03:07, 3.17s/it, loss=0.486, lr=3.7e-5] Steps: 88%|████████▊ | 442/500 [32:26<04:21, 4.51s/it, loss=0.486, lr=3.7e-5] Steps: 88%|████████▊ | 442/500 [32:26<04:21, 4.51s/it, loss=0.9, lr=3.58e-5] Steps: 89%|████████▊ | 443/500 [32:27<03:32, 3.72s/it, loss=0.9, lr=3.58e-5] Steps: 89%|████████▊ | 443/500 [32:27<03:32, 3.72s/it, loss=0.658, lr=3.46e-5] Steps: 89%|████████▉ | 444/500 [32:29<02:57, 3.17s/it, loss=0.658, lr=3.46e-5] Steps: 89%|████████▉ | 444/500 [32:29<02:57, 3.17s/it, loss=0.726, lr=3.34e-5] Steps: 89%|████████▉ | 445/500 [32:37<04:08, 4.52s/it, loss=0.726, lr=3.34e-5] Steps: 89%|████████▉ | 445/500 [32:37<04:08, 4.52s/it, loss=0.512, lr=3.23e-5] Steps: 89%|████████▉ | 446/500 [32:39<03:21, 3.73s/it, loss=0.512, lr=3.23e-5] Steps: 89%|████████▉ | 446/500 [32:39<03:21, 3.73s/it, loss=0.468, lr=3.11e-5] Steps: 89%|████████▉ | 447/500 [32:41<02:48, 3.17s/it, loss=0.468, lr=3.11e-5] Steps: 89%|████████▉ | 447/500 [32:41<02:48, 3.17s/it, loss=0.318, lr=3e-5] Steps: 90%|████████▉ | 448/500 [32:48<03:55, 4.52s/it, loss=0.318, lr=3e-5] Steps: 90%|████████▉ | 448/500 [32:48<03:55, 4.52s/it, loss=0.527, lr=2.89e-5] Steps: 90%|████████▉ | 449/500 [32:50<03:10, 3.73s/it, loss=0.527, lr=2.89e-5] Steps: 90%|████████▉ | 449/500 [32:50<03:10, 3.73s/it, loss=0.433, lr=2.79e-5] Steps: 90%|█████████ | 450/500 [32:52<02:38, 3.18s/it, loss=0.433, lr=2.79e-5] Steps: 90%|█████████ | 450/500 [32:52<02:38, 3.18s/it, loss=0.859, lr=2.68e-5] Steps: 90%|█████████ | 451/500 [33:00<03:42, 4.54s/it, loss=0.859, lr=2.68e-5] Steps: 90%|█████████ | 451/500 [33:00<03:42, 4.54s/it, loss=0.539, lr=2.58e-5] Steps: 90%|█████████ | 452/500 [33:02<02:59, 3.74s/it, loss=0.539, lr=2.58e-5] Steps: 90%|█████████ | 452/500 [33:02<02:59, 3.74s/it, loss=0.422, lr=2.47e-5] Steps: 91%|█████████ | 453/500 [33:04<02:29, 3.19s/it, loss=0.422, lr=2.47e-5] Steps: 91%|█████████ | 453/500 [33:04<02:29, 3.19s/it, loss=0.258, lr=2.37e-5] Steps: 91%|█████████ | 454/500 [33:12<03:30, 4.57s/it, loss=0.258, lr=2.37e-5] Steps: 91%|█████████ | 454/500 [33:12<03:30, 4.57s/it, loss=0.517, lr=2.28e-5] Steps: 91%|█████████ | 455/500 [33:13<02:49, 3.76s/it, loss=0.517, lr=2.28e-5] Steps: 91%|█████████ | 455/500 [33:13<02:49, 3.76s/it, loss=1.03, lr=2.18e-5] Steps: 91%|█████████ | 456/500 [33:15<02:20, 3.20s/it, loss=1.03, lr=2.18e-5] Steps: 91%|█████████ | 456/500 [33:15<02:20, 3.20s/it, loss=0.708, lr=2.09e-5] Steps: 91%|█████████▏| 457/500 [33:23<03:14, 4.52s/it, loss=0.708, lr=2.09e-5] Steps: 91%|█████████▏| 457/500 [33:23<03:14, 4.52s/it, loss=0.763, lr=1.99e-5] Steps: 92%|█████████▏| 