Prediction

Model

dribnet/pixray-api:30a982e4

le76voshdvfjzajwbzhavclkla

Status

Succeeded

Source

Web

Hardware

–

Total duration

–

Created

about 4 years ago by dribnet

Webhook

–

Input

settings: # find palettes at https://lospec.com/palette-list palette: https://lospec.com/palette-list/cl8uds-32x.png filters: lookup prompts: "robots at sunset" quality: better seed: god_mode

{
  "settings": "# find palettes at https://lospec.com/palette-list\npalette: https://lospec.com/palette-list/cl8uds-32x.png\nfilters: lookup\nprompts: \"robots at sunset\"\nquality: better\nseed: god_mode\n"
}

Install Replicate’s Node.js client library:

npm install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=r8_ExK**********************************

This is your API token. Keep it to yourself.

Import and set up the client:

import Replicate from "replicate";

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

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

const output = await replicate.run(
  "dribnet/pixray-api:30a982e43193c50480874788bbddcbb3f58397b74c3a580b8f5d3786ab778ade",
  {
    input: {
      settings: "# find palettes at https://lospec.com/palette-list\npalette: https://lospec.com/palette-list/cl8uds-32x.png\nfilters: lookup\nprompts: \"robots at sunset\"\nquality: better\nseed: god_mode\n"
    }
  }
);

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

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=r8_ExK**********************************

This is your API token. Keep it to yourself.

Import the client:

import replicate

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

output = replicate.run(
    "dribnet/pixray-api:30a982e43193c50480874788bbddcbb3f58397b74c3a580b8f5d3786ab778ade",
    input={
        "settings": "# find palettes at https://lospec.com/palette-list\npalette: https://lospec.com/palette-list/cl8uds-32x.png\nfilters: lookup\nprompts: \"robots at sunset\"\nquality: better\nseed: god_mode\n"
    }
)

# The dribnet/pixray-api 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/dribnet/pixray-api/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_ExK**********************************

This is your API token. Keep it to yourself.

Run dribnet/pixray-api 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": "dribnet/pixray-api:30a982e43193c50480874788bbddcbb3f58397b74c3a580b8f5d3786ab778ade",
    "input": {
      "settings": "# find palettes at https://lospec.com/palette-list\\npalette: https://lospec.com/palette-list/cl8uds-32x.png\\nfilters: lookup\\nprompts: \\"robots at sunset\\"\\nquality: better\\nseed: god_mode\\n"
    }
  }' \
  https://api.replicate.com/v1/predictions

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

Output

{
  "outputType": "object",
  "properties": {
    "file": {
      "outputType": "uri",
      "mimeType": "image/png",
      "inferred": false,
      "outputValue": "https://replicate.delivery/mgxm/d1649fbe-e5cb-4374-871b-1c5f51bbd8d2/tempfile.png",
      "typeMismatch": false
    }
  },
  "inferred": false,
  "outputValue": {
    "file": "https://replicate.delivery/mgxm/d1649fbe-e5cb-4374-871b-1c5f51bbd8d2/tempfile.png"
  },
  "typeMismatch": false
}

