dribnet/pixray 📝 → 🖼️
About
Pixray with custom settings
Example Output
Output
[object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object]
Performance Metrics
1778.85s
Prediction Time
1809.46s
Total Time
All Input Parameters
{
"prompts": "Aliens destroying NYC skyline with lasers. #pixelart",
"settings": "drawer: pixel\nquality: better\nsize: [1080, 360]\npixel_size: [270, 90]"
}
Input Parameters
- prompts
- text prompt
- settings
- yaml settings
Output Schema
Output
Example Execution Logs
---> BasePixrayPredictor Predict
Using seed:
12913612667127894451
Running pixeldrawer with 270x90 grid
Using device:
cuda:0
Optimising using:
Adam
Using text prompts:
['Aliens destroying NYC skyline with lasers. #pixelart']
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(
iter: 0, loss: 2.99, losses: 0.997, 0.08, 0.895, 0.0461, 0.922, 0.0469 (-0=>2.988)
0it [00:03, ?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(
iter: 10, loss: 2.82, losses: 0.948, 0.0798, 0.849, 0.0451, 0.852, 0.0459 (-0=>2.819)
0it [01:10, ?it/s]
0it [02:12, ?it/s]
0it [00:00, ?it/s]
iter: 20, loss: 2.52, losses: 0.856, 0.0824, 0.742, 0.0496, 0.738, 0.0478 (-0=>2.516)
0it [00:03, ?it/s]
Caught SIGTERM, exiting...
iter: 30, loss: 2.5, losses: 0.846, 0.0853, 0.736, 0.0494, 0.738, 0.0471 (-1=>2.501)
0it [01:09, ?it/s]
0it [02:10, ?it/s]
0it [00:00, ?it/s]
iter: 40, loss: 2.34, losses: 0.789, 0.0855, 0.689, 0.0514, 0.679, 0.0488 (-0=>2.343)
0it [00:03, ?it/s]
iter: 50, loss: 2.43, losses: 0.829, 0.0844, 0.708, 0.0496, 0.714, 0.0479 (-4=>2.303)
0it [01:09, ?it/s]
0it [02:10, ?it/s]
0it [00:00, ?it/s]
iter: 60, loss: 2.23, losses: 0.752, 0.0878, 0.645, 0.0533, 0.64, 0.0499 (-0=>2.229)
0it [00:03, ?it/s]
iter: 70, loss: 2.2, losses: 0.749, 0.0868, 0.633, 0.0538, 0.631, 0.0511 (-0=>2.205)
0it [01:09, ?it/s]
0it [02:10, ?it/s]
0it [00:00, ?it/s]
iter: 80, loss: 2.29, losses: 0.785, 0.0865, 0.659, 0.0522, 0.657, 0.0502 (-10=>2.205)
0it [00:03, ?it/s]
---> BasePixrayPredictor Predict
Using seed:
706433100638245275
Running pixeldrawer with 270x90 grid
Using device:
cuda:0
Optimising using:
Adam
Using text prompts:
['Aliens destroying NYC skyline with lasers. #pixelart']
0it [00:00, ?it/s]
iter: 90, loss: 2.15, losses: 0.732, 0.0872, 0.609, 0.0557, 0.611, 0.0518 (-0=>2.147)
0it [01:08, ?it/s]
iter: 0, loss: 2.97, losses: 0.983, 0.0807, 0.892, 0.0469, 0.919, 0.0483 (-0=>2.97)
0it [00:03, ?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 [02:10, ?it/s]
0it [00:00, ?it/s]
iter: 10, loss: 2.79, losses: 0.944, 0.0792, 0.839, 0.0465, 0.838, 0.0459 (-1=>2.783)
0it [01:04, ?it/s]
iter: 100, loss: 2.23, losses: 0.764, 0.0864, 0.635, 0.0542, 0.637, 0.0522 (-2=>2.123)
0it [00:03, ?it/s]
0it [02:02, ?it/s]
0it [00:00, ?it/s]
iter: 20, loss: 2.6, losses: 0.892, 0.0807, 0.767, 0.0481, 0.763, 0.0475 (-2=>2.566)
0it [00:03, ?it/s]
iter: 110, loss: 2.21, losses: 0.755, 0.0875, 0.63, 0.0544, 0.631, 0.0521 (-6=>2.119)
0it [01:09, ?it/s]
iter: 30, loss: 2.55, losses: 0.868, 0.082, 0.749, 0.0481, 0.755, 0.0468 (-1=>2.459)
0it [01:04, ?it/s]
0it [02:10, ?it/s]
0it [00:00, ?it/s]
iter: 120, loss: 2.11, losses: 0.715, 0.0885, 0.595, 0.0562, 0.603, 0.0529 (-2=>2.084)
0it [00:03, ?it/s]
0it [02:01, ?it/s]
0it [00:00, ?it/s]
