赞
踩
先为conda 添加源
编辑home目录下的.condarc文件
如果是windows, 在C:\Users\Administrator
新建一个.conda文件夹,在.conda文件夹新建.condarc文件
或在C:\Users\Administrator这个目录下直接新建.condarc
可以都弄
- channels:
- - https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch/
- - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/
- - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main/
- - https://mirrors.ustc.edu.cn/anaconda/pkgs/main/
- - https://mirrors.ustc.edu.cn/anaconda/pkgs/free/
- - defaults
- show_channel_urls: true
- remote_read_timeout_secs: 600.0
注意要把https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch/
写在第一行,因为像https://mirrors.ustc.edu.cn/anaconda/pkgs/main/这个链接或者其他链接
也有pytorch,但有时候会中断,所以把最稳定最快的链接写前面,因为你不写在前面conda在https://mirrors.ustc.edu.cn/anaconda/pkgs/main/这个链接搜索到了pytorch,就不会去https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch/这个链接搜,导致中断
换好源之后,然后最好先建一个虚拟环境
conda create -n torch1.2 python=3.6
再执行:
conda install pytorch==1.2.0 torchvision==0.4.0 cudatoolkit=10.0
相装1.1的话
conda install pytorch==1.1.0 torchvision==0.3.0 cudatoolkit=9.0
它会自动把其他所有依赖比如cudnn也装,记住后面不要加-c pytorch,要不然就不会走中科大的源
这种用conda装pytorch,就不用手动装cuda,
中科大这两句也要加,因为cudatoolkit我下载的时候都是在中科大这两个源下载的
- https://mirrors.ustc.edu.cn/anaconda/pkgs/main/
- https://mirrors.ustc.edu.cn/anaconda/pkgs/free/
然后我又想改装pytorch 1.3.1
我就直接conda install pytorch==1.3.1
后续测试
- In [1]: import torch
-
- In [2]: torch.__version__
- Out[2]: '1.3.1'
-
- In [3]:
-
- In [3]: torch.cuda.is_available()
- Out[3]: True
下面是装的时候自动装的一些包:
- (pytorch) wosai@wosai:~$ conda list
- # packages in environment at /home/wosai/anaconda3/envs/pytorch:
- #
- # Name Version Build Channel
- _libgcc_mutex 0.1 main https://mirrors.ustc.edu.cn/anaconda/pkgs/main
- backcall 0.1.0 <pip>
- blas 1.0 mkl https://mirrors.ustc.edu.cn/anaconda/pkgs/free
- certifi 2016.2.28 py36_0 https://mirrors.ustc.edu.cn/anaconda/pkgs/free
- cffi 1.10.0 py36_0 https://mirrors.ustc.edu.cn/anaconda/pkgs/free
- cudatoolkit 10.0.130 0 https://mirrors.ustc.edu.cn/anaconda/pkgs/main
- decorator 4.4.1 <pip>
- freetype 2.5.5 2 https://mirrors.ustc.edu.cn/anaconda/pkgs/free
- intel-openmp 2019.4 243 https://mirrors.ustc.edu.cn/anaconda/pkgs/main
- ipython 7.10.1 <pip>
- ipython-genutils 0.2.0 <pip>
- jbig 2.1 0 https://mirrors.ustc.edu.cn/anaconda/pkgs/free
- jedi 0.15.1 <pip>
- jpeg 9b 0 https://mirrors.ustc.edu.cn/anaconda/pkgs/free
- libffi 3.2.1 1 https://mirrors.ustc.edu.cn/anaconda/pkgs/free
- libgcc-ng 9.1.0 hdf63c60_0 https://mirrors.ustc.edu.cn/anaconda/pkgs/main
- libgfortran-ng 7.3.0 hdf63c60_0 https://mirrors.ustc.edu.cn/anaconda/pkgs/main
- libpng 1.6.30 1 https://mirrors.ustc.edu.cn/anaconda/pkgs/free
- libstdcxx-ng 9.1.0 hdf63c60_0 https://mirrors.ustc.edu.cn/anaconda/pkgs/main
- libtiff 4.0.6 3 https://mirrors.ustc.edu.cn/anaconda/pkgs/free
- mkl 2019.4 243 https://mirrors.ustc.edu.cn/anaconda/pkgs/main
- mkl-service 2.3.0 py36he904b0f_0 https://mirrors.ustc.edu.cn/anaconda/pkgs/main
- mkl_fft 1.0.15 py36ha843d7b_0 https://mirrors.ustc.edu.cn/anaconda/pkgs/main
- mkl_random 1.1.0 py36hd6b4f25_0 https://mirrors.ustc.edu.cn/anaconda/pkgs/main
