AttributeError: 'Tensor' object does not have '_keras_history' attribute

Question

I looked for the whole “Tensor” object does not have the *** attribute, but none of them are related to Keras (except for TensorFlow: AttributeError: the Tensor object does not have the attribute 'log10' , which did not help) ...

I make a kind of GAN (Generative Adversarial Networks). Here you can find the structure.

Layer (type)                     Output Shape          Param #         Connected to                     
_____________________________________________________________________________
input_1 (InputLayer)             (None, 30, 91)        0                                            
_____________________________________________________________________________
model_1 (Model)                  (None, 30, 1)         12558           input_1[0][0]                    
_____________________________________________________________________________
model_2 (Model)                  (None, 30, 91)        99889           input_1[0][0]                    
                                                                       model_1[1][0]                    
_____________________________________________________________________________
model_3 (Model)                  (None, 1)             456637          model_2[1][0]                    
_____________________________________________________________________________

I previewed model_2 and model_3. The fact is that I previewed model_2 with a list of 0 and 1, but the return values ​​of model_1 are returned. So I looked at rounding model1_output with the following code: K.round () on model1_out.

import keras.backend as K
[...]
def make_gan(GAN_in, model1, model2, model3):
    model1_out = model1(GAN_in)
    model2_out = model2([GAN_in, K.round(model1_out)])
    GAN_out = model3(model2_out)
    GAN = Model(GAN_in, GAN_out)
    GAN.compile(loss=loss, optimizer=model1.optimizer, metrics=['binary_accuracy'])
    return GAN
[...]

I have the following error:

AttributeError: The tensor object does not have the _keras_history attribute

Full trace:

Traceback (most recent call last):
  File "C:\Users\Asmaa\Documents\BillyValuation\GFD.py", line 88, in <module>
GAN = make_gan(inputSentence, G, F, D)
  File "C:\Users\Asmaa\Documents\BillyValuation\GFD.py", line 61, in make_gan
GAN = Model(GAN_in, GAN_out)
  File "C:\ProgramData\Anaconda3\lib\site-packages\keras\legacy\interfaces.py", line 88, in wrapper
return func(*args, **kwargs)
  File "C:\ProgramData\Anaconda3\lib\site-packages\keras\engine\topology.py", line 1705, in __init__
build_map_of_graph(x, finished_nodes, nodes_in_progress)
  File "C:\ProgramData\Anaconda3\lib\site-packages\keras\engine\topology.py", line 1695, in build_map_of_graph
layer, node_index, tensor_index)
  File "C:\ProgramData\Anaconda3\lib\site-packages\keras\engine\topology.py", line 1695, in build_map_of_graph
layer, node_index, tensor_index)
  File "C:\ProgramData\Anaconda3\lib\site-packages\keras\engine\topology.py", line 1665, in build_map_of_graph
layer, node_index, tensor_index = tensor._keras_history
AttributeError: 'Tensor' object has no attribute '_keras_history'

Python 3.6 Spyder 3.1.4 Windows 7. TensorFlow Keras pip. !

Answer 1

"+", "" keras

Answer 2

@'Maëva LC': , None.

model1_out = (lambda x: K.round(x), output_shape=...)(model1_out)

- . , .

round() , None. .

Answer 3

:

Traceback (most recent call last):
  File "C:\Users\Asmaa\Documents\BillyValuation\GFD.py", line 88, in <module>
GAN = make_gan(inputSentence, G, F, D)
  File "C:\Users\Asmaa\Documents\BillyValuation\GFD.py", line 61, in make_gan
GAN = Model(GAN_in, GAN_out)

, , .

In your sample code, please check line by line, regardless of whether you are using the operation without Keras , especially in the last few lines. For example, for elementary additions, you can intuitively use +or even numpy.add, but you should use it instead keras.layers.Add().

Answer 4

Try the following:

def make_gan(GAN_in, model1, model2, model3):
    model1_out = model1(GAN_in)
    model1_out = Lambda(lambda x: K.round(x), output_shape=...)(model1_out)
    model2_out = model2([GAN_in, model1_out])
    GAN_out = model3(model2_out)
    GAN = Model(GAN_in, GAN_out)
    GAN.compile(loss=loss, optimizer=model1.optimizer, metrics=['binary_accuracy'])
    return GAN