Training history and visualization in TensorFlow - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to train the model and save the training history.

TensorFlow

history = model.fit(X_train, y_train, epochs=[1], validation_split=0.2)

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

A100

B5

C50

D10

Attempts:

3 left

💡 Hint

Common Mistakes

Using too many epochs can cause long training times.

Forgetting to set epochs causes an error.

The epochs parameter sets how many times the model will see the entire training data. Here, 10 epochs is a common choice for a simple training.

2fill in blank

medium

Complete the code to plot training accuracy from the history object.

TensorFlow

plt.plot(history.history['[1]'])

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

Aaccuracy

Bval_accuracy

Cval_loss

Dloss

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'loss' instead of 'accuracy' to plot accuracy.

Using validation keys when plotting training metrics.

The history.history dictionary stores metrics by their names. To plot training accuracy, use the key 'accuracy'.

3fill in blank

hard

Fix the error in the code to plot validation loss correctly.

TensorFlow

plt.plot(history.history['[1]'])

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

Aval_loss

Bval_accuracy

Closs_val

Dval_acc

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'val_acc' or 'loss_val' which are invalid keys.

Mixing validation accuracy key with loss key.

The correct key for validation loss in history.history is 'val_loss'. Other keys will cause errors or wrong plots.

4fill in blank

hard

Fill both blanks to plot training and validation accuracy.

TensorFlow

plt.plot(history.history['[1]'], label='train')
plt.plot(history.history['[2]'], label='validation')
plt.legend()

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

Aaccuracy

Bval_accuracy

Closs

Dval_loss

Attempts:

3 left

💡 Hint

Common Mistakes

Plotting loss keys instead of accuracy keys.

Mixing training loss with validation accuracy.

To compare training and validation accuracy, plot 'accuracy' and 'val_accuracy' from the history.

5fill in blank

hard

Fill all three blanks to create a dictionary of training loss, validation loss, and epochs.

TensorFlow

results = {
  '[1]': history.history['loss'],
  '[2]': history.history['val_loss'],
  '[3]': list(range(1, len(history.history['loss']) + 1))
}

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

Atrain_loss

Bvalidation_loss

Cepochs

Dloss

Attempts:

3 left

💡 Hint

Common Mistakes

Using incorrect or inconsistent dictionary keys.

Not creating the epochs list correctly.

This dictionary stores training loss as 'train_loss', validation loss as 'validation_loss', and the epoch numbers as 'epochs'.