我想连接这两个模型

Question

我正在尝试连接以下两个模型：

input_layer = Input(shape=(227,227,3))
model1 = Sequential([
    Conv2D(20, kernel_size=(5,5), activation='relu' ),
    MaxPooling2D((2,2)),
    Conv2D(30, kernel_size=(3,3), activation='relu'),
    MaxPooling2D((2,2)),
    Conv2D(40, kernel_size=(3,3), activation='relu'),
    MaxPooling2D((2,2)),
    Conv2D(50, kernel_size=(3,3), activation='relu'),
    MaxPooling2D((2,2)),
    Conv2D(60, kernel_size=(3,3), activation='relu'),
    MaxPooling2D((2,2)),
    
])(input_layer)

model2 = Sequential([
    Conv2D(20, kernel_size=(5,5), activation='relu', dilation_rate=(3)),
    MaxPooling2D((2,2)),
    Conv2D(30, kernel_size=(3,3), activation='relu', dilation_rate=(2)),
    MaxPooling2D((2,2)),
    Conv2D(40, kernel_size=(3,3), activation='relu', dilation_rate=(2)),
    MaxPooling2D((2,2)),
    Conv2D(50, kernel_size=(3,3), activation='relu', dilation_rate=(1)),
    MaxPooling2D((2,2)),
    Conv2D(60, kernel_size=(3,3), activation='relu', dilation_rate=(1)),
    MaxPooling2D((2,2)),
])(input_layer)

merged_model = Concatenate()([model1, model2])
merged_model = Flatten()(merged_model)
merged_model = Dense(1024, activation='relu')(merged_model)
merged_model = Dense(4, activation='softmax')(merged_model)`

但是显示错误：

Concatenate

层需要具有匹配形状的输入，但连接轴除外。收到：input_shape=[(None, 5, 5, 60), (None, 4, 4, 60)]

我尝试了 ChatGPT，它要求我使用 Flatten 函数并展平模型 2，但随后它将转换为 KerasTensor 并且无法编译。我需要有关如何解决此问题或如何更改膨胀率以使两个输入形状变得相同的建议。 Chat GPT 给了我这个方法：`

model1 = Sequential([
    Conv2D(20, kernel_size=(5,5), activation='relu' ),
    MaxPooling2D((2,2)),
    Conv2D(30, kernel_size=(3,3), activation='relu'),
    MaxPooling2D((2,2)),
    Conv2D(40, kernel_size=(3,3), activation='relu'),
    MaxPooling2D((2,2)),
    Conv2D(50, kernel_size=(3,3), activation='relu'),
    MaxPooling2D((2,2)),
    Conv2D(60, kernel_size=(3,3), activation='relu'),
    MaxPooling2D((2,2)),  
])(input_layer)

model2 = Sequential([
    Conv2D(20, kernel_size=(5,5), activation='relu', dilation_rate=(3)),
    MaxPooling2D((2,2)),
    Conv2D(30, kernel_size=(3,3), activation='relu', dilation_rate=(2)),
    MaxPooling2D((2,2)),
    Conv2D(40, kernel_size=(3,3), activation='relu', dilation_rate=(2)),
    MaxPooling2D((2,2)),
    Conv2D(50, kernel_size=(3,3), activation='relu', dilation_rate=(1)),
    MaxPooling2D((2,2)),
    Conv2D(60, kernel_size=(3,3), activation='relu', dilation_rate=(1)),
    MaxPooling2D((2,2)),
])(input_layer)

model1 = Flatten()(model1)
model2 = Flatten()(model2)

merged_model = Concatenate()([model1, model2])
merged_model = Dense(1024, activation='relu')(merged_model)
merged_model = Dense(4, activation='softmax')(merged_model)`

Answer 1

我编译没有问题。这是代码：

import tensorflow as tf
from tensorflow.keras import layers
from tensorflow import keras

input_layer = layers.Input(shape=(227,227,3))
input_layer = layers.Input(shape=(227,227,3))
model1_out = keras.Sequential([
    layers.Conv2D(20, kernel_size=(5,5), activation='relu' ),
    layers.MaxPooling2D((2,2)),
    layers.Conv2D(30, kernel_size=(3,3), activation='relu'),
    layers.MaxPooling2D((2,2)),
    layers.Conv2D(40, kernel_size=(3,3), activation='relu'),
    layers.MaxPooling2D((2,2)),
    layers.Conv2D(50, kernel_size=(3,3), activation='relu'),
    layers.MaxPooling2D((2,2)),
    layers.Conv2D(60, kernel_size=(3,3), activation='relu'),
    layers.MaxPooling2D((2,2)),  
])(input_layer)

model2_out = keras.Sequential([
    layers.Conv2D(20, kernel_size=(5,5), activation='relu', dilation_rate=(3)),
    layers.MaxPooling2D((2,2)),
    layers.Conv2D(30, kernel_size=(3,3), activation='relu', dilation_rate=(2)),
    layers.MaxPooling2D((2,2)),
    layers.Conv2D(40, kernel_size=(3,3), activation='relu', dilation_rate=(2)),
    layers.MaxPooling2D((2,2)),
    layers.Conv2D(50, kernel_size=(3,3), activation='relu', dilation_rate=(1)),
    layers.MaxPooling2D((2,2)),
    layers.Conv2D(60, kernel_size=(3,3), activation='relu', dilation_rate=(1)),
    layers.MaxPooling2D((2,2)),
])(input_layer)

model1_out = layers.Flatten()(model1_out)
model2_out = layers.Flatten()(model2_out)

merged_out = layers.Concatenate()([model1_out, model2_out])
merged_out = layers.Dense(1024, activation='relu')(merged_out)
merged_out = layers.Dense(4, activation='softmax')(merged_out)

model = keras.Model(input_layer, merged_out)
model.compile(optimizer="adam", loss="sparse_categorical_crossentropy")