anaconda+pytorch+导入

1.需要卸载原本的python 或者 看看教程
2.下载anaconda并安装
3.配置镜像 配置环境
4.在prompt里，命令如下
新建虚拟环境 conda create --prefix=X:\X\X\envs<name> python=3.X
激活 conda activate
官网下载pytorch 输入命令conda install pytorch torchvision torchaudio pytorch-cuda=12.1 -c pytorch -c nvidia
done之后检查
检查有无pytorch conda list
看看python python
检查是否有GPU import torch
torch.cuda.is_available() True

5.导出到PyCharm 见4；主要是改设置里面的解释器

3。https://blog.csdn.net/liuhyusb/article/details/135753864?ops_request_misc=%257B%2522request%255Fid%2522%253A%2522171504889516800222852542%2522%252C%2522scm%2522%253A%252220140713.130102334…%2522%257D&request_id=171504889516800222852542&biz_id=0&utm_medium=distribute.pc_search_result.none-task-blog-2_alltop_positive~default-1-135753864-null-null.142^v100pc_search_result_base8&utm_term=anaconda%E5%AE%89%E8%A3%852024&spm=1018.2226.3001.4187

4。https://blog.csdn.net/weixin_56462041/article/details/130096491?ops_request_misc=%257B%2522request%255Fid%2522%253A%2522171507005616800182794098%2522%252C%2522scm%2522%253A%252220140713.130102334…%2522%257D&request_id=171507005616800182794098&biz_id=0&utm_medium=distribute.pc_search_result.none-task-blog-2_allsobaiduend~default-2-130096491-null-null.142^v100pc_search_result_base6&utm_term=EnvironmentLocationNotFound%3A%20Not%20a%20conda%20environment&spm=1018.2226.3001.4187