keras_neural_net.py

# Create the deep learning layers
model = keras.models.Sequential()
model.add(keras.layers.Dense(256, activation='relu', input_shape=(train_df_scaled.shape[1],), name='raw'))

model.add(keras.layers.BatchNormalization())
model.add(keras.layers.Dense(128, activation='relu'))
model.add(keras.layers.BatchNormalization())
model.add(keras.layers.Dense(64, activation='relu'))
model.add(keras.layers.BatchNormalization())
model.add(keras.layers.Dense(32, activation='relu'))
model.add(keras.layers.BatchNormalization())
model.add(keras.layers.Dense(16, activation='relu'))
model.add(keras.layers.BatchNormalization())
model.add(keras.layers.Dense(1, name='predictions'))

adam = keras.optimizers.Adam(lr=LEARNING_RATE)
model.compile(loss='mse', optimizer=adam, metrics=['mae'])