Automatically configure cudatoolkit=11.1 and torch=1.8.2 scripts in the new anaconda environment. The commented out part below is the mirror setting. If it has not been configured before, the comment should be uncommented.
# echo "Setting tsinghua and froge mirrors to conda..."
# conda config --add channels 'https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/'
# conda config --add channels 'https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main/'
# conda config --add channels https://mirrors.ustc.edu.cn/anaconda/cloud/conda-forge/
# conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch/
# conda config --set show_channel_urls yes
echo "Installing cudatoolkit=11.1..."
conda install cudatoolkit=11.1
echo "updating pip and setting mirror..."
pip install pip -U
pip config set global.index-url https://pypi.tuna.tsinghua.edu.cn/simple
echo "installing torch 1.8.2_cu111 to your env..."
pip3 install torch==1.8.2+cu111 torchvision==0.9.2+cu111 torchaudio==0.8.2 -f https://download.pytorch.org/whl/lts/1.8/torch_lts.html
How to use: Create a new .sh file, copy the code into it, and run it directly in the conda environment. (Of course, you can also enter it manually line by line)