Prediction

Model

pixray/text2image-future:0d01ba09

4porzbc32fbv7ezhe7u7tprg5i

Status

Succeeded

Source

Web

Hardware

–

Total duration

8m 59s

Created

almost 4 years ago by dribnet

Webhook

–

Input

prompts: evil medical device from a post apocalyptic veterinary clinic #pixelart
drawer: pixel
settings: # random number seed can be a word or number seed: reference # higher quality than default quality: better # smooth out the result a bit custom_loss: smoothness:0.5 # enable transparency in image transparent: true # how much to encourage transparency (can also be negative) transparent_weight: 0.1

{
  "drawer": "pixel",
  "prompts": "evil medical device from a post apocalyptic veterinary clinic #pixelart",
  "settings": "# random number seed can be a word or number\nseed: reference\n# higher quality than default\nquality: better\n# smooth out the result a bit\ncustom_loss: smoothness:0.5\n# enable transparency in image\ntransparent: true\n# how much to encourage transparency (can also be negative) \ntransparent_weight: 0.1\n\n"
}

Install Replicate’s Node.js client library:

npm install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=r8_A9F**********************************

This is your API token. Keep it to yourself.

Import and set up the client:

import Replicate from "replicate";
import fs from "node:fs";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run pixray/text2image-future using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "pixray/text2image-future:0d01ba09e8fa182455bc3ccc7b11b834356645dbcfa2712d6e5a4d5615a7a9f0",
  {
    input: {
      drawer: "pixel",
      prompts: "evil medical device from a post apocalyptic veterinary clinic #pixelart",
      settings: "# random number seed can be a word or number\nseed: reference\n# higher quality than default\nquality: better\n# smooth out the result a bit\ncustom_loss: smoothness:0.5\n# enable transparency in image\ntransparent: true\n# how much to encourage transparency (can also be negative) \ntransparent_weight: 0.1\n\n"
    }
  }
);

// To access the file URL:
console.log(output[0].url()); //=> "http://example.com"

// To write the file to disk:
fs.writeFile("my-image.png", output[0]);

To learn more, take a look at the guide on getting started with Node.js.

Install Replicate’s Python client library:

pip install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=r8_A9F**********************************

This is your API token. Keep it to yourself.

Import the client:

import replicate

Run pixray/text2image-future using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "pixray/text2image-future:0d01ba09e8fa182455bc3ccc7b11b834356645dbcfa2712d6e5a4d5615a7a9f0",
    input={
        "drawer": "pixel",
        "prompts": "evil medical device from a post apocalyptic veterinary clinic #pixelart",
        "settings": "# random number seed can be a word or number\nseed: reference\n# higher quality than default\nquality: better\n# smooth out the result a bit\ncustom_loss: smoothness:0.5\n# enable transparency in image\ntransparent: true\n# how much to encourage transparency (can also be negative) \ntransparent_weight: 0.1\n\n"
    }
)

# The pixray/text2image-future model can stream output as it's running.
# The predict method returns an iterator, and you can iterate over that output.
for item in output:
    # https://replicate.com/pixray/text2image-future/api#output-schema
    print(item)

To learn more, take a look at the guide on getting started with Python.

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=r8_A9F**********************************

This is your API token. Keep it to yourself.

Run pixray/text2image-future 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": "pixray/text2image-future:0d01ba09e8fa182455bc3ccc7b11b834356645dbcfa2712d6e5a4d5615a7a9f0",
    "input": {
      "drawer": "pixel",
      "prompts": "evil medical device from a post apocalyptic veterinary clinic #pixelart",
      "settings": "# random number seed can be a word or number\\nseed: reference\\n# higher quality than default\\nquality: better\\n# smooth out the result a bit\\ncustom_loss: smoothness:0.5\\n# enable transparency in image\\ntransparent: true\\n# how much to encourage transparency (can also be negative) \\ntransparent_weight: 0.1\\n\\n"
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

