pytorch安装cpu linux,Ubuntu16.04 系统下的pytorch库（cpu版）安装与demo测试

PyTorch是一个快速，灵活的实验深度学习框架[1]。下面介绍Ubuntu16.04 系统及Anaconda环境下，对于pytorch库(cpu版)的安装与demo测试。

系统环境：

Ubuntu16.04

Anaconda2-5.2.0-Linux-x86_64 [2]

pytorch 安装及环境配置

1、源码下载

git clone --recursive https://github.com/pytorch/pytorch

2、安装

打开，并进入刚下载的源码文件夹

cd ./pytorch

执行安装文件 setup.py

python setup.py install

注意：执行该命令时，可能会提示权限问题，你可使用chmod命令[3]将该问题解决。例如我的pytorch源码文件夹路径是：~/Downloads/pytorch。通过执行以下命令即可解决权限问题：

chmod 777 ~/Downloads/pytorch

如需了解更多有关chmod命令，可以参考文档[1]。

pytorch包导入测试

在Python环境下输入命令：

import torch

如果在导入pytorch包没有提示错误，就说明安装成功了。

注意：本人使用该导入命令时，提示错误：“ImportError: No module named _C”。

Traceback (most recent call last):

File "", line 1, in

File "torch/__init__.py", line 78, in

from torch._C import *

ImportError: No module named _C

解决办法：关闭该终端窗口，打开新的终端窗口输入导入pytorch包测试即可通过。

demo测试

# -*- coding: utf-8 -*-

import torch

dtype = torch.float

device = torch.device("cpu")

# dtype = torch.device("cuda:0") # Uncomment this to run on GPU

# N is batch size; D_in is input dimension;

# H is hidden dimension; D_out is output dimension.

N, D_in, H, D_out = 64, 1000, 100, 10

# Create random input and output data

x = torch.randn(N, D_in, device=device, dtype=dtype)

y = torch.randn(N, D_out, device=device, dtype=dtype)

# Randomly initialize weights

w1 = torch.randn(D_in, H, device=device, dtype=dtype)

w2 = torch.randn(H, D_out, device=device, dtype=dtype)

learning_rate = 1e-6

for t in range(500):

# Forward pass: compute predicted y

h = x.mm(w1)

h_relu = h.clamp(min=0)

y_pred = h_relu.mm(w2)

# Compute and print loss

loss = (y_pred - y).pow(2).sum().item()

print(t, loss)

# Backprop to compute gradients of w1 and w2 with respect to loss

grad_y_pred = 2.0 * (y_pred - y)

grad_w2 = h_relu.t().mm(grad_y_pred)

grad_h_relu = grad_y_pred.mm(w2.t())

grad_h = grad_h_relu.clone()

grad_h[h < 0] = 0

grad_w1 = x.t().mm(grad_h)

# Update weights using gradient descent

w1 -= learning_rate * grad_w1

w2 -= learning_rate * grad_w2

输出结果如下：

(0, 33276068.0)

(1, 33026900.0)

(2, 35287920.0)

(3, 33936460.0)

(4, 26412418.0)

(5, 16149083.0)

......

......

......

(489, 4.269163400749676e-05)

(490, 4.2141629819525406e-05)

(491, 4.168449595454149e-05)

(492, 4.131443711230531e-05)

(493, 4.065494067617692e-05)

(494, 3.98996562580578e-05)

(495, 3.9512655348517e-05)

(496, 3.895666668540798e-05)

(497, 3.858671334455721e-05)

(498, 3.7854093534406275e-05)

(499, 3.753306737053208e-05)

参考链接

[1] PyTorch官网教程

https://pytorch.org/tutorials/index.html

[2]Anaconda镜像下载：

https://mirrors.tuna.tsinghua.edu.cn/anaconda/archive/Anaconda2-5.2.0-Linux-x86_64.sh

[3]chmod命令

https://blog.csdn.net/u010636181/article/details/80678758

更多精彩项目及分享，请关注我的个人公众号：gbxiao992