458/500 [33:25<02:36, 3.73s/it, loss=0.763, lr=1.99e-5] Steps: 92%|█████████▏| 458/500 [33:25<02:36, 3.73s/it, loss=1.03, lr=1.9e-5] Steps: 92%|█████████▏| 459/500 [33:27<02:10, 3.18s/it, loss=1.03, lr=1.9e-5] Steps: 92%|█████████▏| 459/500 [33:27<02:10, 3.18s/it, loss=0.278, lr=1.82e-5] Steps: 92%|█████████▏| 460/500 [33:34<03:01, 4.54s/it, loss=0.278, lr=1.82e-5] Steps: 92%|█████████▏| 460/500 [33:34<03:01, 4.54s/it, loss=0.628, lr=1.73e-5] Steps: 92%|█████████▏| 461/500 [33:36<02:25, 3.74s/it, loss=0.628, lr=1.73e-5] Steps: 92%|█████████▏| 461/500 [33:36<02:25, 3.74s/it, loss=0.912, lr=1.64e-5] Steps: 92%|█████████▏| 462/500 [33:38<02:00, 3.18s/it, loss=0.912, lr=1.64e-5] Steps: 92%|█████████▏| 462/500 [33:38<02:00, 3.18s/it, loss=0.257, lr=1.56e-5] Steps: 93%|█████████▎| 463/500 [33:46<02:48, 4.55s/it, loss=0.257, lr=1.56e-5] Steps: 93%|█████████▎| 463/500 [33:46<02:48, 4.55s/it, loss=1.24, lr=1.48e-5] Steps: 93%|█████████▎| 464/500 [33:48<02:14, 3.75s/it, loss=1.24, lr=1.48e-5] Steps: 93%|█████████▎| 464/500 [33:48<02:14, 3.75s/it, loss=0.45, lr=1.4e-5] Steps: 93%|█████████▎| 465/500 [33:50<01:51, 3.19s/it, loss=0.45, lr=1.4e-5] Steps: 93%|█████████▎| 465/500 [33:50<01:51, 3.19s/it, loss=0.263, lr=1.33e-5] Steps: 93%|█████████▎| 466/500 [33:57<02:33, 4.53s/it, loss=0.263, lr=1.33e-5] Steps: 93%|█████████▎| 466/500 [33:57<02:33, 4.53s/it, loss=0.56, lr=1.25e-5] Steps: 93%|█████████▎| 467/500 [33:59<02:03, 3.73s/it, loss=0.56, lr=1.25e-5] Steps: 93%|█████████▎| 467/500 [33:59<02:03, 3.73s/it, loss=0.629, lr=1.18e-5] Steps: 94%|█████████▎| 468/500 [34:01<01:41, 3.18s/it, loss=0.629, lr=1.18e-5] Steps: 94%|█████████▎| 468/500 [34:01<01:41, 3.18s/it, loss=0.263, lr=1.11e-5] Steps: 94%|█████████▍| 469/500 [34:09<02:20, 4.52s/it, loss=0.263, lr=1.11e-5] Steps: 94%|█████████▍| 469/500 [34:09<02:20, 4.52s/it, loss=0.918, lr=1.04e-5] Steps: 94%|█████████▍| 470/500 [34:11<01:51, 3.73s/it, loss=0.918, lr=1.04e-5] Steps: 94%|█████████▍| 470/500 [34:11<01:51, 3.73s/it, loss=0.991, lr=9.79e-6] Steps: 94%|█████████▍| 471/500 [34:12<01:32, 3.18s/it, loss=0.991, lr=9.79e-6] Steps: 94%|█████████▍| 471/500 [34:12<01:32, 3.18s/it, loss=0.33, lr=9.15e-6] Steps: 94%|█████████▍| 472/500 [34:20<02:06, 4.53s/it, loss=0.33, lr=9.15e-6] Steps: 94%|█████████▍| 472/500 [34:20<02:06, 4.53s/it, loss=0.536, lr=8.54e-6] Steps: 95%|█████████▍| 473/500 [34:22<01:40, 3.73s/it, loss=0.536, lr=8.54e-6] Steps: 95%|█████████▍| 473/500 [34:22<01:40, 3.73s/it, loss=0.487, lr=7.94e-6] Steps: 95%|█████████▍| 474/500 [34:24<01:22, 3.18s/it, loss=0.487, lr=7.94e-6] Steps: 