when run conda create ,the yml will be some question when used the cloud system.
1:when used cloud system it is better tu run
conda env create -f multi-qg.yml -n name
name is your name to create ,because when used virtual nvidia docker system ,you can't create directly by multi-qg.yml
2:the torch-scatter have a strictly restriction on cuda version and python version,you should check your cuda version and
find the torch-scatter at :
https://pytorch-geometric.com/whl/torch-1.5.0.html
pip it by download or change the correct version in multi-qg.yml,i suggest download it !! it is important.i waste lot time here
when run conda create ,the yml will be some question when used the cloud system. 1:when used cloud system it is better tu run conda env create -f multi-qg.yml -n name name is your name to create ,because when used virtual nvidia docker system ,you can't create directly by multi-qg.yml 2:the torch-scatter have a strictly restriction on cuda version and python version,you should check your cuda version and find the torch-scatter at : https://pytorch-geometric.com/whl/torch-1.5.0.html pip it by download or change the correct version in multi-qg.yml,i suggest download it !! it is important.i waste lot time here
name: channels: