Akshay Chawla

@ljm355 I solved this problem by using an older version of keras. My setup is as follows: python 3.4.3 tensorflow-cpu (1.4.1) keras - 2.0.6 (this is what made it work)...

@tangsanli5201 Cool! This is a much better solution as it works for all versions of Keras. Could you dump the json for the latest keras (via pip) and raise a...

Hey @savan77 I'm soo sorry for not replying earlier. I hope I can clear some of your doubts. 1. In line 59, I calculate the adversarial version of x, now...

Hey Rajeev, I've got the deconv layer working in a separate implementation and I'll be submitting a PR sometime in the next 3-4 days. I'm just ironing out some issues...

Okay quick update regarding this.. I had been working on a separate heavily modified implantation of Marc's repository so wasn't able to make any commits earlier. Now I've started integrating...

Hi @withbrightmoon , thank you for the interest in our work. I'll try my best to help you out. It definitely looks like the deep feature statistics loss `loss_r_feature` is...

Thanks for running these experiments Xiu. We can atleast see that the images are being optimized w.r.t the losses that are enabled. The dark images in experiment 2 show that...

Hi @withbrightmoon , Thank you for running these experiments, the results look objectively better than before. I apologize for not responding earlier. Here are a few more things that you...

Thank you for the interest in our work @doublelei ! I have enabled the download permissions for the GCS bucket. Please confirm that you can view and download the data....

Hi Rahul! Sorry for the late response. I think it is because I skip certain filetypes and forcefully add a .jpg at the end if no extension is available. I'm...