Output

31 / 31

{
  "id": "4porzbc32fbv7ezhe7u7tprg5i",
  "model": "pixray/text2image-future",
  "version": "0d01ba09e8fa182455bc3ccc7b11b834356645dbcfa2712d6e5a4d5615a7a9f0",
  "input": {
    "drawer": "pixel",
    "prompts": "evil medical device from a post apocalyptic veterinary clinic #pixelart",
    "settings": "# random number seed can be a word or number\nseed: reference\n# higher quality than default\nquality: better\n# smooth out the result a bit\ncustom_loss: smoothness:0.5\n# enable transparency in image\ntransparent: true\n# how much to encourage transparency (can also be negative) \ntransparent_weight: 0.1\n\n"
  },
  "logs": "---> BasePixrayPredictor Predict\nUsing seed: 3903845079\nRunning pixeldrawer with 80x45 grid\n\n  0%|                                               | 0.00/244M [00:00<?, ?iB/s]\n  0%|▏                                     | 1.19M/244M [00:00<00:21, 11.9MiB/s]\n  5%|█▉                                    | 12.6M/244M [00:00<00:14, 16.3MiB/s]\n 14%|█████▏                                | 33.6M/244M [00:00<00:09, 22.6MiB/s]\n 21%|███████▉                              | 50.6M/244M [00:00<00:06, 30.6MiB/s]\n 27%|██████████▍                           | 67.0M/244M [00:00<00:04, 40.6MiB/s]\n 34%|████████████▊                         | 82.4M/244M [00:00<00:03, 52.4MiB/s]\n 44%|████████████████▉                      | 106M/244M [00:00<00:02, 68.7MiB/s]\n 53%|████████████████████▌                  | 128M/244M [00:00<00:01, 87.0MiB/s]\n 61%|████████████████████████▌               | 149M/244M [00:00<00:00, 106MiB/s]\n 71%|████████████████████████████▍           | 173M/244M [00:01<00:00, 129MiB/s]\n 79%|███████████████████████████████▊        | 194M/244M [00:01<00:00, 146MiB/s]\n 88%|███████████████████████████████████▏    | 214M/244M [00:01<00:00, 161MiB/s]\n 96%|██████████████████████████████████████▍ | 235M/244M [00:01<00:00, 171MiB/s]\n100%|████████████████████████████████████████| 244M/244M [00:01<00:00, 187MiB/s]\nLoaded CLIP RN50: 224x224 and 102.01M params\nLoaded CLIP ViT-B/32: 224x224 and 151.28M params\nLoaded CLIP ViT-B/16: 224x224 and 149.62M params\nUsing device: cuda:0\n\nOptimising using: Adam\nUsing text prompts: ['evil medical device from a post apocalyptic veterinary clinic #pixelart']\nusing custom losses: smoothness:0.5\n0it [00:00, ?it/s]\niter: 0, loss: 3.23, losses: 1.01, 0.0859, 0.928, 0.0617, 0.904, 0.0643, 0.1, 0.0725 (-0=>3.227)\n\n\n0it [00:00, ?it/s]/root/.pyenv/versions/3.8.12/lib/python3.8/site-packages/torch/nn/functional.py:3609: UserWarning: Default upsampling behavior when mode=bilinear is changed to align_corners=False since 0.4.0. Please specify align_corners=True if the old behavior is desired. See the documentation of nn.Upsample for details.