两个串联神经网络训练代码

下面是两个串联神经网络训练的代码示例，一个是基于PyTorch实现的，另一个是基于Keras实现的：

PyTorch代码：

import torch
import torch.nn as nn
import torch.optim as optim

# 定义第一个神经网络
class Net1(nn.Module):
    def __init__(self):
        super(Net1, self).__init__()
        self.fc1 = nn.Linear(10, 20)
        self.fc2 = nn.Linear(20, 10)

    def forward(self, x):
        x = self.fc1(x)
        x = nn.functional.relu(x)
        x = self.fc2(x)
        return x

# 定义第二个神经网络
class Net2(nn.Module):
    def __init__(self):
        super(Net2, self).__init__()
        self.fc1 = nn.Linear(10, 20)
        self.fc2 = nn.Linear(20, 5)

    def forward(self, x):
        x = self.fc1(x)
        x = nn.functional.relu(x)
        x = self.fc2(x)
        return x

# 定义串联神经网络
class ConcatNet(nn.Module):
    def __init__(self):
        super(ConcatNet, self).__init__()
        self.net1 = Net1()
        self.net2 = Net2()

    def forward(self, x):
        x = self.net1(x)
        x = self.net2(x)
        return x

# 训练串联神经网络
net = ConcatNet()
criterion = nn.MSELoss()
optimizer = optim.SGD(net.parameters(), lr=0.01)

for epoch in range(100):
    # 生成训练数据
    inputs = torch.randn(100, 10)
    labels = torch.randn(100, 5)

    # 前向传播
    outputs = net(inputs)

    # 计算损失
    loss = criterion(outputs, labels)

    # 反向传播
    optimizer.zero_grad()
    loss.backward()
    optimizer.step()

    print('Epoch [%d/%d], Loss: %.4f' % (epoch+1, 100, loss.item()))

Keras代码：

import numpy as np
from keras.models import Sequential
from keras.layers import Dense

# 定义第一个神经网络
model1 = Sequential()
model1.add(Dense(units=20, activation='relu', input_dim=10))
model1.add(Dense(units=10, activation='relu'))

# 定义第二个神经网络
model2 = Sequential()
model2.add(Dense(units=20, activation='relu', input_dim=10))
model2.add(Dense(units=5, activation='relu'))

# 定义串联神经网络
model = Sequential()
model.add(model1)
model.add(model2)

# 编译模型
model.compile(loss='mse', optimizer='sgd')

# 训练模型
for epoch in range(100):
    # 生成训练数据
    inputs = np.random.randn(100, 10)
    labels = np.random.randn(100, 5)

    # 训练模型
    loss = model.train_on_batch(inputs, labels)

    print('Epoch [%d/%d], Loss: %.4f' % (epoch+1, 100, loss))