pytorch笔记1

pytorch基础知识

import torch
x  = torch.tensor([2,3,4], dtype=torch.float) # 创建一个Tensor，值为[2.,3.,4.]，类型为 float

# 创建一个需要求 梯度的 tensor。
x2 = torch.tensor([2,3,4], dtype=torch.float, requires_grad=True)
x.size()
torch.Size([3])
a=1;b=2;
a.add_(b) # 所有带 _ 的operation，都会更改调用对象的值，
#例如 a=1;b=2; a.add_(b); a就是3了，没有 _ 的operation就没有这种效果，只会返回运算结果
torch.cuda.is_available()

import torch
x = torch.tensor([1,1,1,1,1], dtype=torch.float, requires_grad=True)
y = x * 2
grads = torch.FloatTensor([1,2,3,4,5])
y.backward(grads)#如果y是scalar的话，那么直接y.backward()，然后通过x.grad方式，就可以得到var的梯度
x.grad           #如果y不是scalar，那么只能通过传参的方式给x指定梯度

神经网络

import torch
import torch.nn as nn
import torch.nn.functional as F

class Net(nn.Module):

    def __init__(self):
        super(Net, self).__init__()
        # 1 input image channel, 6 output channels, 3x3 square convolution
        # kernel
        self.conv1 = nn.Conv2d(1, 6, 3)
        self.conv2 = nn.Conv2d(6, 16, 3)
        # an affine operation: y = Wx + b
        self.fc1 = nn.Linear(16 * 6 * 6, 120)  # 6*6 from image dimension
        self.fc2 = nn.Linear(120, 84)
        self.fc3 = nn.Linear(84, 10)

    def forward(self, x):
        # Max pooling over a (2, 2) window
        x = F.max_pool2d(F.relu(self.conv1(x)), (2, 2))
        # If the size is a square you can only specify a single number
        x = F.max_pool2d(F.relu(self.conv2(x)), 2)
        x = x.view(-1, self.num_flat_features(x))
        x = F.relu(self.fc1(x))
        x = F.relu(self.fc2(x))
        x = self.fc3(x)
        return x

    def num_flat_features(self, x):
        size = x.size()[1:]  # all dimensions except the batch dimension
        num_features = 1
        for s in size:
            num_features *= s
        return num_features

net = Net()
print(net)#打印网络结构

params = list(net.parameters())
print(len(params))
print(params[0].size())  # conv1's .weight
#让我们尝试一个32x32随机输入
input = torch.randn(1, 1, 32, 32)#（1,1,32,32）大小的正态分布
out = net(input)
print(out)
#使用随机梯度将所有参数和反向传播的梯度缓冲区归零
net.zero_grad()
out.backward(torch.randn(1, 10))
#损失函数
output = net(input)
target = torch.randn(10)  # a dummy target, for example
target = target.view(1, -1)  # make it the same shape as output
criterion = nn.MSELoss()

loss = criterion(output, target)
print(loss)
print(loss.grad_fn)  # MSELoss
print(loss.grad_fn.next_functions[0][0])  # Linear
print(loss.grad_fn.next_functions[0][0].next_functions[0][0])  # ReLU
#反向传播
net.zero_grad()     # zeroes the gradient buffers of all parameters

print('conv1.bias.grad before backward')
print(net.conv1.bias.grad)

loss.backward()

print('conv1.bias.grad after backward')
print(net.conv1.bias.grad)
#更新权重
import torch.optim as optim

# create your optimizer
optimizer = optim.SGD(net.parameters(), lr=0.01)

# in your training loop:
optimizer.zero_grad()   # zero the gradient buffers
output = net(input)
loss = criterion(output, target)
loss.backward()
optimizer.step()    # Does the update

一般结构

import torch.nn as nn
import torch.nn.functional as F

class Net(nn.Module):#需要继承这个类
    def __init__(self):
        super(Net, self).__init__()
        #建立了两个卷积层，self.conv1, self.conv2，注意，这些层都是不包含激活函数的
        self.conv1 = nn.Conv2d(1, 6, 5) # 1 input image channel, 6 output channels, 5x5 square convolution kernel
        self.conv2 = nn.Conv2d(6, 16, 5)
        #三个全连接层
        self.fc1   = nn.Linear(16*5*5, 120) # an affine operation: y = Wx + b
        self.fc2   = nn.Linear(120, 84)
        self.fc3   = nn.Linear(84, 10)

    def forward(self, x): #注意，2D卷积层的输入data维数是 batchsize*channel*height*width
        x = F.max_pool2d(F.relu(self.conv1(x)), (2, 2)) # Max pooling over a (2, 2) window
        x = F.max_pool2d(F.relu(self.conv2(x)), 2) # If the size is a square you can only specify a single number
        x = x.view(-1, self.num_flat_features(x))
        x = F.relu(self.fc1(x))
        x = F.relu(self.fc2(x))
        x = self.fc3(x)
        return x
    
    def num_flat_features(self, x):
        size = x.size()[1:] # all dimensions except the batch dimension
        num_features = 1
        for s in size:
            num_features *= s
        return num_features

net = Net()

# create your optimizer
optimizer = optim.SGD(net.parameters(), lr = 0.01)

# in your training loop:
for i in range(num_iteations):
    optimizer.zero_grad() # zero the gradient buffers，如果不归0的话，gradients会累加

    output = net(input) # 这里就体现出来动态建图了，你还可以传入其他的参数来改变网络的结构

    loss = criterion(output, target)
    loss.backward() # 得到grad，i.e.给Variable.grad赋值
    optimizer.step() # Does the update，i.e. Variable.data -= learning_rate*Variable.grad