\n  warnings.warn(\n\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 10, loss: 2.88, losses: 0.916, 0.0816, 0.799, 0.061, 0.798, 0.0613, 0.0905, 0.0707 (-0=>2.878)\n\n\n0it [00:00, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 20, loss: 2.73, losses: 0.874, 0.0823, 0.753, 0.0629, 0.74, 0.0628, 0.0843, 0.0704 (-0=>2.73)\n\n\n0it [00:01, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 30, loss: 2.62, losses: 0.848, 0.0824, 0.724, 0.0635, 0.685, 0.0644, 0.0787, 0.0711 (-0=>2.617)\n\n\n0it [00:01, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 40, loss: 2.53, losses: 0.818, 0.0824, 0.695, 0.0662, 0.66, 0.0663, 0.0731, 0.0716 (-3=>2.533)\n\n\n0it [00:00, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 50, loss: 2.53, losses: 0.815, 0.0837, 0.7, 0.0643, 0.66, 0.0655, 0.0681, 0.0704 (-2=>2.493)\n\n\n0it [00:00, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 60, loss: 2.46, losses: 0.786, 0.083, 0.676, 0.0673, 0.648, 0.0676, 0.0642, 0.0714 (-1=>2.428)\n\n\n0it [00:00, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 70, loss: 2.46, losses: 0.781, 0.0844, 0.682, 0.0658, 0.648, 0.0672, 0.0608, 0.0695 (-1=>2.412)\n\n\n0it [00:00, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 80, loss: 2.46, losses: 0.788, 0.0835, 0.682, 0.0654, 0.645, 0.0668, 0.0579, 0.0716 (-2=>2.376)\n\n\n0it [00:00, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 90, loss: 2.41, losses: 0.771, 0.0833, 0.665, 0.067, 0.631, 0.0677, 0.0552, 0.0682 (-3=>2.364)\n\n\n0it [00:00, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 100, loss: 2.35, losses: 0.746, 0.0847, 0.65, 0.067, 0.619, 0.0692, 0.053, 0.0624 (-0=>2.352)\n\n\n0it [00:00, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 110, loss: 2.35, losses: 0.748, 0.0844, 0.651, 0.067, 0.616, 0.0691, 0.051, 0.0596 (-1=>2.331)\n\n\n0it [00:00, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 120, loss: 2.36, losses: 0.756, 0.0844, 0.651, 0.0662, 0.618, 0.069, 0.0491, 0.0617 (-11=>2.331)\n\n\n0it [00:00, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 130, loss: 2.36, losses: 0.756, 0.0839, 0.654, 0.0666, 0.622, 0.0688, 0.0472, 0.061 (-6=>2.31)\n\n\n0it [00:00, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 140, loss: 2.36, losses: 0.753, 0.0854, 0.652, 0.0665, 0.623, 0.0677, 0.0458, 0.0638 (-6=>2.305)\n\n\n0it [00:00, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 150, loss: 2.33, losses: 0.741, 0.0847, 0.646, 0.0665, 0.614, 0.0693, 0.0447, 0.0621 (-5=>2.297)\n\n\n0it [00:00, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 160, loss: 2.31, losses: 0.734, 0.0842, 0.641, 0.0675, 0.609, 0.0699, 0.0437, 0.061 (-15=>2.297)\n\n\n0it [00:00, ?it/s]\n0it [00:15, ?it/s]\n\n0it [00:00, ?it/s]\niter: 170, loss: 2.3, losses: 0.736, 