- ninja 1.7.2 0 https://mirrors.ustc.edu.cn/anaconda/pkgs/free
- numpy 1.17.4 py36hc1035e2_0 https://mirrors.ustc.edu.cn/anaconda/pkgs/main
- numpy-base 1.17.4 py36hde5b4d6_0 https://mirrors.ustc.edu.cn/anaconda/pkgs/main
- olefile 0.44 py36_0 https://mirrors.ustc.edu.cn/anaconda/pkgs/free
- openssl 1.0.2l 0 https://mirrors.ustc.edu.cn/anaconda/pkgs/free
- parso 0.5.1 <pip>
- pexpect 4.7.0 <pip>
- pickleshare 0.7.5 <pip>
- pillow 4.2.1 py36_0 https://mirrors.ustc.edu.cn/anaconda/pkgs/free
- pip 9.0.1 py36_1 https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free
- prompt-toolkit 3.0.2 <pip>
- ptyprocess 0.6.0 <pip>
- pycparser 2.18 py36_0 https://mirrors.ustc.edu.cn/anaconda/pkgs/free
- Pygments 2.5.2 <pip>
- python 3.6.2 0 https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free
- pytorch 1.3.1 py3.6_cuda10.0.130_cudnn7.6.3_0 https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch
- readline 6.2 2 https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free
- setuptools 36.4.0 py36_1 https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free
- six 1.10.0 py36_0 https://mirrors.ustc.edu.cn/anaconda/pkgs/free
- sqlite 3.13.0 0 https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free
- tk 8.5.18 0 https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free
- traitlets 4.3.3 <pip>
- wcwidth 0.1.7 <pip>
- wheel 0.29.0 py36_0 https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free
- xz 5.2.3 0 https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free
- zlib 1.2.11 0 https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free

之后运行pytorch版的yolo3之后,报错:
python detect.py
Traceback (most recent call last):
File "detect.py", line 5, in <module>
from utils.datasets import *
File "/home/wosai/2020/PyTorch-YOLOv3/utils/datasets.py", line 12, in <module>
import torchvision.transforms as transforms
File "/home/wosai/anaconda3/envs/pytorch/lib/python3.6/site-packages/torchvision/__init__.py", line 1, in <module>
from torchvision import models
File "/home/wosai/anaconda3/envs/pytorch/lib/python3.6/site-packages/torchvision/models/__init__.py", line 11, in <module>
from . import detection
File "/home/wosai/anaconda3/envs/pytorch/lib/python3.6/site-packages/torchvision/models/detection/__init__.py", line 1, in <module>
from .faster_rcnn import *
File "/home/wosai/anaconda3/envs/pytorch/lib/python3.6/site-packages/torchvision/models/detection/faster_rcnn.py", line 7, in <module>
from torchvision.ops import misc as misc_nn_ops
File "/home/wosai/anaconda3/envs/pytorch/lib/python3.6/site-packages/torchvision/ops/__init__.py", line 1, in <module>
from .boxes import nms, box_iou
File "/home/wosai/anaconda3/envs/pytorch/lib/python3.6/site-packages/torchvision/ops/boxes.py", line 2, in <module>
from torchvision import _C
ImportError: libcudart.so.9.0: cannot open shared object file: No such file or directory
竟然提示要cuda9,我明明我装了cuda10,测试了cuda是可用的,后来想想可能是torchvision版本的原因,因为我装的是torchvision0.3
于是pip install --upgrade torchvision
升级到了最新版本的0.4.2
再测试,成功!所以我在想是不是torch0.3要和cuda9对应,torchvision要和cuda10对应
ubuntu18对cuda10支持的比较好
Copyright © 2003-2013 www.wpsshop.cn 版权所有,并保留所有权利。