Tip: After the article is written, the table of contents can be automatically generated. How to generate it can refer to the help document on the right
Table of contents
1. How to check your own pytorch and cuda versions?
foreword
Because the mmcv-full version does not match the pytorch and cuda versions, the mmcv-full installation fails.
1. How to check your own pytorch and cuda versions?
First open Anaconda3, enter your own python environment, and check the pytorch and cuda versions.
The code is as follows (example):
activate python(自己的python环境名称)
输入python
torch
cuda
2. Install mmcv-full
1. Enter the MMCV official website
URL: mirrors/open-mmlab/mmcv GitCode
找到See the documentation for more features and usage.
Click install mmcv-full
You will find a tool that automatically matches the version, enter your environment version as shown in the figure, copy the installation command of mmcv-full, install it, and you are done!
Summarize
Tip: Before installing mmcv-full, uninstall mmcv first
For example: the above is what I will talk about today. This article briefly introduces the installation of mmcv-full, and I hope it will be helpful to you.