PyG(pytorch-geometric)安裝

1、新建環境：

conda create -n pytorch(PyG) python=3.6

2、激活環境

activate pytorch(PyG)

3、安裝pytorch（GPU）版本。

4、檢查pytorch版本與CUDA版本：

>>python -c "import torch; print(torch.__version__)"  

>>1.10.1

>>python -c "import torch; print(torch.version.cuda)"

>>10.2

5、安裝pytorch-geometric：

pip install torch-scatter -f https://pytorch-geometric.com/whl/torch-${TORCH}+${CUDA}.html
pip install torch-sparse -f https://pytorch-geometric.com/whl/torch-${TORCH}+${CUDA}.html
pip install torch-cluster -f https://pytorch-geometric.com/whl/torch-${TORCH}+${CUDA}.html
pip install torch-spline-conv -f https://pytorch-geometric.com/whl/torch-${TORCH}+${CUDA}.html
pip install torch-geometric

其中torch換成自己的torch版本，CUDA換成自己的CUDA版本。

例如，我的實例：

pip install torch-scatter -f https://pytorch-geometric.com/whl/torch-1.10.1+cu102.html

pip install torch-sparse -f https://pytorch-geometric.com/whl/torch-1.10.1+cu102.html

pip install torch-cluster -f https://pytorch-geometric.com/whl/torch-1.10.1+cu102.html

pip install torch-spline-conv -f https://pytorch-geometric.com/whl/torch-1.10.1+cu102.html

pip install torch-geometric 

6、測試：

     import torch_geometric.transforms as T

　 from torch_geometric.nn import GCNConv,ChebConv