Igor Susmelj
Igor Susmelj
**ABC-GAN**: Adaptive Blur and Control for improved training stability of Generative Adversarial Networks - We present two methods only affecting the way you train a GAN to improve it's stability,...
We didn't notice any effect. The test accuracy is on par with other models like SimCLR, MoCo. You can easily introduce split batch norm using [lightly](https://github.com/lightly-ai/lightly). Just change line 86...
We did some quick experiments and it seems that Barlow Twins works well with a similar model configuration as SimSiam. We essentially ended up using the SimSiam model just replacing...
Ok, thanks for the quick reply. That would indeed be a another way to deal with it. However, I would further investigate the distribution of weights and features to figure...
No, it might only be the same for your current code since you resize the input data to 224x224 pixels. The **adaptive version ensures a predefined output shape** whereas the...
Hi, thanks for your interest in contributing. I like your ideas and inputs (Especially the first 3 points). I would like to keep this style guide as simple and lean...
Hi Francesco, you're right. The code there is not nice. It was a leftover to showcase that what you pass to sequential can be broken up into parts (items of...
Thanks, I didn't know that. I added the changes. Do you know whether it only affects the "hooks" or whether it has an impact on the forward/ backward pass?
thanks for the explanation!