0.0851, 0.633, 0.0673, 0.606, 0.0703, 0.0428, 0.0597 (-3=>2.289)\n\n\n0it [00:00, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 180, loss: 2.31, losses: 0.741, 0.0849, 0.637, 0.0668, 0.61, 0.0689, 0.0418, 0.0622 (-5=>2.28)\n\n\n0it [00:00, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 190, loss: 2.32, losses: 0.745, 0.0844, 0.642, 0.0659, 0.613, 0.0683, 0.0409, 0.0629 (-15=>2.28)\n\n\n0it [00:00, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 200, loss: 2.26, losses: 0.716, 0.0857, 0.624, 0.0677, 0.599, 0.0703, 0.0401, 0.0618 (-0=>2.265)\n\n\n0it [00:01, ?it/s]\n0it [00:17, ?it/s]\n\n0it [00:00, ?it/s]\niter: 210, loss: 2.33, losses: 0.746, 0.085, 0.646, 0.0659, 0.616, 0.0683, 0.0392, 0.0641 (-1=>2.255)\n\n\n0it [00:00, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 220, loss: 2.27, losses: 0.724, 0.0844, 0.628, 0.0674, 0.601, 0.0704, 0.0386, 0.0578 (-5=>2.252)\n\n\nDropping learning rate\n0it [00:01, ?it/s]\n0it [00:17, ?it/s]\n\n0it [00:00, ?it/s]\niter: 230, loss: 2.3, losses: 0.739, 0.085, 0.634, 0.0671, 0.606, 0.0693, 0.0384, 0.0602 (-1=>2.264)\n\n\n0it [00:01, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 240, loss: 2.26, losses: 0.716, 0.0849, 0.627, 0.0678, 0.599, 0.07, 0.0383, 0.0565 (-5=>2.25)\n\n\n0it [00:01, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 250, loss: 2.25, losses: 0.718, 0.0857, 0.619, 0.0679, 0.592, 0.0703, 0.0382, 0.056 (-0=>2.247)\n\n\n0it [00:01, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 260, loss: 2.27, losses: 0.719, 0.0846, 0.628, 0.067, 0.6, 0.0704, 0.0382, 0.0594 (-5=>2.244)\n\n\n0it [00:01, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 270, loss: 2.27, losses: 0.724, 0.0859, 0.63, 0.0669, 0.598, 0.0703, 0.0381, 0.0555 (-1=>2.241)\n\n\n0it [00:01, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 280, loss: 2.26, losses: 0.72, 0.0853, 0.625, 0.0674, 0.597, 0.0703, 0.0381, 0.0581 (-11=>2.241)\n\n\n0it [00:00, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 290, loss: 2.25, losses: 0.716, 0.0855, 0.621, 0.0681, 0.593, 0.0701, 0.0381, 0.0574 (-21=>2.241)\n\n\n0it [00:00, ?it/s]\n0it [00:16, ?it/s]\n\n0it [00:00, ?it/s]\niter: 300, finished (-31=>2.241)\n\n\n0it [00:00, ?it/s]\n0it [00:00, ?it/s]",
  "output": [
    "https://replicate.delivery/mgxm/ba9efa8b-0536-484d-bb61-c22d801e7ee1/tempfile.png",
    "https://replicate.delivery/mgxm/808b18a0-2632-47b3-b74e-72d62602d03a/tempfile.png",
    "https://replicate.delivery/mgxm/615eb647-7688-4636-aecd-f39119ddc241/tempfile.png",
    "https://replicate.delivery/mgxm/0be4c412-fbf5-43bc-a3f7-c15d0e334b09/tempfile.png",
    "https://replicate.delivery/mgxm/bfdd74ed-3bf4-4696-bfbc-59bd2273cdd4/tempfile.png",
    "https://replicate.delivery/mgxm/6797d76f-5e30-48cc-898c-454a2759d59b/tempfile.png",
