Tensorflow Keras -并行向多个模型层提供输入

import tensorflow as tf
from tensorflow.keras.layers import Input, Dense, Concatenate
from tensorflow.keras.models import Model

# 定义两个不同的输入
input_A = Input(shape=(64,), name='input_A')
input_B = Input(shape=(64,), name='input_B')

# 对输入A应用一个密集层
x = Dense(32, activation='relu')(input_A)

# 对输入B应用另一个密集层
y = Dense(32, activation='relu')(input_B)

# 将两个层的结果合并
combined = Concatenate()([x, y])

# 添加输出层
output = Dense(1, activation='sigmoid')(combined)

# 创建模型，指定输入和输出
model = Model(inputs=[input_A, input_B], outputs=output)

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

# 打印模型结构
model.summary()

import numpy as np

# 生成示例数据
data_A = np.random.random((1000, 64))
data_B = np.random.random((1000, 64))
labels = np.random.randint(2, size=(1000, 1))

# 训练模型
model.fit([data_A, data_B], labels, epochs=10, batch_size=32)