Nik Li

Hi, when will the image-to-image retrieval be released, os is there any alternative way (e.g. change some configurations) to train my own image-to-image dataset using the current version? Thanks.

I have noticed that there is a requirement of z-dimension >=36. I wonder where is the limitation implemented in the code? Is it possible to modify the code to make...

generate white images during training

1

Hi, I found the all of the saved reconstructed images are white images under the directory ./logs/ during training(shown below). Is there anyone can tell me the potential reason? 

非常感谢你的分享

2

这是我今年star的最好的一个project 非常感谢你的分享，非常详细 希望以后有机会可以多交流合作