Inar TImiryasov
Results
2
issues of
Inar TImiryasov
The `DeepIce` model contains a method called `no_weight_decay()` which is intended to specify that the `cls_token` parameter should not be subject to weight decay during training: ``` @torch.jit.ignore def no_weight_decay(self)...
bug
Update PyTorch requirement to use CUDA 12.1 The previous requirements file specified PyTorch 2.2.0 with CUDA 11.8 (cu118). However, based on the available packages from the PyTorch stable and PyTorch...