使用PyTorch/TensorFlow搭建简单全连接神经网络

import torch
import torch.nn as nn
import torch.optim as optim
 
# 定义一个简单的全连接神经网络
class SimpleFCNN(nn.Module):
    def __init__(self, input_size, hidden_size, output_size):
        super(SimpleFCNN, self).__init__()
        self.fc1 = nn.Linear(input_size, hidden_size)  # 输入层到隐藏层
        self.relu = nn.ReLU()                           # 激活函数
        self.fc2 = nn.Linear(hidden_size, output_size)  # 隐藏层到输出层
 
    def forward(self, x):
        out = self.fc1(x)
        out = self.relu(out)
        out = self.fc2(out)
        return out
 
# 超参数设置
input_size = 10      # 输入特征数量
hidden_size = 5      # 隐藏层节点数
output_size = 1      # 输出节点数
learning_rate = 0.001
num_epochs = 100
 
# 实例化模型、定义损失函数和优化器
model = SimpleFCNN(input_size, hidden_size, output_size)
criterion = nn.MSELoss()  # 均方误差损失，用于回归问题
optimizer = optim.Adam(model.parameters(), lr=learning_rate)
 
# 构造一些模拟数据
x_train = torch.randn(100, input_size)
y_train = torch.randn(100, output_size)
 
# 训练模型
for epoch in range(num_epochs):
    outputs = model(x_train)
    loss = criterion(outputs, y_train)
 
    optimizer.zero_grad()  # 梯度清零
    loss.backward()        # 反向传播
    optimizer.step()       # 更新参数
 
    if (epoch + 1) % 10 == 0:
        print(f'Epoch [{epoch + 1}/{num_epochs}], Loss: {loss.item():.4f}')

import tensorflow as tf
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense
 
# 超参数设置
input_size = 10      # 输入特征数量
hidden_size = 5      # 隐藏层节点数
output_size = 1      # 输出节点数
learning_rate = 0.001
num_epochs = 100
batch_size = 10
 
# 构建顺序模型
model = Sequential([
    Dense(hidden_size, activation='relu', input_shape=(input_size,)),  # 输入层到隐藏层
    Dense(output_size)                                                 # 输出层
])
 
# 编译模型：指定优化器和损失函数（这里依然使用均方误差）
model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=learning_rate),
              loss='mse')
 
# 构造一些模拟数据
import numpy as np
x_train = np.random.randn(100, input_size)
y_train = np.random.randn(100, output_size)
 
# 训练模型
history = model.fit(x_train, y_train, epochs=num_epochs, batch_size=batch_size, verbose=2)