pytorch基本概念
PyTorch 是由 Torch7 团队开源的，这也是Facebook 的 AI 研究团队发布了一个 Python 工具包，据该项目官网介绍，是一个 Python 优先的深度学习框架，能够在强大的 GPU 加速基础上实现张量和动态神经网络。
github:https://github.com/pytorch/pytorch
pytorch的安装
官网 推荐基于anaconda的pytorch安装.
conda install pytorch torchvision cudatoolkit=9.0 -c pytorch配置python环境
python windows环境下载地址: https://www.python.org/downloads/windows/
基于anaconda管理python环境(推荐)
国外的anaconda源下载速度比较满，windows环境的exe安装文件我分享出来了
链接：https://pan.baidu.com/s/1AN2fKb1LpuQzvv2Us5ctvw
提取码：0wlu
或者使用中科大的镜像地址: https://mirrors.ustc.edu.cn/anaconda/archive/
常用命令
# 查看版本号
conda --version
# 创建环境
conda create --name your_env_name python=version
# 查看环境
conda info -e
# 切换环境
source activate your_env_name通过命令行安装pytorch
conda install pytorch torchvision cudatoolkit=9.0 -c pytorchpytorch的基本操作
import torch
# 构造一个未初始化的5x3矩阵
x = torch.empty(5, 3)
print(x)
# 构造一个随机初始化的矩阵
x = torch.rand(5, 3)
print(x)
# 构造一个矩阵填充的零和dtype long
x = torch.zeros(5, 3, dtype=torch.long)
print(x)
# 直接构造张量
x = torch.tensor([5, 3], dtype=torch.float)
print(x)
# 根据现有的张量创建张量。除非用户提供新值，否则这些方法将重用输入张量的属性，例如dtype
x = x.new_ones(5, 3, dtype=torch.double) # new_* methods take in sizes
print(x)
# 操作符
x = torch.tensor([[1, 2], [3,4]])
y = torch.tensor([[5, 6], [7,8]])
torch.add(x, y)
print(x + y)
print(x * y)
# 转置矩阵
print(torch.t(x))
# 将NumPy数组转换为Torch Tensor
import numpy as np
a = np.ones(5)
b = torch.from_numpy(a)
print(b)
代码实现一个简单的神经网络(GPU)-后再优化
import torch.nn as nn
import torch.nn.functional as F
import torch
import torchvision
import torchvision.transforms as transforms
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
print(device)
# torchvision 是一个和 torch配合使用的Python包, 提供了几个常用的数据集,
# 几种已经搭建好的经典网络模型, 以及一些图像数据处理方面的工具
# 训练一个图像的分类器
transform = transforms.Compose(
[transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])
trainset = torchvision.datasets.CIFAR10(root='./data', train=True,
download=True, transform=transform)
trainloader = torch.utils.data.DataLoader(trainset, batch_size=4,
shuffle=True, num_workers=0, timeout=30000)
testset = torchvision.datasets.CIFAR10(root='./data', train=False,
download=True, transform=transform)
testloader = torch.utils.data.DataLoader(testset, batch_size=4,
shuffle=False, num_workers=0, timeout=30000)
classes = ('plane', 'car', 'bird', 'cat',
'deer', 'dog', 'frog', 'horse', 'ship', 'truck')
# 模型
class Net(nn.Module):
def __init__(self):
super(Net, self).__init__()
self.conv1 = nn.Conv2d(3, 6, 5)
self.pool = nn.MaxPool2d(2, 2)
self.conv2 = nn.Conv2d(6, 16, 5)
self.fc1 = nn.Linear(16 * 5 * 5, 120)
self.fc2 = nn.Linear(120, 84)
self.fc3 = nn.Linear(84, 10)
def forward(self, x):
x = self.pool(F.relu(self.conv1(x)))
x = self.pool(F.relu(self.conv2(x)))
x = x.view(-1, 16 * 5 * 5)
x = F.relu(self.fc1(x))
x = F.relu(self.fc2(x))
x = self.fc3(x)
return x
net = Net()
# 在GPU上训练
net.to(device)
# 定义损失函数和优化器
import torch.optim as optim
# 交叉熵损失函数
criterion = nn.CrossEntropyLoss()
optimizer = optim.SGD(net.parameters(), lr=0.001, momentum=0.9)
# 训练网络
for epoch in range(2):
running_loss = 0.0
for i, data in enumerate(trainloader, 0):
# 获取输入
inputs, labels = data
inputs, labels = inputs.to(device), labels.to(device)
# 梯度置 0
optimizer.zero_grad()
# 正向传播 反向传播 优化
outputs = net(inputs)
loss = criterion(outputs, labels)
loss.backward()
optimizer.step()
# 打印状态信息
running_loss += loss.item()
if i % 2000 == 1999: # 每2000批次打印一次
print('[%d, %5d] loss: %.3f' %
(epoch + 1, i + 1, running_loss / 2000))
running_loss = 0.0
print('Finished Training')
correct = 0
total = 0
with torch.no_grad():
for data in testloader:
images, labels = data
images, labels = inputs.to(device), labels.to(device)
outputs = net(images)
_, predict = torch.max(outputs.data, 1)
total += labels.size(0)
correct += (predict == labels).sum().item()
print('Accuracy of the network on the 10000 test images: %d %%' % (
100 * correct / total))