15 / 15

{
  "id": "le76voshdvfjzajwbzhavclkla",
  "model": "dribnet/pixray-api",
  "version": "30a982e43193c50480874788bbddcbb3f58397b74c3a580b8f5d3786ab778ade",
  "input": {
    "settings": "# find palettes at https://lospec.com/palette-list\npalette: https://lospec.com/palette-list/cl8uds-32x.png\nfilters: lookup\nprompts: \"robots at sunset\"\nquality: better\nseed: god_mode\n"
  },
  "logs": "---> BasePixrayPredictor Predict\nFound 8 colors in https://lospec.com/palette-list/cl8uds-32x.png\nUsing seed:\n392990954\nWorking with z of shape (1, 256, 16, 16) = 65536 dimensions.\nloaded pretrained LPIPS loss from taming/modules/autoencoder/lpips/vgg.pth\nVQLPIPSWithDiscriminator running with hinge loss.\nRestored from models/vqgan_imagenet_f16_16384.ckpt\ncolor table has 8 entries like [[0.6470588235294118, 0.7176470588235294, 0.8313725490196079], [0.9882352941176471, 0.6901960784313725, 0.5490196078431373], [0.6039215686274509, 0.6705882352941176, 0.788235294117647], [0.5607843137254902, 0.6274509803921569, 0.7490196078431373], [0.9372549019607843, 0.615686274509804, 0.4980392156862745]]\nUsing device:\ncuda:0\nOptimising using:\nAdam\nUsing text prompts:\n['robots at sunset']\n\n0it [00:00, ?it/s]\n/root/.pyenv/versions/3.8.12/lib/python3.8/site-packages/torch/nn/functional.py:3631: 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/root/.pyenv/versions/3.8.12/lib/python3.8/site-packages/torch/functional.py:445: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at  ../aten/src/ATen/native/TensorShape.cpp:2157.)\n  return _VF.meshgrid(tensors, **kwargs)  # type: ignore[attr-defined]\niter: 0, loss: 3.4, losses: 0.395, 0.994, 0.0795, 0.915, 0.047, 0.916, 0.0481 (-0=>3.395)\n\n0it [00:00, ?it/s]\niter: 10, loss: 2.84, losses: 0.0314, 0.945, 0.0797, 0.85, 0.0455, 0.839, 0.047 (-0=>2.838)\n\n0it [00:11, ?it/s]\n\n0it [00:22, ?it/s]\n\n0it [00:00, ?it/s]\niter: 20, loss: 2.8, losses: 0.0259, 0.935, 0.0811, 0.838, 0.0445, 0.831, 0.0458 (-0=>2.802)\n\n0it [00:00, ?it/s]\niter: 30, loss: 2.79, losses: 0.0302, 0.931, 0.0805, 0.835, 0.0446, 0.823, 0.0454 (-2=>2.789)\n\n0it [00:12, ?it/s]\n\n0it [00:22, ?it/s]\n\n0it [00:00, ?it/s]\niter: 40, loss: 2.78, losses: 0.031, 0.926, 0.081, 0.83, 0.0443, 0.82, 0.0458 (-3=>2.76)\n\n0it [00:00, ?it/s]\niter: 50, loss: 2.76, losses: 0.0377, 0.918, 0.0814, 0.825, 0.0437, 0.814, 0.0452 (-13=>2.76)\n\n0it [00:12, ?it/s]\n\n0it [00:22, ?it/s]\n\n0it [00:00, ?it/s]\niter: 60, loss: 2.75, losses: 0.0688, 0.9, 0.0821, 0.813, 0.0428, 0.801, 0.0451 (-6=>2.732)\n\n0it [00:00, ?it/s]\niter: 70, loss: 2.7, losses: 0.0738, 0.879, 0.0798, 0.798, 0.0439, 0.777, 0.0456 (-0=>2.697)\n\n0it [00:12, ?it/s]\n\n0it [00:22, ?it/s]\n\n0it [00:00, ?it/s]\niter: 80, loss: 2.75, losses: 0.138, 0.877, 0.0817, 0.791, 0.0444, 0.776, 0.0466 (-10=>2.697)\n\n0it [00:00, ?it/s]\niter: 90, loss: 2.76, losses: 0.128, 0.893, 0.0822, 0.795, 0.044, 0.772, 0.0458 (-20=>2.697)\n\n0it [00:12, ?it/s]\n\n0it [00:22, ?it/s]\n\n0it [00:00, ?it/s]\niter: 100, loss: 2.74, losses: 0.143, 0.88, 0.0824, 0.783, 0.0459, 0.759, 0.0481 (-6=>2.672)\n\n0it [00:00, ?it/s]\niter: 110, loss: 2.71, losses: 0.116, 0.874, 0.0821, 0.782, 0.0443, 0.762, 0.047 (-16=>2.672)\n\n0it [00:12, ?it/s]\n\n0it [00:22, ?it/s]\n\n0it [00:00, ?it/s]\niter: 120, loss: 2.72, losses: 0.147, 0.862, 0.0833, 0.778, 0.0459, 0.755, 0.0483 (-4=>2.661)\n\n0it [00:00, ?it/s]\niter: 130, loss: 2.65, losses: 0.152, 0.843, 0.0834, 0.744, 0.0468, 0.734, 0.0485 (-6=>2.652)\n\n0it [00:12, ?it/s]\n\n0it [00:22, ?it/s]\n\n0it [00:00, ?it/s]\niter: 140, loss: 2.69, losses: 0.145, 0.852, 0.0822, 0.767, 0.0457, 0.748, 0.0485 (-6=>2.633)\n\n0it [00:00, ?it/s]\niter: 150, loss: 2.68, losses: 0.173, 0.849, 0.0832, 0.749, 0.046, 0.738, 0.047 (-8=>2.628)\n\n0it [00:12, ?it/s]\n\n0it [00:22, ?it/s]\n\n0it [00:00, ?it/s]\niter: 160, loss: 2.73, losses: 0.17, 0.859, 0.0837, 0.773, 0.0461, 0.751, 0.0474 (-18=>2.628)\n\n0it [00:00, ?it/s]\niter: 170, loss: 2.73, losses: 0.166, 0.87, 0.0834, 0.765, 0.0466, 0.756, 0.0472 (-28=>2.628)\n\n0it [00:12, ?it/s]\n\n0it [00:22, ?it/s]\n\n0it [00:00, ?it/s]\niter: 180, loss: 2.68, losses: 0.164, 0.848, 0.0844, 0.744, 0.0476, 0.746, 0.0485 (-38=>2.628)\n\n0it [00:00, ?it/s]\niter: 190, loss: 2.6, losses: 0.156, 0.818, 0.0838, 0.727, 0.0475, 0.72, 0.0484 (-0=>2.6)\n\n0it [00:12, ?it/s]\n\n0it [00:22, ?it/s]\n\n0it [00:00, ?it/s]\niter: 200, loss: 2.7, losses: 0.177, 0.851, 0.0837, 0.752, 0.047, 0.739, 0.0484 (-10=>2.6)\n\n0it [00:00, ?it/s]\niter: 210, loss: 2.7, losses: 0.175, 0.85, 0.0843, 0.751, 0.0473, 0.746, 0.0487 (-20=>2.6)\n\n0it [00:12, ?it/s]\n\n0it [00:22, ?it/s]\n\n0it [00:00, ?it/s]\niter: 220, loss: 2.7, losses: 0.166, 0.858, 0.0853, 0.751, 0.0464, 0.743, 0.0481 (-30=>2.6)\n\n0it [00:00, ?it/s]\nDropping learning rate\niter: 230, loss: 2.6, losses: 0.153, 0.824, 0.0858, 0.722, 0.048, 0.717, 0.0488 (-0=>2.599)\n\n0it [00:12, ?it/s]\n\n0it [00:22, ?it/s]\n\n0it [00:00, ?it/s]\niter: 240, loss: 2.65, losses: 0.149, 0.843, 0.0851, 0.74, 0.0471, 0.738, 0.0493 (-1=>2.574)\n\n0it [00:00, ?it/s]\niter: 250, loss: 2.65, losses: 0.151, 0.842, 0.0842, 0.744, 0.0474, 0.736, 0.0486 (-11=>2.574)\n\n0it [00:12, ?it/s]\n\n0it [00:22, ?it/s]\n\n0it [00:00, ?it/s]\niter: 260, loss: 2.64, losses: 0.154, 0.836, 0.0851, 0.737, 0.0476, 0.731, 0.0481 (-21=>2.574)\n\n0it [00:00, ?it/s]\niter: 