akaniklaus

I am getting the error with Python3, PyTorch 1.0 on Ubuntu 18.04 (where I installed Anaconda3) Yes I am not having any error with PyTorch 0.4 so it should be...

By the way, would you like me to integrate decoupled weight decay into your work? Also, this paper is very relevant to yours: https://openreview.net/forum?id=BJll6o09tm I was actually planning to decay...

It would be compatible but just let me know if I should directly switch it to 'decoupled weight' decay, which is proven better especially for Adam; or enable the control...

Ok will do that. On Wed 6. Mar 2019 at 12:33, LoLo wrote: > It seems that the decoupled weight decay version for Adam is not > integrated into Adam...

Hello @Luolc, FYI, they want to add this to pytorch/contrib if you can make a pull request: https://github.com/pytorch/pytorch/issues/17560

Hello, I also didn't experience the speed improvement. There is authors' implementation in the below link but it seems much more complex to plug & play'; and they only provided...

Anybody managed to reproduce GE-theta plus? I am unable to since that models provided here quite cryptic to me, whilst all we require to reproduce the core block of the...

I have tried just changing convolutions, pooling, etc to 1d but I keep getting the following error: ``` File "pd_archs.py", line 95, in __init__ self.conv = OctConv(in_channels, out_channels, kernel_size =...

Do you know what happens when you don't use enumerate but get batches via next(iter(data_loader))?

@WojciechMigda You wouldn't like to buy / sell while it is still on a trend. You would rather buy / sell on a trend reversal when the trend (e.g. ADX)...