How to fix "Tensor object is not callable" error in keras code?

Question

I am making a multi layer auto-encoder using Keras and I am getting the error where I make the Model of decoder.

I saw a matching problem in Stackoverflow but my code isn't incorrect and is complete, so not missing anything

encoding_dim=16


i=Input(shape=(122,))




encoded=Dense(64,activation='relu')(i)

encoded1=Dense(32,activation='relu')(encoded)

encoded2=Dense(16,activation='relu')(encoded1)


decoded=Dense(32,activation='relu')(encoded2)

decoded2 =Dense(64,activation='relu')(decoded)

decoded3 =Dense(122,activation='relu')(decoded2)

autoencoder = Model(i, decoded3)


ec = Model(i,encoded)

encoded_input=Input(shape=(encoding_dim,))

decoder_layer=autoencoder.layers[-3](encoded_input)
decoder_layer=autoencoder.layers[-2](decoder_layer)
decoder_layer=autoencoder.layers[-1](decoder_layer)


**decoder=Model(encoded_input,decoder_layer(encoded_input))**

OverLordGoldDragon · Accepted Answer · 2019-10-11 19:18:09Z

1

The problem is here:

decoder = Model(encoded_input, decoder_layer(encoded_input))

decoder_layer                 # <-- is a LAYER
decoder_layer(encoded_input)  # <-- is a TENSOR

Model expects inputs= and outputs= to be both layers, not tensors; a layer fed an input evalutes to a tensor. Fix:

decoder = Model(encoded_input, decoder_layer)

answered Oct 11, 2019 at 19:18

OverLordGoldDragon

20k13 gold badges62 silver badges122 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

How to fix "Tensor object is not callable" error in keras code?

1 Answer 1

Comments

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Related