search

huggingface / pytorch-pretrained-BigGAN

🦋A PyTorch implementation of BigGAN with pretrained weights and conversion scripts.
MIT License
1.02k stars 180 forks source link

Output does not match TF implementation #3

Closed darknoon closed 5 years ago

darknoon commented 5 years ago

First, I want to thank you for doing this! There are a lot of little pieces to get right and the results so far are pretty amazing.

That said, images are quite different from the Tensorflow implementation. I'm interested in using this with points found with Ganbreeder, which means that I want to match the TF version as closely as possible. I think it's all using cuDNN underneath right? If so, I imagine we could get this closer aligned.

My main question is how to modify the Tensorflow graph to output the intermediate calculations so we can compare activations and figure out which layer(s) are causing the biggest error. I would be happy you with this.

Examples:

TF pytorch URL
https://ganbreeder.app/info?k=ff84584479eb1cc90e4af4a4
Unknown-56 https://ganbreeder.app/info?k=e1a780eba7bd551dfeb43789
Unknown-57 https://ganbreeder.app/info?k=42999d4b4849f0c0852e6ecd
darknoon commented 5 years ago

Just realized I'm an idiot, this model is BigGAN-deep, not BigGAN. Will redo matching to see if it corresponds better. Sorry for the noise.