教程Python代碼如下：

# 1) Design model(input, output size, forward pass)

# 2) Construct loss and optimizer

# 3) Training loop 訓(xùn)練循環(huán)

#??- forward pass: compute prediction

#??- backward pass: gradients

#??- update weights

import torch

import torch.nn as nn?# 神經(jīng)網(wǎng)絡(luò)模塊

import numpy as np?# 進(jìn)行數(shù)據(jù)轉(zhuǎn)換

from sklearn import datasets?# 加載二進(jìn)制分類數(shù)據(jù)集

from sklearn.preprocessing import StandardScaler?# 縮放features

from sklearn.model_selection import train_test_split?#將訓(xùn)練集和測試集分離開

"""二進(jìn)制分類問題，根據(jù)輸入數(shù)據(jù)特征，預(yù)測乳腺癌"""

# 0) prepare data

bc = datasets.load_breast_cancer()?# 乳腺癌數(shù)據(jù)集

X, y = bc.data, bc.target

n_samples, n_features = X.shape

#print(n_samples, n_features)

"""train_test_split函數(shù)用于將矩陣隨機(jī)劃分為訓(xùn)練子集和測試子集，并返回劃分好的訓(xùn)練集測試集樣本和訓(xùn)練集測試集標(biāo)簽"""

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=1234)

# scale 縮放特征

sc = StandardScaler()

X_train = sc.fit_transform(X_train)

X_test = sc.transform(X_test)

X_train = torch.from_numpy(X_train.astype(np.float32))

X_test = torch.from_numpy(X_test.astype(np.float32))

y_train = torch.from_numpy(y_train.astype(np.float32))

y_test = torch.from_numpy(y_test.astype(np.float32))

# 重塑y_train

y_train = y_train.view(y_train.shape[0],1)?#現(xiàn)在y_train只有1行

y_test = y_test.view(y_test.shape[0],1)

# 1) model

"""f = wx + b, sigmoid at the end"""

class LogisticRegression(nn.Module):

??def __init__(self,n_input_features):

????super(LogisticRegression, self).__init__()

????self.linear = nn.Linear(n_input_features, 1)?#輸出僅一個(gè)值

??#前向傳遞，激活函數(shù)sigmoid

??def forward(self, x):

????y_predicted = torch.sigmoid(self.linear(x))

????return y_predicted

model = LogisticRegression(n_features)

# 2) loss and optimizer

learning_rate = 0.01

criterion = nn.BCELoss()?# 二分類交叉熵?fù)p失

optimizer = torch.optim.SGD(model.parameters(), lr=learning_rate)

# 3) training loop

num_epoch = 500

for epoch in range(num_epoch):

??# forward pass and loss

??y_predicted = model(X_train)

??loss = criterion(y_predicted,y_train)

??# backward pass

??loss.backward()

??# update weigths

??optimizer.step()

??# zero gradients

??optimizer.zero_grad()

??if(epoch+1) % 10 == 0:

????print(f'epoch: {epoch+1}, loss = {loss.item():.4f}')

# 評估模型

with torch.no_grad():

??y_predicted = model(X_test)

??y_predicted_cls = y_predicted.round()

??acc = y_predicted_cls.eq(y_test).sum() / float(y_test.shape[0])

??print(f'accuracy: {acc:.4f}')