iter: 40, loss: 2.43, losses: 0.827, 0.0839, 0.714, 0.0503, 0.71, 0.048 (-2=>2.359)
0it [00:03, ?it/s]
iter: 130, loss: 2.26, losses: 0.773, 0.087, 0.648, 0.0538, 0.647, 0.0524 (-4=>2.079)
0it [01:08, ?it/s]
iter: 50, loss: 2.43, losses: 0.829, 0.0831, 0.709, 0.0501, 0.707, 0.0487 (-3=>2.319)
0it [01:03, ?it/s]
0it [02:09, ?it/s]
0it [00:00, ?it/s]
iter: 140, loss: 2.06, losses: 0.705, 0.0889, 0.58, 0.0574, 0.579, 0.0546 (-0=>2.064)
0it [00:03, ?it/s]
0it [02:00, ?it/s]
0it [00:00, ?it/s]
iter: 60, loss: 2.36, losses: 0.808, 0.0837, 0.686, 0.051, 0.687, 0.0491 (-8=>2.288)
0it [00:03, ?it/s]
iter: 150, loss: 2.17, losses: 0.744, 0.0877, 0.613, 0.0556, 0.617, 0.0532 (-10=>2.064)
0it [01:08, ?it/s]
iter: 70, loss: 2.21, losses: 0.746, 0.0862, 0.639, 0.0547, 0.637, 0.0514 (-0=>2.214)
0it [01:03, ?it/s]
0it [02:09, ?it/s]
0it [00:00, ?it/s]
iter: 160, loss: 2.17, losses: 0.743, 0.0876, 0.615, 0.0558, 0.614, 0.0534 (-20=>2.064)
0it [00:03, ?it/s]
0it [02:00, ?it/s]
0it [00:00, ?it/s]
iter: 80, loss: 2.27, losses: 0.775, 0.0863, 0.653, 0.0532, 0.651, 0.051 (-4=>2.204)
0it [00:03, ?it/s]
iter: 170, loss: 2.07, losses: 0.699, 0.089, 0.581, 0.0576, 0.585, 0.0546 (-4=>2.064)
0it [01:08, ?it/s]
---> BasePixrayPredictor Predict
Using seed:
11971428367999728664
Running pixeldrawer with 270x90 grid
iter: 90, loss: 2.21, losses: 0.755, 0.0871, 0.636, 0.0546, 0.628, 0.0525 (-5=>2.192)
0it [01:03, ?it/s]
Caught SIGTERM, exiting...
0it [02:09, ?it/s]
0it [00:00, ?it/s]
iter: 180, loss: 2.14, losses: 0.733, 0.0882, 0.606, 0.0564, 0.604, 0.0538 (-14=>2.064)
0it [00:03, ?it/s]
0it [02:00, ?it/s]
0it [00:00, ?it/s]
Using device:
cuda:0
Optimising using:
Adam
Using text prompts:
['Aliens destroying NYC skyline with lasers. #pixelart']
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(
iter: 100, loss: 2.2, losses: 0.749, 0.0873, 0.632, 0.0538, 0.63, 0.0524 (-3=>2.13)
0it [00:03, ?it/s]
iter: 0, loss: 3, losses: 0.999, 0.0801, 0.897, 0.0461, 0.927, 0.0473 (-0=>2.996)
0it [00:03, ?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(
iter: 190, loss: 2.04, losses: 0.694, 0.0887, 0.57, 0.0584, 0.572, 0.0553 (-0=>2.039)
0it [01:07, ?it/s]
iter: 110, loss: 2.2, losses: 0.754, 0.0858, 0.625, 0.0549, 0.626, 0.0525 (-13=>2.13)
0it [01:03, ?it/s]
iter: 10, loss: 2.81, losses: 0.938, 0.0792, 0.85, 0.0464, 0.853, 0.047 (-0=>2.813)
0it [01:09, ?it/s]
0it [02:08, ?it/s]
0it [00:00, ?it/s]
iter: 200, loss: 2.2, losses: 0.756, 0.0865, 0.624, 0.0542, 0.626, 0.0528 (-10=>2.039)
0it [00:03, ?it/s]
0it [02:00, ?it/s]
0it [00:00, ?it/s]
iter: 120, loss: 2.21, losses: 0.761, 0.0852, 0.627, 0.0541, 0.629, 0.0519 (-2=>2.11)
0it [00:03, ?it/s]
0it [02:08, ?it/s]
0it [00:00, ?it/s]
iter: 20, loss: 2.69, losses: 0.922, 0.0796, 0.8, 0.0459, 0.797, 0.0464 (-1=>2.666)
0it [00:03, ?it/s]
iter: 210, loss: 2.14, losses: 0.733, 0.0869, 0.607, 0.0558, 0.598, 0.054 (-20=>2.039)
0it [01:08, ?it/s]
iter: 130, loss: 2.17, losses: 0.739, 0.0867, 0.619, 0.0555, 0.619, 0.0536 (-12=>2.11)
0it [01:03, ?it/s]
iter: 30, loss: 2.52, losses: 0.858, 0.0828, 0.736, 0.0484, 0.747, 0.0471 (-2=>2.441)
0it [01:05, ?it/s]
0it [02:09, ?it/s]
0it [00:00, ?it/s]
iter: 220, loss: 2.12, losses: 0.722, 0.0878, 0.6, 0.0556, 0.596, 0.0542 (-9=>2.032)
0it [00:03, ?it/s]
0it [01:59, ?it/s]