270, loss: 2.65, losses: 0.145, 0.843, 0.0853, 0.742, 0.047, 0.735, 0.0482 (-4=>2.562)\n\n0it [00:12, ?it/s]\n\n0it [00:22, ?it/s]\n\n0it [00:00, ?it/s]\niter: 280, loss: 2.61, losses: 0.16, 0.825, 0.0853, 0.721, 0.049, 0.716, 0.049 (-14=>2.562)\n\n0it [00:00, ?it/s]\niter: 290, loss: 2.64, losses: 0.155, 0.835, 0.0849, 0.735, 0.0479, 0.734, 0.0484 (-24=>2.562)\n\n0it [00:12, ?it/s]\n\n0it [00:22, ?it/s]\n\n0it [00:00, ?it/s]\niter: 300, finished (-34=>2.562)\n\n0it [00:00, ?it/s]\n\n0it [00:00, ?it/s]",
  "output": [
    {
      "file": "https://replicate.delivery/mgxm/908a2b66-f39c-45b5-80e0-2c3bd1eb89e6/tempfile.png"
    },
    {
      "file": "https://replicate.delivery/mgxm/c1639b8a-8d93-47b3-b99a-154f64effefc/tempfile.png"
    },
    {
      "file": "https://replicate.delivery/mgxm/0c1f6a24-8065-4f0e-959c-39a33d9c1ac3/tempfile.png"
    },
    {
      "file": "https://replicate.delivery/mgxm/4ecde433-c878-47b9-8d34-2208a1fcddfe/tempfile.png"
    },
    {
      "file": "https://replicate.delivery/mgxm/ef7a3e70-62a2-4a97-965a-2c458d952f60/tempfile.png"
    },
    {
      "file": "https://replicate.delivery/mgxm/ec99a125-29cd-4bd1-ad8c-efd52cdb37bd/tempfile.png"
    },
    {
      "file": "https://replicate.delivery/mgxm/cb02501c-13a7-4886-a130-93893136bc0b/tempfile.png"
    },
    {
      "file": "https://replicate.delivery/mgxm/f37939e4-4f3e-4fe9-987a-655259370aac/tempfile.png"
    },
    {
      "file": "https://replicate.delivery/mgxm/194d00ed-685e-40a6-befc-b4e5634ea7a9/tempfile.png"
    },
    {
      "file": "https://replicate.delivery/mgxm/5d69d8ad-6c23-4217-8d6a-3e4d298b3e8a/tempfile.png"
    },
    {
      "file": "https://replicate.delivery/mgxm/556c728b-e8f0-45cc-abf1-593571e6fbf7/tempfile.png"
    },
    {
      "file": "https://replicate.delivery/mgxm/5d4ba414-7230-4344-aeff-380712b9ae0c/tempfile.png"
    },
    {
      "file": "https://replicate.delivery/mgxm/e86c93b8-644c-48ac-a047-bc586d664e5e/tempfile.png"
    },
    {
      "file": "https://replicate.delivery/mgxm/fd4faa26-6bee-4f32-aeb9-30ef070c854c/tempfile.png"
    },
    {
      "file": "https://replicate.delivery/mgxm/d1649fbe-e5cb-4374-871b-1c5f51bbd8d2/tempfile.png"
    }
  ],
  "data_removed": false,
  "error": null,
  "source": "web",
  "status": "succeeded",
  "created_at": "2021-11-28T10:00:49.325906Z",
  "started_at": "2021-11-30T22:00:25.336971Z",
  "completed_at": "2021-11-28T10:06:55.200045Z",
  "urls": {
    "cancel": "https://api.replicate.com/v1/predictions/le76voshdvfjzajwbzhavclkla/cancel",
    "get": "https://api.replicate.com/v1/predictions/le76voshdvfjzajwbzhavclkla"
  },
  "metrics": {
    "total_time": 365.874139
  }
}

