Alexandra

Hello @JaydenKing32 , did you find a cleaner solution for `h.num_classes = 90`? We have the same issue. Thanks

Hello, To avoid this problem I downloaded cuda9.1 using this link [Cuda 9.1 on Ubuntu 18.04](https://gist.github.com/DaneGardner/accd6fd330348543167719002a661bd5). Then don't forget to install all recommended packages + `conda install pytorch=0.4.1 cuda91 -c...

Hello did you finally find it the "trained DPATCH" ? @linsencc

Yes, since I didn't find the code neither the `untargeted_DPATCH.npy`. I train the model again, `python train.py`

Same question here @QUJIJI ? My email [email protected]

1. Force the device = torch.device('cpu') 2. Remove all .to(0) that is used for GPU only It should work (at least it work on my Mac)

I have a macbook with M1 chip, I installed every packages/versions with `poetry install`