Why regularization prevents overfitting in TensorFlow - Test Your Understanding

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'

Al1

Bl2

Cl1_l2

Ddropout

Attempts:

3 left

💡 Hint

Common Mistakes

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

L2 regularization adds a penalty proportional to the square of the weights, helping to prevent overfitting by discouraging large weights.

2fill in blank

medium

Complete the code to compile the model with mean squared error loss and Adam optimizer.

TensorFlow

model.compile(optimizer='[1]', loss='mse', metrics=['mae'])

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

Aadagrad

Bsgd

Cadam

Drmsprop

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'sgd' which is less adaptive than Adam.

Adam optimizer is commonly used for training neural networks and works well with regularization.

3fill in blank

hard

Fix the error in the code to add dropout regularization after a Dense layer.

TensorFlow

model.add(tf.keras.layers.Dense(128, activation='relu'))
model.add(tf.keras.layers.[1](0.5))

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

ADropout

BFlatten

CBatchNormalization

DDense

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'Dense' instead of 'Dropout' for regularization.

Dropout randomly disables neurons during training to prevent overfitting.

4fill in blank

hard

Fill both blanks to create a dictionary comprehension that filters features with length greater than 3.

TensorFlow

filtered_features = {feature: data[feature] for feature in features if [1](feature) [2] 3}

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

Alen

B>

C<

Dsum

Attempts:

3 left

💡 Hint

Common Mistakes

Using '<' instead of '>' for filtering.

We use len(feature) to get the length and '>' to filter features longer than 3 characters.

5fill in blank

hard

Fill all three blanks to create a dictionary comprehension that includes features with values greater than 0.

TensorFlow

positive_features = { [1]: [2] for [3] in data if data[[3]] > 0 }

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

Afeature

Bdata[feature]

Dvalue

Attempts:

3 left

💡 Hint

Common Mistakes

Using different variable names inconsistently.

The dictionary comprehension uses 'feature' as key, 'data[feature]' as value, and iterates over 'feature' in data.