Tweak it Report

---> BasePixrayPredictor Predict Found 8 colors in https://lospec.com/palette-list/cl8uds-32x.png Using seed: 392990954 Working with z of shape (1, 256, 16, 16) = 65536 dimensions. loaded pretrained LPIPS loss from taming/modules/autoencoder/lpips/vgg.pth VQLPIPSWithDiscriminator running with hinge loss. Restored from models/vqgan_imagenet_f16_16384.ckpt color table has 8 entries like [[0.6470588235294118, 0.7176470588235294, 0.8313725490196079], [0.9882352941176471, 0.6901960784313725, 0.5490196078431373], [0.6039215686274509, 0.6705882352941176, 0.788235294117647], [0.5607843137254902, 0.6274509803921569, 0.7490196078431373], [0.9372549019607843, 0.615686274509804, 0.4980392156862745]] Using device: cuda:0 Optimising using: Adam Using text prompts: ['robots at sunset'] 0it [00:00, ?it/s] /root/.pyenv/versions/3.8.12/lib/python3.8/site-packages/torch/nn/functional.py:3631: 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( /root/.pyenv/versions/3.8.12/lib/python3.8/site-packages/torch/functional.py:445: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at ../aten/src/ATen/native/TensorShape.cpp:2157.) return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] iter: 0, loss: 3.4, losses: 0.395, 0.994, 0.0795, 0.915, 0.047, 0.916, 0.0481 (-0=>3.395) 0it [00:00, ?it/s] iter: 10, loss: 2.84, losses: 0.0314, 0.945, 0.0797, 0.85, 0.0455, 0.839, 0.047 (-0=>2.838) 0it [00:11, ?it/s] 0it [00:22, ?it/s] 0it [00:00, ?it/s] iter: 20, loss: 2.8, losses: 0.0259, 0.935, 0.0811, 0.838, 0.0445, 0.831, 0.0458 (-0=>2.802) 0it [00:00, ?it/s] iter: 30, loss: 2.79, losses: 0.0302, 0.931, 0.0805, 0.835, 0.0446, 0.823, 0.0454 (-2=>2.789) 0it [00:12, ?it/s] 0it [00:22, ?it/s] 0it [00:00, ?it/s] iter: 40, loss: 2.78, losses: 0.031, 0.926, 0.081, 0.83, 0.0443, 0.82, 0.0458 (-3=>2.76) 0it [00:00, ?it/s] iter: 50, loss: 2.76, losses: 0.0377, 0.918, 0.0814, 0.825, 0.0437, 0.814, 0.0452 (-13=>2.76) 0it [00:12, ?it/s] 0it [00:22, ?it/s] 0it [00:00, ?it/s] iter: 60, loss: 2.75, losses: 0.0688, 0.9, 0.0821, 0.813, 0.0428, 0.801, 0.0451 (-6=>2.732) 0it [00:00, ?it/s] iter: 70, loss: 2.7, losses: 0.0738, 0.879, 0.0798, 0.798, 0.0439, 0.777, 0.0456 (-0=>2.697) 0it [00:12, ?it/s] 0it [00:22, ?it/s] 0it [00:00, ?it/s] iter: 80, loss: 2.75, losses: 0.138, 0.877, 0.0817, 0.791, 0.0444, 0.776, 0.0466 (-10=>2.697) 0it [00:00, ?it/s] iter: 90, loss: 2.76, losses: 0.128, 0.893, 0.0822, 0.795, 0.044, 0.772, 0.0458 (-20=>2.697) 0it [00:12, ?it/s] 0it [00:22, ?it/s] 0it [00:00, ?it/s] iter: 100, loss: 2.74, losses: 0.143, 0.88, 0.0824, 0.783, 0.0459, 0.759, 0.0481 (-6=>2.672) 0it [00:00, ?it/s] iter: 110, loss: 2.71, losses: 0.116, 0.874, 0.0821, 0.782, 0.0443, 0.762, 0.047 (-16=>2.672) 0it [00:12, ?it/s] 0it [00:22, ?it/s] 0it [00:00, ?it/s] iter: 120, loss: 2.72, losses: 0.147, 0.862, 0.0833, 0.778, 0.0459, 0.755, 0.0483 (-4=>2.661) 0it [00:00, ?it/s] iter: 130, loss: 2.65, losses: 0.152, 0.843, 0.0834, 0.744, 0.0468, 0.734, 0.0485 (-6=>2.652) 0it [00:12, ?it/s] 0it [00:22, ?it/s] 0it [00:00, ?it/s] iter: 140, loss: 2.69, losses: 0.145, 0.852, 0.0822, 0.767, 0.0457, 0.748, 0.0485 (-6=>2.633) 0it [00:00, ?it/s] iter: 150, loss: 2.68, losses: 0.173, 0.849, 0.0832, 0.749, 0.046, 0.738, 0.047 (-8=>2.628) 0it [00:12, ?it/s] 0it [00:22, ?it/s] 0it [00:00, ?it/s] iter: 160, loss: 2.73, losses: 0.17, 0.859, 0.0837, 0.773, 0.0461, 0.751, 0.0474 (-18=>2.628) 0it [00:00, ?it/s] iter: 170, loss: 2.73, losses: 0.166, 0.87, 0.0834, 0.765, 0.0466, 0.756, 0.0472 (-28=>2.628) 0it [00:12, ?it/s] 0it [00:22, ?it/s] 0it [00:00, ?it/s] iter: 180, loss: 2.68, losses: 0.164, 0.848, 0.0844, 0.744, 0.0476, 0.746, 0.0485 (-38=>2.628) 0it [00:00, ?it/s] iter: 190, loss: 2.6, losses: 0.156, 0.818, 0.0838, 0.727, 0.0475, 0.72, 0.0484 (-0=>2.6) 0it [00:12, ?it/s] 0it [00:22, ?it/s] 0it [00:00, ?it/s] iter: 200, loss: 2.7, losses: 0.177, 0.851, 0.0837, 0.752, 0.047, 0.739, 0.0484 (-10=>2.6) 0it [00:00, ?it/s] iter: 210, loss: 2.7, losses: 0.175, 0.85, 0.0843, 0.751, 0.0473, 0.746, 0.0487 (-20=>2.6) 0it [00:12, ?it/s] 0it [00:22, ?it/s] 0it [00:00, ?it/s] iter: 220, loss: 2.7, losses: 0.166, 0.858, 0.0853, 0.751, 0.0464, 0.743, 0.0481 (-30=>2.6) 0it [00:00, ?it/s] Dropping learning rate iter: 230, loss: 2.6, losses: 0.153, 0.824, 0.0858, 0.722, 0.048, 0.717, 0.0488 (-0=>2.599) 0it [00:12, ?it/s] 0it [00:22, ?it/s] 0it [00:00, ?it/s] iter: 240, loss: 2.65, losses: 0.149, 0.843, 0.0851, 0.74, 0.0471, 0.738, 0.0493 (-1=>2.574) 0it [00:00, ?it/s] iter: 250, loss: 2.65, losses: 0.151, 0.842, 0.0842, 0.744, 0.0474, 0.736, 0.0486 (-11=>2.574) 0it [00:12, ?it/s] 0it [00:22, ?it/s] 0it [00:00, ?it/s] iter: 260, loss: 2.64, losses: 0.154, 0.836, 0.0851, 0.737, 0.0476, 0.731, 0.0481 (-21=>2.574) 0it [00:00, ?it/s] iter: 270, loss: 2.65, losses: 0.145, 0.843, 0.0853, 0.742, 0.047, 0.735, 0.0482 (-4=>2.562) 0it [00:12, ?it/s] 0it [00:22, ?it/s] 0it [00:00, ?it/s] iter: 280, loss: 2.61, losses: 0.16, 0.825, 0.0853, 0.721, 0.049, 0.716, 0.049 (-14=>2.562) 0it [00:00, ?it/s] iter: 290, loss: 2.64, losses: 0.155, 0.835, 0.0849, 0.735, 0.0479, 0.734, 0.0484 (-24=>2.562) 0it [00:12, ?it/s] 0it [00:22, ?it/s] 0it [00:00, ?it/s] iter: 300, finished (-34=>2.562) 0it [00:00, ?it/s] 0it [00:00, ?it/s]