參考

3.10. 多重感知機的簡潔實現

import torch
from torch import nn
from torch.nn import init
import numpy as np
import sys
sys.path.append("..") 
import d2lzh_pytorch as d2l

3.10.1. 定義模型

num_inputs, num_outputs, num_hiddens = 784, 10, 256
# 參數都存在 net.parameters()中
net = nn.Sequential(d2l.FlattenLayer(),nn.Linear(num_inputs, num_hiddens),nn.ReLU(),nn.Linear(num_hiddens, num_outputs),
)# 初始化權重值
for param in net.parameters():init.normal_(param, mean=0, std=0.01)

3.10.2 讀取數據并訓練模型

batch_size = 256
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size)
loss = torch.nn.CrossEntropyLoss()optimizer = torch.optim.SGD(net.parameters(), lr = 0.5)num_epochs = 5
d2l.train_ch3(net, train_iter, test_iter, loss, num_epochs, batch_size, None, None, optimizer)

### 3.10.3 預測# 測試
X, y = iter(test_iter).next()true_labels = d2l.get_fashion_mnist_labels(y.numpy())
pred_labels = d2l.get_fashion_mnist_labels(net(X).argmax(dim=1).numpy())
titles = [true + '\n' + pred for true, pred in zip(true_labels, pred_labels)]d2l.show_fashion_mnist(X[0:9], titles[0:9])