CODEWITHSUNDEEP
machine-learningdeep-learningkerasconv-neural-network
Akhilesh
Akhilesh

Reputation: 1034

How to add two sequential model in keras?

I have two different model, one is only hidden layer and another model is only dense layer.

I can load those model like 

model_HL = load_model('model_hl.hdf5')
model_DL = load_model('model_dl.hdf5')

and can use it as

output_HL = model_HL.predict (input)
output_HL_flatten = features.reshape((output_HL.shape[0],np.prod(self.outputShape)))
output_DL = model_DL.predict (output_HL_flatten)

But now my requirement has changed. I want to add those model in such a way that I can use it like 

output_DL = model.predict (input)

Please help me to do the same.

Upvotes: 1

Views: 1492

Answers (1)

Jakub Bartczuk
Jakub Bartczuk

Reputation: 2378

You can easily define new model in Keras using the fact that models can be composed like layers:

from keras.models import Model

composed_model = Model(
    inputs=[model_HL.input],
    outputs=[model_DL(model_HL.output)]
)

Upvotes: 1

Related Questions