95%|█████████▍| 474/500 [34:24<01:22, 3.18s/it, loss=0.262, lr=7.37e-6] Steps: 95%|█████████▌| 475/500 [34:32<01:53, 4.54s/it, loss=0.262, lr=7.37e-6] Steps: 95%|█████████▌| 475/500 [34:32<01:53, 4.54s/it, loss=0.487, lr=6.81e-6] Steps: 95%|█████████▌| 476/500 [34:34<01:29, 3.74s/it, loss=0.487, lr=6.81e-6] Steps: 95%|█████████▌| 476/500 [34:34<01:29, 3.74s/it, loss=0.797, lr=6.28e-6] Steps: 95%|█████████▌| 477/500 [34:35<01:13, 3.18s/it, loss=0.797, lr=6.28e-6] Steps: 95%|█████████▌| 477/500 [34:35<01:13, 3.18s/it, loss=0.99, lr=5.77e-6] Steps: 96%|█████████▌| 478/500 [34:43<01:38, 4.47s/it, loss=0.99, lr=5.77e-6] Steps: 96%|█████████▌| 478/500 [34:43<01:38, 4.47s/it, loss=0.612, lr=5.28e-6] Steps: 96%|█████████▌| 479/500 [34:45<01:17, 3.70s/it, loss=0.612, lr=5.28e-6] Steps: 96%|█████████▌| 479/500 [34:45<01:17, 3.70s/it, loss=0.422, lr=4.82e-6] Steps: 96%|█████████▌| 480/500 [34:47<01:03, 3.15s/it, loss=0.422, lr=4.82e-6] Steps: 96%|█████████▌| 480/500 [34:47<01:03, 3.15s/it, loss=0.384, lr=4.37e-6] Steps: 96%|█████████▌| 481/500 [34:54<01:25, 4.48s/it, loss=0.384, lr=4.37e-6] Steps: 96%|█████████▌| 481/500 [34:54<01:25, 4.48s/it, loss=1.03, lr=3.95e-6] Steps: 96%|█████████▋| 482/500 [34:56<01:06, 3.70s/it, loss=1.03, lr=3.95e-6] Steps: 96%|█████████▋| 482/500 [34:56<01:06, 3.70s/it, loss=0.76, lr=3.54e-6] Steps: 97%|█████████▋| 483/500 [34:58<00:53, 3.16s/it, loss=0.76, lr=3.54e-6] Steps: 97%|█████████▋| 483/500 [34:58<00:53, 3.16s/it, loss=0.366, lr=3.16e-6] Steps: 97%|█████████▋| 484/500 [35:06<01:12, 4.52s/it, loss=0.366, lr=3.16e-6] Steps: 97%|█████████▋| 484/500 [35:06<01:12, 4.52s/it, loss=0.491, lr=2.8e-6] Steps: 97%|█████████▋| 485/500 [35:08<00:55, 3.73s/it, loss=0.491, lr=2.8e-6] Steps: 97%|█████████▋| 485/500 [35:08<00:55, 3.73s/it, loss=0.633, lr=2.46e-6] Steps: 97%|█████████▋| 486/500 [35:09<00:44, 3.18s/it, loss=0.633, lr=2.46e-6] Steps: 97%|█████████▋| 486/500 [35:09<00:44, 3.18s/it, loss=1.02, lr=2.15e-6] Steps: 97%|█████████▋| 487/500 [35:17<00:58, 4.48s/it, loss=1.02, lr=2.15e-6] Steps: 97%|█████████▋| 487/500 [35:17<00:58, 4.48s/it, loss=0.521, lr=1.85e-6] Steps: 98%|█████████▊| 488/500 [35:19<00:44, 3.70s/it, loss=0.521, lr=1.85e-6] Steps: 98%|█████████▊| 488/500 [35:19<00:44, 3.70s/it, loss=0.488, lr=1.58e-6] Steps: 98%|█████████▊| 489/500 [35:21<00:34, 3.16s/it, loss=0.488, lr=1.58e-6] Steps: 98%|█████████▊| 489/500 [35:21<00:34, 3.16s/it, loss=0.295, lr=1.33e-6] Steps: 98%|█████████▊| 490/500 [35:28<00:44, 4.48s/it, loss=0.295, lr=1.33e-6] Steps: 98%|█████████▊| 490/500 [35:28<00:44, 