来自这个论文:
来自这个仓库:
https://github.com/chao1224/Geom3D
之前按照README.md中的指导来安装我会出现这个问题:
AttributeError: 'NoneType' object has no attribute 'origin'
而且不知道为什么使用conda安装环境很慢……
所以尝试全部使用pip进行安装:
创建并激活环境:
conda create -n Geom3D python=3.7
conda activate Geom3D
安装rdkit、numpy、networkx、sklearn:
pip install rdkit
pip numpy networkx scikit-learn
安装pytorch(我选择1.12.1版本,可以根据自己的系统调整):
# CUDA 11.6
pip install torch==1.12.1+cu116 torchvision==0.13.1+cu116 torchaudio==0.12.1 --extra-index-url https://download.pytorch.org/whl/cu116
安装pyg:
pip install torch_geometric
根据pytorch版本安装pyg对应的相关库:
https://data.pyg.org/whl/这个网址下找到torch对应的版本
https://data.pyg.org/whl/torch-1.12.1%2Bcu116.html这是torch1.12.1 cuda为11.6的对应whl安装网址
pip install pyg_lib torch_scatter torch_sparse torch_cluster torch_spline_conv -f https://data.pyg.org/whl/torch-1.12.1%2Bcu116.html
下载其余的:
pip install ogb==1.2.1
pip install sympy
pip install ase
pip install lie_learn # for TFN and SE3-Trans
pip install packaging # for SEGNN
pip3 install e3nn # for SEGNNpip install transformers # for smiles
pip install selfies # for selfiespip install atom3d # for Atom3D
pip install cffi # for Atom3D
pip install biopython # for Atom3Dpip install cython # for pyximport
pip install ogb sympy ase lie_learn packaging e3nn transformers selfies atom3d cffi biopython cython
最后那个py-xgboost-cpu我是没安装,先看看效果吧。