    "https://replicate.delivery/mgxm/e0c4f446-5ec0-4ca8-bd3e-a6b87c607097/tempfile.png",
    "https://replicate.delivery/mgxm/9a4db59e-ed71-4d90-aee2-057ad3e813e6/tempfile.png",
    "https://replicate.delivery/mgxm/173ba724-6b33-4f1d-ae90-a5749d8d8b10/tempfile.png",
    "https://replicate.delivery/mgxm/8e4c4aa2-1b9a-472a-8c8d-6ef98b0ea8ba/tempfile.png",
    "https://replicate.delivery/mgxm/025527a9-ca98-4731-857e-996512e4bc38/tempfile.png",
    "https://replicate.delivery/mgxm/23012112-765f-4961-9a72-25f6ef716fd2/tempfile.png",
    "https://replicate.delivery/mgxm/2a8dc4f5-ae14-4f96-9bec-893067d31c2c/tempfile.png",
    "https://replicate.delivery/mgxm/343ed30a-ad54-410a-9f42-e7b144c4d194/tempfile.png",
    "https://replicate.delivery/mgxm/2d6f7f08-1738-4280-aa4b-b1c4a4d821a9/tempfile.png",
    "https://replicate.delivery/mgxm/078b8255-7ea7-4f9c-8ddb-093cf3b908d6/tempfile.png",
    "https://replicate.delivery/mgxm/a35a83af-5ca2-482a-9387-eec3d2e335c7/tempfile.png",
    "https://replicate.delivery/mgxm/7db822f6-2ed6-4bfa-b2aa-78c366982c39/tempfile.png",
    "https://replicate.delivery/mgxm/38f81049-d86b-4bd9-b93a-41fa3b25850b/tempfile.png",
    "https://replicate.delivery/mgxm/0fcc4303-c779-493b-92d1-3ec5157f274f/tempfile.png",
    "https://replicate.delivery/mgxm/29c8c9dd-d7a0-41d0-97f1-23115f6a795f/tempfile.png",
    "https://replicate.delivery/mgxm/e527168a-625f-496c-b949-54e647ebcfda/tempfile.png",
    "https://replicate.delivery/mgxm/d75fb245-a66f-4012-b25d-5658acd2ef90/tempfile.png",
    "https://replicate.delivery/mgxm/b89eef0b-9952-4c01-aa07-554600250659/tempfile.png",
    "https://replicate.delivery/mgxm/213b5ab1-8013-42ad-adc9-30f42295ec34/tempfile.png",
    "https://replicate.delivery/mgxm/86b6363c-21cd-4c86-a929-2d0cd098b415/tempfile.png",
    "https://replicate.delivery/mgxm/9e225cb0-a6f0-456b-b2ed-116a8d7e6646/tempfile.png",
    "https://replicate.delivery/mgxm/918247df-9b61-48d6-9a1c-0b26994f79fd/tempfile.png",
    "https://replicate.delivery/mgxm/209549da-c442-42c1-b96f-e010ae14f07a/tempfile.png",
    "https://replicate.delivery/mgxm/03247e3f-2265-48ef-9915-4a25b131a727/tempfile.png",
    "https://replicate.delivery/mgxm/55ebb21e-006f-4b53-a489-1f968c7d64e6/tempfile.png"
  ],
  "data_removed": false,
  "error": null,
  "source": "web",
  "status": "succeeded",
  "created_at": "2022-03-10T01:16:41.185854Z",
  "started_at": "2022-03-10T01:16:41.403543Z",
  "completed_at": "2022-03-10T01:25:39.824286Z",
  "urls": {
    "cancel": "https://api.replicate.com/v1/predictions/4porzbc32fbv7ezhe7u7tprg5i/cancel",
    "get": "https://api.replicate.com/v1/predictions/4porzbc32fbv7ezhe7u7tprg5i"
  },
  "metrics": {
    "predict_time": 538.420743,
    "total_time": 538.638432
  }
}

