ZhangTianrong

Same here.

With CUDA 10.0.130, cudnn v7.6, PyTorch 1.1.0, I trined the model with the code you provided in readme on real data. I am getting as low as 0.438 highest val_unseen...

I am facing the same problem here yielding almost same perforance with you. Have you found any clue?

@mforbes Did you find a way to add the option to the demo app without modifying the source and hosting it locally?

@jfthuong I guess so. Is there a solution?

Any update on this. Now I can confirm that this happens whenever I use rdp. Maybe it's because of the virtual monitor in rdp session?

Of course not. I tried the beta version and it seems that the error goes away.

Ah, and here it comes back. The same error. It is interesting as I see the error on both my desktop and laptop. Maybe there are some software which I...

That probably is a problem with compiler version. When I switched from `gcc 9` to `gcc 4`, it got installed in `python 3.8` (but not `3.9`), except that it is...

@mohabouje Any update on this?