參考

3.6 softmax回歸的從零開始實現

import torch
import torchvision
import numpy as np
import sys
sys.path.append("..")
import d2lzh_pytorch as d2l

3.6.1. 獲取和讀取數據

batch_size = 256
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size)

3.6.2. 初始化模型參數

num_inputs = 784
num_outputs = 10W = torch.tensor(np.random.normal(0, 0.01, (num_inputs, num_outputs)), dtype=torch.float)  # torch.Size([784, 10])
b = torch.zeros(num_outputs, dtype=torch.float)   # torch.Size([10])# 同之前一樣,我們需要模型參數梯度。
W.requires_grad_(requires_grad=True)
b.requires_grad_(requires_grad=True)

3.6.3. 實現softmax運算

def softmax(X):X_exp = X.exp()partition = X_exp.sum(dim=1, keepdim=True)return X_exp / partition

3.6.4. 定義模型

# 傳入特征,給出預測值
def net(X):return softmax(torch.mm(X.view((-1, num_inputs)), W) + b)

3.6.5. 定義損失函數

def cross_entropy(y_hat, y):return -torch.log(y_hat.gather(1, y.view(-1, 1)))

3.6.6. 計算分類準確率

def accuracy(y_hat, y):return (y_hat.argmax(dim=1) ==y).float().mean().item()def evaluate_accuracy(data_iter, net):acc_sum, n = 0.0, 0for X, y in data_iter:acc_sum += (net(X).argmax(dim=1) == y).float().sum().item()n += y.shape[0]return acc_sum /n

3.6.7. 訓練模型

• d2lzh

num_epochs, lr = 5, 0.1# 本函數已保存在d2lzh包中方便以后使用
def train_ch3(net, train_iter, test_iter, loss, num_epochs, batch_size,params=None, lr=None, optimizer=None):for epoch in range(num_epochs):train_l_sum, train_acc_sum, n = 0.0, 0.0, 0for X, y in train_iter:y_hat = net(X)l = loss(y_hat, y).sum()# 梯度清零if optimizer is not None:optimizer.zero_grad()elif params is not None and params[0].grad is not None:for param in params:param.grad.data.zero_()l.backward()if optimizer is None:d2l.sgd(params, lr, batch_size)else:optimizer.step()  # “softmax回歸的簡潔實現”一節將用到train_l_sum += l.item()train_acc_sum += (y_hat.argmax(dim=1) == y).sum().item()n += y.shape[0]test_acc = evaluate_accuracy(test_iter, net)print('epoch %d, loss %.4f, train acc %.3f, test acc %.3f'% (epoch + 1, train_l_sum / n, train_acc_sum / n, test_acc))train_ch3(net, train_iter, test_iter, cross_entropy, num_epochs, batch_size, [W, b], lr)

3.6.8. 預測

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])