查看某个安装包XXX的依赖项的conda指令为:
conda search XXX -info
比如XXX为pytorch0.3.1,就会有如下输出:
pytorch 0.3.1 py36hfbe7015_1
----------------------------
file name : pytorch-0.3.1-py36hfbe7015_1.tar.bz2
name : pytorch
version : 0.3.1
build string: py36hfbe7015_1
build number: 1
size : 216.1 MB
constrains : ()
license : BSD 3-Clause
subdir : linux-64
url : https://mirrors.ustc.edu.cn/anaconda/pkgs/main/linux-64/pytorch-0.3.1-py36hfbe7015_1.tar.bz2
md5 : 5889bb2d4b5b2f14c2fb510857746361
dependencies:
- cffi
- cudatoolkit 8.0.*
- cudnn >=7.0.5,<=8.0a0
- libgcc-ng >=5.4.0
- libstdcxx-ng >=5.4.0
- nccl <2
- numpy >=1.11.3,<2.0a0
- python >=3.6,<3.7.0a0
- mkl >=2018.0.2,<2019.0a0
手动安装dependencies中的项能够解决conflict错误。