多次使用不同输入的 ResNet50（权重共享）

Question

I would like to use the same ResNet50 multiple times with different inputs, ie weights shared. Below is my coce but I'm getting the error message AttributeError: 'Tensor' object has no attribute 'output' for the line resnet_x = resnet_x.output .

What do I have to change to make it work?

from tensorflow.keras.applications.resnet50 import ResNet50
from tensorflow.keras.layers import Input
from tensorflow.keras.layers import GlobalAveragePooling2D
import tensorflow as tf
from tensorflow.keras.models import Model
from tensorflow.keras.optimizers import Adam

input_tensor_x = Input(shape=(96,96,3))
input_tensor_y = Input(shape=(96,96,3))
input_tensor_z = Input(shape=(96,96,3))

base_model = ResNet50(weights=None, include_top=False, input_shape=(96,96,3))
resnet_x = base_model(input_tensor_x)
resnet_x = resnet_x.output
resnet_x = GlobalAveragePooling2D()(resnet_x)
resnet_x = Dropout(0.5)(resnet_x)

resnet_y = base_model(input_tensor_y)
resnet_y = resnet_y.output
resnet_y = GlobalAveragePooling2D()(resnet_y)
resnet_y = Dropout(0.5)(resnet_y)

resnet_z = base_model(input_tensor_z)
resnet_y = resnet_y.output
resnet_y = GlobalAveragePooling2D()(resnet_y)
resnet_y = Dropout(0.5)(resnet_y)

merge_layer = tf.keras.layers.Concatenate()([resnet_x, resnet_y, resnet_z])

output_tensor = Dense(self.num_classes, activation='softmax')(merge_layer)

# instantiate and compile model
cnn_model = Model(inputs=[input_tensor_x, input_tensor_y, input_tensor_z], outputs=output_tensor)
opt = Adam()
cnn_model.compile(loss=categorical_crossentropy, optimizer=opt, metrics=['accuracy', tf.keras.metrics.AUC()])

Answer 1

simply removing the lines resnet_XXX = resnet_XXX.output does the job. pay attention to the name of the variables (below resnet_z layer)

input_tensor_x = Input(shape=(96,96,3))
input_tensor_y = Input(shape=(96,96,3))
input_tensor_z = Input(shape=(96,96,3))

base_model = ResNet50(weights=None, include_top=False, input_shape=(96,96,3))
resnet_x = base_model(input_tensor_x)
resnet_x = GlobalAveragePooling2D()(resnet_x)
resnet_x = Dropout(0.5)(resnet_x)

resnet_y = base_model(input_tensor_y)
resnet_y = GlobalAveragePooling2D()(resnet_y)
resnet_y = Dropout(0.5)(resnet_y)

resnet_z = base_model(input_tensor_z)
resnet_z = GlobalAveragePooling2D()(resnet_z)
resnet_z = Dropout(0.5)(resnet_z)

merge_layer = tf.keras.layers.Concatenate()([resnet_x, resnet_y, resnet_z])

output_tensor = Dense(10, activation='softmax')(merge_layer)

# instantiate and compile model
cnn_model = Model(inputs=[input_tensor_x, input_tensor_y, input_tensor_z], outputs=output_tensor)
opt = Adam()
cnn_model.compile(loss='categorical_crossentropy', optimizer=opt, metrics=['accuracy', tf.keras.metrics.AUC()])