张宏伦

how to sample the 20w images from EmotioNet

4

hello, thanks for your wonderful work. I have obtained the 95w images of EmotioNet, extracted the AUs with OpenFace and cropped the faces with face_recognition. The problem is, how to...

different update frequencies of different modules

2

Hi~ thanks a lot for your awesome work! I find that during each iteration, D is optimized for **twice**: ``` d_loss, d_losses_latent = compute_d_loss( nets, args, x_real, y_org, y_trg, z_trg=z_trg,...

Comparisons between PATN and PATN_Fine

1

Hello. Thanks for your awesome work. I found that the generator of PATN uses instance norm, while that of PATN_Fine uses batch norm (models/networks.py L119 define_G). What's more, PATN uses...

MAMP中无法使用phpMyAdmin

2

可能是phpMyAdmin和PHP版本不兼容，可以尝试在MAMP设置中更改PHP版本

我提供的豆瓣电影数据某些行存在缺失值，所以有的行的列数不一样，在mac下会自动处理，但是在windows下会报错。 可以尝试使用data.table包的fread()函数，读文件功能更强： `douban_movie_clean

tensorboard visualization of different loss curves

1

hello, thanks a lot for your efforts on reproducing BeautyGAN. I am also trying to reproduce this paper and can you provide a screenshot of different loss curves during training...

Thanks a lot for your awesome release. May I ask a question about the implementation of SpectralNormalization? The call function of it is as follows, where the the line `self.restore_weights()`...