Feedforward Net convert from Python

Hi.

I have an example of a feedforward network written in Python using an ADAM optimizer which I want to replicate in Matlab. The basics are

network = models.Sequential()
network.add(layers.Dense(units=64, activation='relu', input_shape=(len(features.columns),)))
network.add(layers.Dense(units=32, activation='relu'))
network.add(layers.Dense(units=1, activation='sigmoid'))

network.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy']) 

es = EarlyStopping(monitor='val_loss', mode='min', verbose=0, patience=500)
mc = ModelCheckpoint('data/best_model.h5', monitor='val_loss', mode='min', verbose=2, save_best_only=True)

history = network.fit(train_features, train_target, 
            epochs=1000, verbose=0, batch_size=128, 
            validation_data=(test_features, test_target), callbacks=[es, mc]) 

I believe I cannot use the Adam optimizer in the feedforward function so can I directly convert this or woud I have to create some layers myself rather than use the feedforward function?

Answer 1

[Srivardhan Gadila]

You can train the above network in keras framework and import it to matlab using the importKerasLayers, importKerasNetwork functions.

Alternatively you can define the above network in matlab using the Deep Learning Layers in MATLAB and mention the 'adam' optimizer as the sovlerName in the trainingOptions.

Also refer to Deep Learning Toolbox — Examples.