I started with pytorch version 1.9 but I had to update it for my project so I ran
conda install pytorch torchvision torchaudio cudatoolkit=10.2 -c pytorch
to get it to version 1.11. But then I found out my gpu is too new for this version of pytorch
with CUDA capability sm_86 is not compatible with the current PyTorch installation. The current PyTorch install supports CUDA capabilities sm_37 sm_50 sm_60 sm_70.
so I try
conda install pytorch torchvision torchaudio cudatoolkit=11.3 -c pytorch
to use cuda 11.3
and now it complains about conflicts and how it has to use flexible solver and its taking forever examining files one by one
Collecting package metadata (current_repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Solving environment: failed with repodata from current_repodata.json, will retry with next repodata source.
Collecting package metadata (repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Solving environment: \
Found conflicts! Looking for incompatible packages.
This can take several minutes. Press CTRL-C to abort.
I tried uninstalling pytorch
`conda uninstall pytorch`
but its still tripping up and going ot the hours long flexible solver installation.