adding_regularization.py

def shallow_cnn(x):
  # Reshape to use within a convolutional neural net
  x_image = tf.reshape(x, [-1, data_width, data_height, 1])
  
  # Define Regularizer
  regularizer = tf.keras.constraints.MaxNorm(max_value=2)

  # Convolutional Layer #1
  W_conv1 = weight_variable([5, 5, 1, 32])
  b_conv1 = bias_variable([32])
  
  h_conv1 = tf.nn.relu(tf.nn.conv2d(x_image, regularizer.__call__(w=W_conv1), 
      strides=[1, 1, 1, 1], padding='SAME') + b_conv1)

  # Pooling Layer #2
  h_pool1 = tf.nn.max_pool(h_conv1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')

  # Convolutional Layer #2
  W_conv2 = weight_variable([5, 5, 32, 64])
  b_conv2 = bias_variable([64])
  
  h_conv2 = tf.nn.relu(tf.nn.conv2d(h_pool1, regularizer.__call__(w=W_conv2), 
      strides=[1, 1, 1, 1], padding='SAME') + b_conv2)
  
  # Batch Norm
  batch_norm = tf.layers.batch_normalization(h_conv2,momentum=0.1,epsilon=1e-5)

  # Pooling Layer #2
  h_pool2 = tf.nn.max_pool(batch_norm, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')

  # Dense Layer
  W_fc1 = weight_variable([7 * 7 * 64, 1024])
  b_fc1 = bias_variable([1024])
  
  # Flatten
  h_pool1_flat = tf.reshape(h_pool2, [-1, 7 * 7 * 64])
  h_fc1 = tf.nn.relu(tf.matmul(h_pool1_flat, W_fc1) + b_fc1)
  
  # Dropout
  keep_prob = tf.placeholder(tf.float32)
  h_fc1_drop = tf.nn.dropout(h_fc1, keep_prob)

  # Logits Layer
  W_fc2 = weight_variable([1024, label_count])
  b_fc2 = bias_variable([label_count])
  
  # Max Norm
  regularizer = tf.keras.constraints.MaxNorm(max_value=0.5)
  
  y_conv = tf.nn.softmax(tf.matmul(h_fc1_drop, regularizer.__call__(w=W_fc2)) + b_fc2)

  return y_conv, keep_prob