Skip to content

Hsintien-Ng/idinvert_pytorch-reproduced

Repository files navigation

idinvert_pytorch-reproduced

Our project is an edited version of https://github.com/genforce/idinvert_pytorch

In this project, we reproduce the encoder training according to the tf version (https://github.com/genforce/idinvert) with celeba-hq dataset. We add a discriminator including models/base_discriminator.py, models/stylegan_discriminator_network.py, models/stylegan_discriminator.py, and the encoder training including training/training_loop_encoder.py, train_encoder.py.

In order to achieve a parallel training, we also edit some lines in models/base_module.py, models/base_generator.py, models/base_encoder.py.

Since I reproduced it based on my own knowledge, there may be some areas where my understanding is not in place. Welcome everyone to correct me.

About

We add the discriminator and encoder training in https://github.com/genforce/idinvert_pytorch

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages