L1 and L2 regularization in TensorFlow - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to add L2 regularization to a Dense layer in TensorFlow.

TensorFlow

layer = tf.keras.layers.Dense(64, kernel_regularizer=tf.keras.regularizers.[1](0.01))

Drag options to blanks, or click blank then click option'

Al2

Bl1_l2

Cl1

Dnone

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'l1' instead of 'l2' for L2 regularization.

Forgetting to specify the regularization parameter inside the parentheses.

The l2 function applies L2 regularization, which adds a penalty proportional to the square of the weights.

2fill in blank

medium

Complete the code to add L1 regularization with strength 0.005 to a Dense layer.

TensorFlow

layer = tf.keras.layers.Dense(32, kernel_regularizer=tf.keras.regularizers.[1](0.005))

Drag options to blanks, or click blank then click option'

Al1

Bl2

Cl1_l2

Dnone

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'l2' instead of 'l1' for L1 regularization.

Not specifying the regularization strength correctly.

The l1 function applies L1 regularization, which adds a penalty proportional to the absolute value of the weights.

3fill in blank

hard

Fix the error in the code to correctly apply both L1 and L2 regularization.

TensorFlow

layer = tf.keras.layers.Dense(128, kernel_regularizer=tf.keras.regularizers.[1](l1=0.001, l2=0.002))

Drag options to blanks, or click blank then click option'

Al2

Bl1_l2

Cnone

Dl1

Attempts:

3 left

💡 Hint

Common Mistakes

Using only 'l1' or 'l2' when both are needed.

Not passing both l1 and l2 parameters.

The l1_l2 function allows applying both L1 and L2 regularization together by specifying both penalties.

4fill in blank

hard

Fill both blanks to create a model with L1 regularization on the first layer and L2 on the second.

TensorFlow

model = tf.keras.Sequential([
    tf.keras.layers.Dense(64, kernel_regularizer=tf.keras.regularizers.[1](0.01), activation='relu'),
    tf.keras.layers.Dense(10, kernel_regularizer=tf.keras.regularizers.[2](0.02), activation='softmax')
])

Drag options to blanks, or click blank then click option'

Al1

Bl2

Cl1_l2

Dnone

Attempts:

3 left

💡 Hint

Common Mistakes

Mixing up L1 and L2 regularizers.

Using the same regularizer for both layers.

The first layer uses l1 regularization to encourage sparsity, and the second uses l2 to penalize large weights.

5fill in blank

hard

Fill all three blanks to create a dictionary comprehension that maps layer names to their regularization type strings ('L1', 'L2', or 'L1_L2').

TensorFlow

layers = {'dense1': tf.keras.regularizers.l1(0.01), 'dense2': tf.keras.regularizers.l2(0.02), 'dense3': tf.keras.regularizers.l1_l2(l1=0.001, l2=0.002)}
reg_types = {name: [1] for name, reg in layers.items() if isinstance(reg, [2])}
reg_types.update({name: 'L2' for name, reg in layers.items() if isinstance(reg, [3])})

Drag options to blanks, or click blank then click option'

A'L1'

Btf.keras.regularizers.L1

Ctf.keras.regularizers.L2

Dtf.keras.regularizers.L1L2

Attempts:

3 left

💡 Hint

Common Mistakes

Using incorrect class names for isinstance checks.

Putting the string label in the wrong blank.

This code maps layer names to strings describing their regularization type by checking the instance type of each regularizer.