EugenHotaj/pytorch-generative

Issues

• how to use multiple GPUs

  #36 opened 2 years ago by Eureka-Y
  1

• How to get the labels of generated images

  #37 opened 2 years ago by Eureka-Y
  1

• [Q] Need an example to show how to generate an image.

  #35 opened 2 years ago by showmin
  13

• Create Lightweight Tests for Important Notebooks

  #14 opened 2 years ago by EugenHotaj
  1

• Add PixelCNN++ improvements to GatedPixelCNN

  #7 opened 2 years ago by EugenHotaj
  1

• How does generation work for autoregressive models?

  #34 opened 2 years ago by YHYeooooong
  2

• Typo in README

  #33 opened 3 years ago by qgallouedec
  1

• Update KDE models to handle multi-dimensional inputs

  #32 opened 3 years ago by qgallouedec
  4

• Are pretrained weights for any of these models in repo available anywhere?

  #29 opened 3 years ago by ethancaballero
  1

• VectorQuantizedVAE2

  #30 opened 3 years ago by adammoss
  1

• Positional encoding is replicated across all channels.

  #31 opened 3 years ago by mattiasxu
  1

• Sampling on 3 channels looks corrupted

  #20 opened 3 years ago by eyalbetzalel
  2

• Applying Style Transfer to a GIF

  #25 opened 3 years ago by thistleknot
  2

• How to view progress during training?

  #28 opened 3 years ago by AlexanderGroeger
  2

• Add unit tests for library code

  #5 opened 4 years ago by EugenHotaj
  1

• How can I extract the average style from a set of images instead of a single image?

  #1 opened 4 years ago by hadaev8
  2

• How to run the style_transfer notebook without square images

  #27 opened 4 years ago by RandallJEllis
  6

• Add slider bars for the different parameters in Neural Style Transfer

  #24 opened 4 years ago by thistleknot
  4

• Replicating NLL Results for PIXEL CNN

  #21 opened 4 years ago by anton-jeran
  2

• Look Into Adding Support For Weights & Biases Logging

  #8 opened 4 years ago by EugenHotaj
  1

• Sampling via py scripts

  #16 opened 4 years ago by eyalbetzalel
  4

• Sample Images During Training to Check Generation Quality

  #17 opened 4 years ago by EugenHotaj
  0

• Training ImageGPT on 64 * 64 size images

  #22 opened 4 years ago by ysig
  2

• Using Multiple Style Images in Neural Style Transfer

  #23 opened 4 years ago by thistleknot
  4

• Trainer is (Probably) Slow Due to Excessive Logging

  #10 opened 4 years ago by EugenHotaj
  1

• train.py file is gone along with others

  #19 opened 4 years ago by eyalbetzalel
  5

• Update trainer to use max_steps instead of epochs

  #3 opened 4 years ago by EugenHotaj
  1

• Loading 3 Channels Images to PixelSnail

  #18 opened 4 years ago by eyalbetzalel
  3

• Get Rid of Most Colab Notebooks

  #12 opened 4 years ago by EugenHotaj
  1

• PixelSnail model fails on GPU

  #15 opened 4 years ago by eyalbetzalel
  4

• Style transfer notebook removed / not functional

  #13 opened 4 years ago by ioannist
  2

• Update Readme

  #9 opened 4 years ago by EugenHotaj
  1

• Move code out of colab notebooks and into modules.

  #4 opened 4 years ago by EugenHotaj
  1

• Tune tqdm logging output

  #6 opened 4 years ago by EugenHotaj
  1

• Create a Trainer object to encapsulate the training functions

  #2 opened 4 years ago by EugenHotaj
  0