4.48s/it, loss=0.598, lr=1.1e-6] Steps: 98%|█████████▊| 491/500 [35:30<00:33, 3.70s/it, loss=0.598, lr=1.1e-6] Steps: 98%|█████████▊| 491/500 [35:30<00:33, 3.70s/it, loss=0.925, lr=8.88e-7] Steps: 98%|█████████▊| 492/500 [35:32<00:25, 3.16s/it, loss=0.925, lr=8.88e-7] Steps: 98%|█████████▊| 492/500 [35:32<00:25, 3.16s/it, loss=0.259, lr=7.01e-7] Steps: 99%|█████████▊| 493/500 [35:40<00:31, 4.53s/it, loss=0.259, lr=7.01e-7] Steps: 99%|█████████▊| 493/500 [35:40<00:31, 4.53s/it, loss=0.579, lr=5.37e-7] Steps: 99%|█████████▉| 494/500 [35:42<00:22, 3.74s/it, loss=0.579, lr=5.37e-7] Steps: 99%|█████████▉| 494/500 [35:42<00:22, 3.74s/it, loss=0.505, lr=3.95e-7] Steps: 99%|█████████▉| 495/500 [35:44<00:15, 3.18s/it, loss=0.505, lr=3.95e-7] Steps: 99%|█████████▉| 495/500 [35:44<00:15, 3.18s/it, loss=0.429, lr=2.74e-7] Steps: 99%|█████████▉| 496/500 [35:51<00:18, 4.50s/it, loss=0.429, lr=2.74e-7] Steps: 99%|█████████▉| 496/500 [35:51<00:18, 4.50s/it, loss=0.827, lr=1.75e-7] Steps: 99%|█████████▉| 497/500 [35:53<00:11, 3.72s/it, loss=0.827, lr=1.75e-7] Steps: 99%|█████████▉| 497/500 [35:53<00:11, 3.72s/it, loss=0.435, lr=9.87e-8] Steps: 100%|█████████▉| 498/500 [35:55<00:06, 3.17s/it, loss=0.435, lr=9.87e-8] Steps: 100%|█████████▉| 498/500 [35:55<00:06, 3.17s/it, loss=0.473, lr=4.39e-8] Steps: 100%|█████████▉| 499/500 [36:03<00:04, 4.53s/it, loss=0.473, lr=4.39e-8] Steps: 100%|█████████▉| 499/500 [36:03<00:04, 4.53s/it, loss=1.04, lr=1.1e-8] Steps: 100%|██████████| 500/500 [36:05<00:00, 3.73s/it, loss=1.04, lr=1.1e-8] Steps: 100%|██████████| 500/500 [36:05<00:00, 3.73s/it, loss=0.847, lr=0] Steps: 100%|██████████| 500/500 [36:09<00:00, 4.34s/it, loss=0.847, lr=0] ---Tar up output directory--- mochi-lora/ mochi-lora/pytorch_lora_weights.safetensors Uploading to Hugging Face: lucataco/mochi-lora-melty HF Repo URL: https://huggingface.co/lucataco/mochi-lora-melty pytorch_lora_weights.safetensors: 0%| | 0.00/76.1M [00:00<?, ?B/s] pytorch_lora_weights.safetensors: 11%|█ | 8.52M/76.1M [00:00<00:00, 85.2MB/s] pytorch_lora_weights.safetensors: 22%|██▏ | 17.0M/76.1M [00:00<00:01, 51.7MB/s] pytorch_lora_weights.safetensors: 42%|████▏ | 32.0M/76.1M [00:00<00:00, 55.8MB/s] pytorch_lora_weights.safetensors: 63%|██████▎ | 48.0M/76.1M [00:00<00:00, 56.2MB/s] pytorch_lora_weights.safetensors: 84%|████████▍ | 64.0M/76.1M [00:01<00:00, 61.6MB/s] pytorch_lora_weights.safetensors: 100%|██████████| 76.1M/76.1M [00:01<00:00, 54.3MB/s] Successfully uploaded model to https://huggingface.co/lucataco/mochi-lora-melty
Want to make some of these yourself?
Run this model