The layer sequential has never been called and thus has no defined input error when trying to extract layers from sequential model

Question

I am trying to extract the layers from a sequential model to build an autoencoder. I trained the model on some data but when I try to get model.input from my model I get an error saying that it has never been called.

I set up and train the model here.

from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Input
from tensorflow.keras.optimizers import Adam

model = Sequential()
model.add(Input(shape=(X_train_scaled.shape[1],)))  # Input layer
model.add(Dense(128, activation='relu'))  # First hidden layer
model.add(Dense(64, activation='relu'))  # Second hidden layer
model.add(Dense(1))  # Output layer

optimiser = Adam(learning_rate = 0.001)
model.compile(optimizer = optimiser, loss='mse')

history = model.fit(X_train_scaled, y_train, validation_data=(X_test_scaled, y_test), epochs=100, batch_size=16)

And the following line causes the error.

from tensorflow.keras.models import Model

encoder = Model(inputs=model.input, outputs=model.layers[-2].output)  # -2 to exclude the last layer

I am very new to this, so any advice or direction would be appreciated.

Sagar · Accepted Answer · 2024-11-08 10:18:19Z

3

You are creating initial model using Sequential API and trying to access the input attribute of the model on another model via functional API. As sequential model don't have an accessible input attribute. To resolve this you need to define the input implicitly by using model.layers[0].input or model.get_layer(index=0).input or else switch the sequential model to functional model. please refer this gist.

answered Nov 8, 2024 at 10:18

Sagar

4613 silver badges10 bronze badges

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

Collectives™ on Stack Overflow

The layer sequential has never been called and thus has no defined input error when trying to extract layers from sequential model

1 Answer 1

Comments

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Related