pytorch安装与入门（一）

 好久没有更新pytorch，这次更新做个笔记

 

 

1.安装CUDA

https://developer.nvidia.com/cuda-toolkit-archive

 

设置环境变量后输入nvcc -V

C:\Users\WQBin>nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2019 NVIDIA Corporation
Built on Fri_Feb__8_19:08:26_Pacific_Standard_Time_2019
Cuda compilation tools, release 10.1, V10.1.105

 

 2.安装pytorch

• PyTorch on Windows only supports Python 3.x

 

2.1解决慢到无法装的问题

最有效方法是添加镜像源，常见的清华或中科大。先查看是否已经安装相关镜像源，windows系统在cmd窗口中执行命令：

conda config --show

使用【conda config --add channels 】添加清华数据源

channels:
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch/
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/Paddle/
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge/

安装：

conda install pytorch torchvision cudatoolkit=10.1

 

测试pytorch

>>> import torch
>>> torch.cuda.is_available()
True

如果是false原因可能是：

• CUDA的版本与cudatoolkit=10.1不匹配
• 需要新版本的驱动安装--https://www.nvidia.cn/Download/index.aspx?lang=cn

 

>>> from __future__ import print_function
>>> import torch
>>> x = torch.rand(5, 3)
>>> print(x)
tensor([[0.1863, 0.4498, 0.0765],
        [0.6291, 0.2132, 0.5534],
        [0.8320, 0.9772, 0.7931],
        [0.8859, 0.9128, 0.0536],
        [0.5445, 0.5895, 0.1608]])
>>>  

其他注意事项：

1.要求电脑的显卡驱动大于396.26。输入 nvidia-smi，查看自己的 Driver Version。

(deeplearning) C:\Users\WQBin>nvidia-smi
Sat Apr 11 11:45:39 2020
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 445.75       Driver Version: 445.75       CUDA Version: 11.0     |
|-------------------------------+----------------------+----------------------+
| GPU  Name            TCC/WDDM | Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  GeForce GTX 1060   WDDM  | 00000000:01:00.0 Off |                  N/A |
| N/A   45C    P8     6W /  N/A |     92MiB /  6144MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+

完美。。开始搞事情