I installed pytorch using the command
conda install pytorch torchvision torchaudio cudatoolkit=11.1 -c pytorch -c conda-forge. This is the second installation of pytorch. I think the first one was the default installed by
pip install -Uqq fastbook. However, after the second installation of pytorch because I wanted my GPU to be used during training, my jupyter kernel keeps restarting everytime I run a code block in Juypter Notebook. I get the error
Kernel started: efe53fe5-e86c-4674-9257-b8aec10b10f6, name: python3 Key already registered with the same priority: GroupSpatialSoftmax
Does anyone know what’s up with this and how to fix this? Thanks!