Generated in

8 minutes 58 seconds

Tweak it Report

---> BasePixrayPredictor Predict Using seed: 3903845079 Running pixeldrawer with 80x45 grid 0%| | 0.00/244M [00:00<?, ?iB/s] 0%|▏ | 1.19M/244M [00:00<00:21, 11.9MiB/s] 5%|█▉ | 12.6M/244M [00:00<00:14, 16.3MiB/s] 14%|█████▏ | 33.6M/244M [00:00<00:09, 22.6MiB/s] 21%|███████▉ | 50.6M/244M [00:00<00:06, 30.6MiB/s] 27%|██████████▍ | 67.0M/244M [00:00<00:04, 40.6MiB/s] 34%|████████████▊ | 82.4M/244M [00:00<00:03, 52.4MiB/s] 44%|████████████████▉ | 106M/244M [00:00<00:02, 68.7MiB/s] 53%|████████████████████▌ | 128M/244M [00:00<00:01, 87.0MiB/s] 61%|████████████████████████▌ | 149M/244M [00:00<00:00, 106MiB/s] 71%|████████████████████████████▍ | 173M/244M [00:01<00:00, 129MiB/s] 79%|███████████████████████████████▊ | 194M/244M [00:01<00:00, 146MiB/s] 88%|███████████████████████████████████▏ | 214M/244M [00:01<00:00, 161MiB/s] 96%|██████████████████████████████████████▍ | 235M/244M [00:01<00:00, 171MiB/s] 100%|████████████████████████████████████████| 244M/244M [00:01<00:00, 187MiB/s] Loaded CLIP RN50: 224x224 and 102.01M params Loaded CLIP ViT-B/32: 224x224 and 151.28M params Loaded CLIP ViT-B/16: 224x224 and 149.62M params Using device: cuda:0 Optimising using: Adam Using text prompts: ['evil medical device from a post apocalyptic veterinary clinic #pixelart'] using custom losses: smoothness:0.5 0it [00:00, ?it/s] iter: 0, loss: 3.23, losses: 1.01, 0.0859, 0.928, 0.0617, 0.904, 0.0643, 0.1, 0.0725 (-0=>3.227) 0it [00:00, ?it/s]/root/.pyenv/versions/3.8.12/lib/python3.8/site-packages/torch/nn/functional.py:3609: UserWarning: Default upsampling behavior when mode=bilinear is changed to align_corners=False since 0.4.0. Please specify align_corners=True if the old behavior is desired. See the documentation of nn.Upsample for details. warnings.warn( 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 10, loss: 2.88, losses: 0.916, 0.0816, 0.799, 0.061, 0.798, 0.0613, 0.0905, 0.0707 (-0=>2.878) 0it [00:00, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 20, loss: 2.73, losses: 0.874, 0.0823, 0.753, 0.0629, 0.74, 0.0628, 0.0843, 0.0704 (-0=>2.73) 0it [00:01, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 30, loss: 2.62, losses: 0.848, 0.0824, 0.724, 0.0635, 0.685, 0.0644, 0.0787, 0.0711 (-0=>2.617) 0it [00:01, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 40, loss: 2.53, losses: 0.818, 0.0824, 0.695, 0.0662, 0.66, 0.0663, 0.0731, 0.0716 (-3=>2.533) 0it [00:00, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 50, loss: 2.53, losses: 0.815, 0.0837, 0.7, 0.0643, 0.66, 0.0655, 0.0681, 0.0704 (-2=>2.493) 0it [00:00, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 60, loss: 2.46, losses: 0.786, 0.083, 0.676, 0.0673, 0.648, 0.0676, 0.0642, 0.0714 (-1=>2.428) 0it [00:00, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 70, loss: 2.46, losses: 0.781, 0.0844, 0.682, 0.0658, 0.648, 0.0672, 0.0608, 0.0695 (-1=>2.412) 0it [00:00, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 80, loss: 2.46, losses: 0.788, 0.0835, 0.682, 0.0654, 0.645, 0.0668, 0.0579, 0.0716 (-2=>2.376) 0it [00:00, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 90, loss: 2.41, losses: 0.771, 0.0833, 0.665, 0.067, 0.631, 0.0677, 0.0552, 0.0682 (-3=>2.364) 0it [00:00, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 