softmax回归

• Softmax 回归是一个多类分类模型
• 使用softmax操作得到每个类的预测置信度
• 使用交叉熵来衡量预测和标号的区别

import torch
import random
import torchvision
from torch import nn
from torch.utils.data import Dataset,DataLoader
from torch.utils.tensorboard import SummaryWriter
from torchvision import transforms
from d2l import torch as d2l
import matplotlib.pyplot as plt

#读取数据集
minst_train = torchvision.datasets.FashionMNIST(root='./data',train=True,transform=transforms.ToTensor(),download=True)
minst_test = torchvision.datasets.FashionMNIST(root='./data',train=False,transform=transforms.ToTensor(),download=True)

#数据集的长度
minst_train_len = len(minst_train)
minst_test_len = len(minst_test)
print('训练集的长度：{} ,测试集的长度：{}'.format(minst_train_len,minst_test_len))

#加载数据集
dataloader = DataLoader(dataset=minst_train,batch_size=256,shuffle=True)

for X,y in dataloader:
    print(X.shape,X.dtype,y.shape,y.dtype)
    break

#定义模型
net = nn.Sequential(nn.Flatten(),nn.Linear(784,10))

#初始化参数
def init_weights(m):
    if type(m) == nn.Linear:
        nn.init.normal_(m.weight,std=0.01)

net.apply(init_weights)

#定义损失函数
loss = nn.CrossEntropyLoss(reduction='none')

#优化算法
optimer = torch.optim.SGD(net.parameters(),lr=0.1)

#训练
epoches = 10

for epoch in range(epoches):
    for X,y in dataloader:
        y1 = net(X)
        l = loss(y1,y)
        optimer.zero_grad()
        l.mean().backward()
        optimer.step()
    print("epoch:{}  loss:{}".format(epoch+1,l))