Early stopping in TensorFlow - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the EarlyStopping callback from TensorFlow.

TensorFlow

from tensorflow.keras.callbacks import [1]

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

ATensorBoard

BReduceLROnPlateau

CEarlyStopping

DModelCheckpoint

Attempts:

3 left

💡 Hint

Common Mistakes

Importing the wrong callback like ModelCheckpoint or TensorBoard.

Forgetting to import from keras.callbacks.

The EarlyStopping callback is imported from tensorflow.keras.callbacks to stop training when a monitored metric stops improving.

2fill in blank

medium

Complete the code to create an EarlyStopping callback that monitors validation loss.

TensorFlow

early_stop = EarlyStopping(monitor='[1]', patience=3)

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

Aval_loss

Baccuracy

Closs

Dval_accuracy

Attempts:

3 left

💡 Hint

Common Mistakes

Monitoring training loss instead of validation loss.

Using accuracy instead of loss for early stopping.

We monitor val_loss to stop training when validation loss stops improving, which helps prevent overfitting.

3fill in blank

hard

Fix the error in the code to correctly use EarlyStopping in model.fit.

TensorFlow

model.fit(X_train, y_train, epochs=50, callbacks=[[1]])

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

AEarlyStopping()

Bearly_stop()

CEarlyStopping

Dearly_stop

Attempts:

3 left

💡 Hint

Common Mistakes

Calling the callback instance with parentheses.

Passing the class name instead of the instance.

The callback instance early_stop should be passed without parentheses in the list to callbacks. Using early_stop() calls the instance as a function, which is incorrect.

4fill in blank

hard

Fill both blanks to create EarlyStopping that stops if validation accuracy does not improve for 5 epochs and restores best weights.

TensorFlow

early_stop = EarlyStopping(monitor='[1]', patience=[2], restore_best_weights=True)

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

Aval_accuracy

Bval_loss

C3

D5

Attempts:

3 left

💡 Hint

Common Mistakes

Using validation loss instead of accuracy.

Setting patience too low or forgetting restore_best_weights.

We monitor val_accuracy to stop training if validation accuracy does not improve. Setting patience=5 waits 5 epochs before stopping. restore_best_weights=True loads the best model weights after stopping.

5fill in blank

hard

Fill all three blanks to create EarlyStopping monitoring validation loss with patience 4 and verbose output.

TensorFlow

early_stop = EarlyStopping(monitor='[1]', patience=[2], verbose=[3])

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

Aval_accuracy

Bval_loss

C4

D1

Attempts:

3 left

💡 Hint

Common Mistakes

Using verbose=0 which hides messages.

Monitoring training loss instead of validation loss.

The monitor is set to val_loss to watch validation loss. patience=4 waits 4 epochs before stopping. verbose=1 enables printing messages when early stopping occurs.