[Pytorch] MNIST

import torch
from torchvision import transforms
from torchvision import datasets
from torch.utils.data import DataLoader
import torch.nn.functional as F #选择激活函数
import torch.optim as optim #选择优化器

在输出的时候，%d表示输出整型数值，
%5d输出的整型值占5位宽度，不足5位的话，在左边用空格补充
%.3f 中， % 引导后面的符号，表示用来格式化字符串后面 % 之后的变量变成字符串。 .3 表示小数点后面保留三位。

图像有通道，长，宽三个特性 通道为 channel WHC pytorch为CWH
Z 2828 pixel 为0-255
R 128*28 pixel为0-1

import torch
from torchvision import transforms
from torchvision import datasets
from torch.utils.data import DataLoader
import torch.nn.functional as F #选择激活函数
import torch.optim as optim #选择优化器

#准备数据
batch_size=64
trans=transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.1307,),(0.3081,))])

#数据载入

train_dataset=datasets.MNIST(root=’…/MNNIST/mnist/’,train=True,download=True,transform=trans)

train_loader=DataLoader(train_dataset,shuffle=True,batch_size=batch_size)

test_dataset=datasets.MNIST(root=’…/MNNIST/mnist/’,train=False,download=True,transform=trans)

test_loader=DataLoader(test_dataset,shuffle=True,batch_size=batch_size)

class Net(torch.nn.Module):
def init(self):
super(Net, self).init()
self.linear1 = torch.nn.Linear(784, 512)
self.linear2 = torch.nn.Linear(512, 64)
self.linear3 = torch.nn.Linear(64, 10)

def forward(self, x):
    x = x.view(-1, 784)
    x = F.relu(self.linear1(x))
    x = F.relu(self.linear2(x))
    x = self.linear3(x)
    return x

model=Net()

criterion=torch.nn.CrossEntropyLoss() #计算损失
optimizer=optim.SGD(model.parameters(),lr=0.01,momentum=0.5) #momentum带冲量的？？？

def train(epoch):
running_loss=0.0

for batch_idx,data in enumerate(train_loader,0):
    input,targe=data

    optimizer.zero_grad()

    output=model(input)
    loss=criterion(output,targe)
    loss.backward()

    optimizer.step()


    running_loss+=loss.item()  #计算输出

    if batch_idx%300==299:
        print("[%d,%d]  loss: %.4f" % (epoch+1,batch_idx+1,running_loss/300))
        running_loss=0

def test():
correct=0
total=0
with torch.no_grad():
for data in test_loader:
images,labels=data
outputs=model(images)

        _,predicted=torch.max(outputs.data,dim=1)  #数据和数据下标

        total+=labels.size(0)

        correct+=(predicted==labels).sum().item()

print("Accuracy on test set:  %d %%" % (100*correct/total))

if name==‘main’:
for epoch in range(1):
train(epoch)
test()

Result:
[1,300] loss: 0.9973
[1,600] loss: 0.3751
[1,900] loss: 0.2958
Accuracy on test set: 92 %