100, loss: 2.35, losses: 0.746, 0.0847, 0.65, 0.067, 0.619, 0.0692, 0.053, 0.0624 (-0=>2.352) 0it [00:00, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 110, loss: 2.35, losses: 0.748, 0.0844, 0.651, 0.067, 0.616, 0.0691, 0.051, 0.0596 (-1=>2.331) 0it [00:00, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 120, loss: 2.36, losses: 0.756, 0.0844, 0.651, 0.0662, 0.618, 0.069, 0.0491, 0.0617 (-11=>2.331) 0it [00:00, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 130, loss: 2.36, losses: 0.756, 0.0839, 0.654, 0.0666, 0.622, 0.0688, 0.0472, 0.061 (-6=>2.31) 0it [00:00, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 140, loss: 2.36, losses: 0.753, 0.0854, 0.652, 0.0665, 0.623, 0.0677, 0.0458, 0.0638 (-6=>2.305) 0it [00:00, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 150, loss: 2.33, losses: 0.741, 0.0847, 0.646, 0.0665, 0.614, 0.0693, 0.0447, 0.0621 (-5=>2.297) 0it [00:00, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 160, loss: 2.31, losses: 0.734, 0.0842, 0.641, 0.0675, 0.609, 0.0699, 0.0437, 0.061 (-15=>2.297) 0it [00:00, ?it/s] 0it [00:15, ?it/s] 0it [00:00, ?it/s] iter: 170, loss: 2.3, losses: 0.736, 0.0851, 0.633, 0.0673, 0.606, 0.0703, 0.0428, 0.0597 (-3=>2.289) 0it [00:00, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 180, loss: 2.31, losses: 0.741, 0.0849, 0.637, 0.0668, 0.61, 0.0689, 0.0418, 0.0622 (-5=>2.28) 0it [00:00, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 190, loss: 2.32, losses: 0.745, 0.0844, 0.642, 0.0659, 0.613, 0.0683, 0.0409, 0.0629 (-15=>2.28) 0it [00:00, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 200, loss: 2.26, losses: 0.716, 0.0857, 0.624, 0.0677, 0.599, 0.0703, 0.0401, 0.0618 (-0=>2.265) 0it [00:01, ?it/s] 0it [00:17, ?it/s] 0it [00:00, ?it/s] iter: 210, loss: 2.33, losses: 0.746, 0.085, 0.646, 0.0659, 0.616, 0.0683, 0.0392, 0.0641 (-1=>2.255) 0it [00:00, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 220, loss: 2.27, losses: 0.724, 0.0844, 0.628, 0.0674, 0.601, 0.0704, 0.0386, 0.0578 (-5=>2.252) Dropping learning rate 0it [00:01, ?it/s] 0it [00:17, ?it/s] 0it [00:00, ?it/s] iter: 230, loss: 2.3, losses: 0.739, 0.085, 0.634, 0.0671, 0.606, 0.0693, 0.0384, 0.0602 (-1=>2.264) 0it [00:01, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 240, loss: 2.26, losses: 0.716, 0.0849, 0.627, 0.0678, 0.599, 0.07, 0.0383, 0.0565 (-5=>2.25) 0it [00:01, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 250, loss: 2.25, losses: 0.718, 0.0857, 0.619, 0.0679, 0.592, 0.0703, 0.0382, 0.056 (-0=>2.247) 0it [00:01, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 260, loss: 2.27, losses: 0.719, 0.0846, 0.628, 0.067, 0.6, 0.0704, 0.0382, 0.0594 (-5=>2.244) 0it [00:01, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 270, loss: 2.27, losses: 0.724, 0.0859, 0.63, 0.0669, 0.598, 0.0703, 0.0381, 0.0555 (-1=>2.241) 0it [00:01, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 280, loss: 2.26, losses: 0.72, 0.0853, 0.625, 0.0674, 0.597, 0.0703, 0.0381, 0.0581 (-11=>2.241) 0it [00:00, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 290, loss: 2.25, losses: 0.716, 0.0855, 0.621, 0.0681, 0.593, 0.0701, 0.0381, 0.0574 (-21=>2.241) 0it [00:00, ?it/s] 0it [00:16, ?it/s] 0it [00:00, ?it/s] iter: 300, finished (-31=>2.241) 0it [00:00, ?it/s] 0it [00:00, ?it/s]