Contrast the code we used in the lecture with the one below…
Contrast the code we used in the lecture with the one below that cannot be used in a classification problem.1)inputs = tf.keras.layers.Dense(units=32, activation=’relu’, input_shape=) 2)hidden = tf.keras.layers.Dense(units=32, activation=’relu’) 3)hidden2 = tf.keras.layers.Dense(units=32, activation=’relu’) 4)outputs = tf.keras.layers.Dense(units=1, activation=’relu’) 5)model = tf.keras.Sequential()6)model.compile(loss=loss, optimizer=optimizer, metrics=) 7) history = model.fit(x_train, y_train, validation_split=0.2, epochs=16, batch_size=512, verbose=0)Which line prevent us from using the code above in a binary classification problem?