python - 属性错误: 'tuple' object has no attribute 'layer' when trying transfer learning with keras

Question

我正在尝试使用 keras 模型进行迁移学习， 但在向模型添加新层时陷入困境。 我尝试过以下代码:

prev_model = load_model('final_model.h5') # loading the previously saved model.

new_model = Sequential()
new_model.add(prev_model)
new_model.add(Dense(256,activation='relu'))
new_model.add(Dropout(0.5))
new_model.add(Dense(1,activation='sigmoid'))

但得到:

TypeError: The added layer must be an instance of class Layer. Found: <tensorflow.python.keras.layers.core.Flatten object at 0x00000000B74364A8>

每当我使用 .add() 添加图层时都会发生这种情况。

然后我发现

number_of_layers_to_freeze = 10
vgg_model = VGG16(include_top=False)
for i in range(number_of_layers_to_freeze):
    vgg_model.layers[i].trainable = False
vgg_output = vgg_model.outputs[0]
output = keras.layers.Dense(10, activation="softmax")(vgg_output)

model = keras.models.Model(inputs=vgg_model.inputs, outputs=output)

在其他帖子。 但这会导致

 AttributeError: 'tuple' object has no attribute 'layer'

我目前正在使用

keras 2.2.5 
tensorflow-gpu 1.14.0

是否是版本冲突导致的？

<小时/>

完整回溯:(AttributeError: 'tuple' 对象没有属性 'layer')

    AttributeError                            Traceback (most recent call last)
<ipython-input-15-afcad6e65f32> in <module>
      4 #     vgg_model.layers[i].trainable = False
      5 vgg_output = conv_base.outputs[0]
----> 6 output = tensorflow.keras.layers.Dropout(dropout_rate, name="dropout_out")(vgg_output)
      7 
      8 model1 = models.Model(inputs=conv_base.inputs, outputs=output)

G:\ProgramData\Anaconda3\envs\tensorf\lib\site-packages\tensorflow\python\keras\engine\base_layer.py in __call__(self, inputs, *args, **kwargs)
    661               kwargs.pop('training')
    662             inputs, outputs = self._set_connectivity_metadata_(
--> 663                 inputs, outputs, args, kwargs)
    664           self._handle_activity_regularization(inputs, outputs)
    665           self._set_mask_metadata(inputs, outputs, previous_mask)

G:\ProgramData\Anaconda3\envs\tensorf\lib\site-packages\tensorflow\python\keras\engine\base_layer.py in _set_connectivity_metadata_(self, inputs, outputs, args, kwargs)
   1706     kwargs.pop('mask', None)  # `mask` should not be serialized.
   1707     self._add_inbound_node(
-> 1708         input_tensors=inputs, output_tensors=outputs, arguments=kwargs)
   1709     return inputs, outputs
   1710 

G:\ProgramData\Anaconda3\envs\tensorf\lib\site-packages\tensorflow\python\keras\engine\base_layer.py in _add_inbound_node(self, input_tensors, output_tensors, arguments)
   1793     """
   1794     inbound_layers = nest.map_structure(lambda t: t._keras_history.layer,
-> 1795                                         input_tensors)
   1796     node_indices = nest.map_structure(lambda t: t._keras_history.node_index,
   1797                                       input_tensors)

G:\ProgramData\Anaconda3\envs\tensorf\lib\site-packages\tensorflow\python\util\nest.py in map_structure(func, *structure, **kwargs)
    513 
    514   return pack_sequence_as(
--> 515       structure[0], [func(*x) for x in entries],
    516       expand_composites=expand_composites)
    517 

G:\ProgramData\Anaconda3\envs\tensorf\lib\site-packages\tensorflow\python\util\nest.py in <listcomp>(.0)
    513 
    514   return pack_sequence_as(
--> 515       structure[0], [func(*x) for x in entries],
    516       expand_composites=expand_composites)
    517 

G:\ProgramData\Anaconda3\envs\tensorf\lib\site-packages\tensorflow\python\keras\engine\base_layer.py in <lambda>(t)
   1792             `call` method of the layer at the call that created the node.
   1793     """
-> 1794     inbound_layers = nest.map_structure(lambda t: t._keras_history.layer,
   1795                                         input_tensors)
   1796     node_indices = nest.map_structure(lambda t: t._keras_history.node_index,

AttributeError: 'tuple' object has no attribute 'layer'

fulltraceback :(TypeError: 添加的图层必须是 Layer 类的实例。)

TypeError                                 Traceback (most recent call last)
<ipython-input-42-b5858637ba91> in <module>
      2 # model.add(prev_model)
----> 3 model.add(tensorflow.keras.layers.GlobalMaxPooling2D(name="gap"))
      4 model.add(Flatten(name="flatten"))
      5 if dropout_rate > 0:
      6     model.add(layers.Dropout(dropout_rate, name="dropout_out"))

G:\ProgramData\Anaconda3\envs\tensorf\lib\site-packages\keras\engine\sequential.py in add(self, layer)
    131             raise TypeError('The added layer must be '
    132                             'an instance of class Layer. '
--> 133                             'Found: ' + str(layer))
    134         self.built = False
    135         if not self._layers:

TypeError: The added layer must be an instance of class Layer. Found: <tensorflow.python.keras.layers.core.Flatten object at 0x00000000B74364A8>

Answer 1

TypeError: The added layer must be an instance of class Layer

在从 tensorflow.keras.layers 添加层时，您已使用 keras.models.Model 创建了模型。
请注意，keras 和 tensorflow.keras 是不同的。确保坚持其中之一。