学习笔记：动手学深度学习 13 softmax回归的简洁实现

from torch import nn
from d2l import torch as d2l
batch_size = 256
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size)
Backend Qt5Agg is interactive backend. Turning interactive mode on.
""""拿进数据"""
Out[3]: '"拿进数据'
"""初始化模型参数""" 
# PyTorch不会隐式地调整输入的形状。因此，
# 我们在线性层前定义了展平层（flatten），来调整网络输入的形状
net = nn.Sequential(nn.Flatten(), nn.Linear(784, 10))#28*28
def init_weights(m):
    if type(m) == nn.Linear:
        nn.init.normal_(m.weight, std=0.01)
"""如果是线性网络，以均值0和标准差0.01随机初始化权重"""
net.apply(init_weights);
"""在交叉熵损失函数中传递未归一化的预测，并同时计算softmax及其对数"""
Out[5]: '在交叉熵损失函数中传递未归一化的预测，并同时计算softmax及其对数'
loss = nn.CrossEntropyLoss()
"""优化算法--学习率为0.1的小批量随机梯度下降作为优化算法"""
Out[7]: '优化算法--学习率为0.1的小批量随机梯度下降作为优化算法'
trainer = torch.optim.SGD(net.parameters(), lr=0.1)
"""训练"""
Out[9]: '训练'
num_epochs = 10
d2l.train_ch3(net, train_iter, test_iter, loss, num_epochs, trainer)
<Figure size 350x250 with 1 Axes>
<Figure size 350x250 with 1 Axes>
<Figure size 350x250 with 1 Axes>
<Figure size 350x250 with 1 Axes>
<Figure size 350x250 with 1 Axes>
<Figure size 350x250 with 1 Axes>
<Figure size 350x250 with 1 Axes>
<Figure size 350x250 with 1 Axes>
<Figure size 350x250 with 1 Axes>
<Figure size 350x250 with 1 Axes>