出现 “module ‘pip‘ has no attribute ‘pep425tags‘
出错原因:
网上说查询pip支持,但出现 “module ‘pip‘ has no attribute ‘pep425tags‘ “的错误解决办法,但出现 “module ‘pip‘ has no attribute ‘pep425tags‘ “的错误解决办法
有用:(240条消息) 想要查询pip支持,但出现 “module ‘pip‘ has no attribute ‘pep425tags‘ “的错误解决办法_天天想着怎么转行的吱的博客-CSDN博客
autodl上下载dgl-安装pytorch1.12 +cuda11.3+dgl
1、安装pytorch1.12+cuda11.3(网址:Previous PyTorch Versions | PyTorch)
# CUDA 11.3
pip install torch==1.12.0+cu113 torchvision==0.13.0+cu113 torchaudio==0.12.0 --extra-index-url https://download.pytorch.org/whl/cu113
2、安装dgl:(网址:[写文章-CSDN创作中心](https://mp.csdn.net/mp_blog/creation/editor/130822955?not_checkout=1))
conda install -c dglteam/label/cu113 dgl
pip install --upgrade charset_normalizer
就成功啦!!
autodl上下载dgl总是失败,应该把保存的文件放在哪里?
1、首先从这个网站上下载你想要的dgl版本:
Index of /anaconda/cloud/dglteam/win-64/ | 清华大学开源软件镜像站 | Tsinghua Open Source Mirror
2、接着解压缩后放进下方红字的文件夹中就可以啦!
最新(23.6.19):在这个里dgl才安装成功:
/miniconda3/envs/pya100/lib/python3.8/site-packages#
————————————————————————————————————
miniconda3/lib/python3.8/site-packages/
上面的文件夹里面放:dgl-0.9.0-py3.8.egg-info和dgl (这都是解压缩后一直打开就能找到的!info可以放在miniconda3的info文件夹里)
如果本地下载:在学校服务器,放的位置如下:
如果在线下载:服务器上下载DGL:
# If you have installed dgl-cudaXX.X package, please uninstall it first.
conda install -c dglteam/label/cu113 dglconda install cudatoolkit
#pip install dgl-cu113 dglgo -f https://data.dgl.ai/wheels/repo.html
除了下载这个之外,还要下载cudatoolkit,不然就会报错OSError: libcublas.so.11: cannot open shared object file: No such file or directory
RuntimeError: Failed to import transformers.models.auto because of the following error (look up to see its traceback):
Failed to import transformers.configuration_utils because of the following error (look up to see its traceback):
No module named 'T5Config':
查了一下,大概是版本不兼容:
解决:下载对应的版本 torch1.11.0+cuda11.3+ transformers4.13.0
接着又出新的错误了:
ModuleNotFoundError: No module named 'torch_scatter'
解决:
https://pytorch-geometric.com/whl/torch-1.11.0%2Bcu113.html
文章评论