[Tensorflow 2][Keras] Week 3 - Assignment: Custom Layers

본 포스팅은 다음 과정을 정리 한 글입니다.

 

Custom Models, Layers, and Loss Functions with TensorFlow

www.coursera.org/specializations/tensorflow-advanced-techniques

 

지난시간 리뷰

2021.03.27 - [Artificial Intelligence/Keras] - [Tensorflow 2][Keras] Week 3 - Activating Custom Layers

[Tensorflow 2][Keras] Week 2 - Activating Custom Layers

Week 3 Assignment: Implement a Quadratic Layer

# Please uncomment all lines in this cell and replace those marked with `# YOUR CODE HERE`.
# You can select all lines in this code cell with Ctrl+A (Windows/Linux) or Cmd+A (Mac), then press Ctrl+/ (Windows/Linux) or Cmd+/ (Mac) to uncomment.



class SimpleQuadratic(Layer):

    def __init__(self, units=32, activation=None):
        '''Initializes the class and sets up the internal variables'''
        super(SimpleQuadratic, self).__init__()
        self.units = units
        self.activation = tf.keras.activations.get(activation)
    
    def build(self, input_shape):
        '''Create the state of the layer (weights)'''
        # a and b should be initialized with random normal, c (or the bias) with zeros.
        # remember to set these as trainable.
        a_init = tf.random_normal_initializer()
        self.a = tf.Variable(name="kernel", 
            initial_value=a_init(shape=(input_shape[-1], self.units),
                                 dtype='float32'),
            trainable=True)
        b_init = tf.random_normal_initializer()
        self.b = tf.Variable(name="kernel", 
            initial_value=b_init(shape=(input_shape[-1], self.units),
                                 dtype='float32'),
            trainable=True)
        c_init = tf.zeros_initializer()
        self.c = tf.Variable(name="bias", 
            initial_value=c_init(shape=(self.units,),
                                 dtype='float32'),
            trainable=True)
   
    def call(self, inputs):
        '''Defines the computation from inputs to outputs'''
        x_squared = tf.math.square(inputs)
        x_squared_times_a = tf.matmul(x_squared, self.a)
        x_times_b = tf.matmul(inputs, self.b)
        x2a_plus_xb_plus_c = x_squared_times_a + x_times_b + self.c
        activated_x2a_plus_xb_plus_c = self.activation(x2a_plus_xb_plus_c)
        return activated_x2a_plus_xb_plus_c

# THIS CODE SHOULD RUN WITHOUT MODIFICATION
# AND SHOULD RETURN TRAINING/TESTING ACCURACY at 97%+

mnist = tf.keras.datasets.mnist

(x_train, y_train),(x_test, y_test) = mnist.load_data()
x_train, x_test = x_train / 255.0, x_test / 255.0

model = tf.keras.models.Sequential([
  tf.keras.layers.Flatten(input_shape=(28, 28)),
  SimpleQuadratic(128, activation='relu'),
  tf.keras.layers.Dropout(0.2),
  tf.keras.layers.Dense(10, activation='softmax')
])

model.compile(optimizer='adam',
              loss='sparse_categorical_crossentropy',
              metrics=['accuracy'])

model.fit(x_train, y_train, epochs=5)
model.evaluate(x_test, y_test)