从零到一构建AI项目实战教程第六篇：构建基础神经网络模型

import numpy as np
import tensorflow as tf
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense
from tensorflow.keras.optimizers import Adam
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.metrics import accuracy_score

# 生成示例数据
np.random.seed(42)
X = np.random.rand(1000, 2)  # 1000个样本，每个样本2个特征
y = (np.sum(X, axis=1) > 1).astype(int)  # 标签：特征之和大于1的为正类，否则为负类

# 数据预处理
scaler = StandardScaler()
X_scaled = scaler.fit_transform(X)

# 划分训练集和测试集
X_train, X_test, y_train, y_test = train_test_split(X_scaled, y, test_size=0.2, random_state=42)

# 构建MLP模型
model = Sequential()
model.add(Dense(16, input_dim=2, activation='relu'))  # 输入层和第一个隐藏层，16个神经元，ReLU激活函数
model.add(Dense(8, activation='relu'))  # 第二个隐藏层，8个神经元，ReLU激活函数
model.add(Dense(1, activation='sigmoid'))  # 输出层，1个神经元，Sigmoid激活函数（用于二分类）

# 编译模型
model.compile(optimizer=Adam(learning_rate=0.001), loss='binary_crossentropy', metrics=['accuracy'])

# 训练模型
model.fit(X_train, y_train, epochs=50, batch_size=32, validation_split=0.2)

# 评估模型
y_pred = (model.predict(X_test) > 0.5).astype(int).flatten()  # 将预测概率转换为标签
accuracy = accuracy_score(y_test, y_pred)
print(f'Test Accuracy: {accuracy:.4f}')