Drew Arad Hudson

Hi, the public GANformer model supports unconditional generation but indeed I do plan to release the extended model, GANformer2, https://arxiv.org/abs/2111.08960, that can perform conditional generation from a layout to an...

Thank you so much for the kind words! :-)

It may be a good idea to change the default in adam from 0.001 to 0.01. In experiments I've done so far 0.01 was consistently better..@null-a you changed the default...

@null-a right. moved my comment to there.

Regarding the stepSize - I also got significant improvement with 0.01, especially for dream..

What about the default number of forward samples? is 1 good?

@ngoodman is there a reason not to change the default from 0.001 to 0.01? In all cases I checked 0.01 was better. @null-a changed the default some time ago from...

Hello everybody! Thank you very much for the interest in the dataset and the challenge, and I would like to begin by the most honest and sincere apology for replying...