str(_D + 1))(tmp_in)
else:
x = tf.keras.layers.Dense(N,
activation='relu',
name='dense_' +
str(_D + 1))(tmp_in)
xn = tf.keras.layers.BatchNormalization()(x)
xd = tf.keras.layers.Dropout(0.5)(xn)
tmp_in = xd
outputs = tf.keras.layers.Dense(1, name='predictions')(xd)
model = tf.keras.Model(inputs=inputs, outputs=outputs)
if fold_n == 0:
model.summary()
model.compile(optimizer=tf.keras.optimizers.Adam(),
loss=tf.keras.losses.MeanAbsoluteError(),
metrics=[tf.keras.metrics.MeanAbsoluteError()])
overfitCallback = tf.keras.callbacks.EarlyStopping(
monitor='val_loss',
min_delta=0,
patience=20,
restore_best_weights=True)
history = model.fit(X_train,
Y_train,
batch_size=64,
epochs=999,
verbose=0,
validation_data=(X_valid, Y_valid),
callbacks=[overfitCallback])
Y_pred_valid = model.predict(X_valid).reshape(-1, )