0it [00:00, ?it/s]
iter: 140, loss: 2.1, losses: 0.714, 0.0866, 0.595, 0.0563, 0.595, 0.0546 (-4=>2.09)
0it [00:03, ?it/s]
0it [02:06, ?it/s]
0it [00:00, ?it/s]
iter: 40, loss: 2.35, losses: 0.794, 0.0855, 0.691, 0.0508, 0.68, 0.0482 (-0=>2.349)
0it [00:03, ?it/s]
Dropping learning rate
iter: 150, loss: 2.07, losses: 0.702, 0.0861, 0.587, 0.0578, 0.584, 0.0554 (-0=>2.073)
0it [01:03, ?it/s]
iter: 230, loss: 2.15, losses: 0.737, 0.0872, 0.608, 0.0558, 0.604, 0.0545 (-2=>2.03)
0it [01:09, ?it/s]
iter: 50, loss: 2.41, losses: 0.818, 0.0858, 0.702, 0.0505, 0.703, 0.0482 (-4=>2.301)
0it [01:06, ?it/s]
0it [02:00, ?it/s]
0it [00:00, ?it/s]
iter: 160, loss: 2.18, losses: 0.749, 0.0849, 0.621, 0.0552, 0.62, 0.0527 (-2=>2.062)
0it [00:03, ?it/s]
0it [02:10, ?it/s]
0it [00:00, ?it/s]
iter: 240, loss: 2.12, losses: 0.724, 0.0878, 0.598, 0.057, 0.596, 0.0547 (-6=>2.018)
0it [00:03, ?it/s]
0it [02:04, ?it/s]
0it [00:00, ?it/s]
iter: 60, loss: 2.35, losses: 0.806, 0.0878, 0.676, 0.0512, 0.679, 0.0494 (-2=>2.25)
0it [00:03, ?it/s]
iter: 170, loss: 2.19, losses: 0.753, 0.0857, 0.623, 0.0544, 0.623, 0.0528 (-12=>2.062)
0it [01:03, ?it/s]
iter: 250, loss: 2.15, losses: 0.734, 0.0869, 0.609, 0.0556, 0.605, 0.0541 (-2=>2.017)
0it [01:09, ?it/s]
iter: 70, loss: 2.33, losses: 0.799, 0.0856, 0.676, 0.0512, 0.673, 0.0499 (-5=>2.235)
0it [01:05, ?it/s]
0it [02:00, ?it/s]
0it [00:00, ?it/s]
iter: 180, loss: 2.2, losses: 0.754, 0.0856, 0.627, 0.0545, 0.627, 0.0529 (-6=>2.057)
0it [00:03, ?it/s]
0it [02:10, ?it/s]
0it [00:00, ?it/s]
iter: 260, loss: 2.12, losses: 0.722, 0.0886, 0.599, 0.0566, 0.594, 0.0547 (-12=>2.017)
0it [00:03, ?it/s]
0it [02:06, ?it/s]
0it [00:00, ?it/s]
iter: 80, loss: 2.28, losses: 0.779, 0.086, 0.657, 0.0523, 0.654, 0.0498 (-8=>2.208)
0it [00:03, ?it/s]
0it [00:38, ?it/s]
Traceback (most recent call last):
File "/root/.pyenv/versions/3.8.12/lib/python3.8/site-packages/cog/server/redis_queue.py", line 163, in start
self.handle_message(response_queue, message, cleanup_functions)
File "/root/.pyenv/versions/3.8.12/lib/python3.8/site-packages/cog/server/redis_queue.py", line 232, in handle_message
result = next(return_value)
File "/src/cogrun.py", line 90, in predict
yield from super().predict(settings="pixrayraw", prompts=prompts, **ydict)
File "/src/cogrun.py", line 48, in predict
run_complete = pixray.do_run(settings, return_display=True)
File "/src/pixray.py", line 1492, in do_run
keep_going = train(args, cur_iteration)
File "/src/pixray.py", line 1368, in train
opt.step()
File "/root/.pyenv/versions/3.8.12/lib/python3.8/site-packages/torch/optim/optimizer.py", line 88, in wrapper
return func(*args, **kwargs)
File "/root/.pyenv/versions/3.8.12/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 28, in decorate_context
return func(*args, **kwargs)
File "/root/.pyenv/versions/3.8.12/lib/python3.8/site-packages/torch/optim/adam.py", line 103, in step
state['step'] += 1
File "/root/.pyenv/versions/3.8.12/lib/python3.8/site-packages/cog/server/redis_queue.py", line 35, in handle_timeout
raise TimeoutError(self.error_message)
TimeoutError: Prediction timed out
Version Details
- Version ID
f72145829507ace382fad7f3c22bd19df152f00005a3e678c652ab052dcd5005- Version Created
- October 27, 2022