如何在keras中实现Conv1D输出与Conv2D输出的结合

from keras.models import Model
from keras.layers import Input, Conv1D, Conv2D, MaxPooling1D, MaxPooling2D, concatenate

input_shape = (input_length, input_dim)  # 输入数据的形状
input_layer = Input(shape=input_shape)  # 输入层

conv1d_layer = Conv1D(filters=32, kernel_size=3, activation='relu')(input_layer)  # Conv1D层
conv2d_layer = Conv2D(filters=64, kernel_size=(3, 3), activation='relu')(input_layer)  # Conv2D层

merged_layer = concatenate([conv1d_layer, conv2d_layer], axis=1)  # 合并层

maxpool1d_layer = MaxPooling1D(pool_size=2)(merged_layer)  # MaxPooling1D层
maxpool2d_layer = MaxPooling2D(pool_size=(2, 2))(merged_layer)  # MaxPooling2D层

# 添加其他层...

# 最后，定义模型并编译
model = Model(inputs=input_layer, outputs=output_layer)
model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])