Champion-challenger model comparison in MLOps - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to load the champion model from the model registry.

MLOps

champion_model = model_registry.[1]('champion')

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

Aload

Bdelete

Csave

Dregister

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'save' instead of 'load' will try to store a model, not retrieve it.

Using 'delete' removes the model, which is not desired here.

The load method retrieves the champion model from the registry for evaluation or deployment.

2fill in blank

medium

Complete the code to compare the challenger model's accuracy with the champion's.

MLOps

if challenger_accuracy [1] champion_accuracy:

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

A<

B==

C<=

D>

Attempts:

3 left

💡 Hint

Common Mistakes

Using '<' would check if challenger is worse, which is not the goal.

Using '==' only checks equality, not improvement.

We check if the challenger model's accuracy is greater than the champion's to decide if it should replace the champion.

3fill in blank

hard

Fix the error in the code that registers the challenger model as the new champion.

MLOps

model_registry.[1](challenger_model, name='champion')

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

Aload

Bregister

Cdelete

Dsave

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'save' might store the model but not update the registry name.

Using 'load' tries to retrieve a model, not register it.

The register method is used to add or update a model in the registry under a specific name.

4fill in blank

hard

Fill both blanks to create a function that evaluates model performance and decides if the challenger should replace the champion.

MLOps

def should_replace(challenger_score, champion_score):
    return challenger_score [1] champion_score and challenger_score [2] 0.8

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

A>

B<

C>=

D<=

Attempts:

3 left

💡 Hint

Common Mistakes

Using '<' in the first blank would reverse the logic.

Using '<=' in the second blank would allow scores below 0.8.

The function returns True if the challenger score is greater than the champion's and also meets a minimum threshold of 0.8 or higher.

5fill in blank

hard

Fill all three blanks to update the model registry only if the challenger outperforms the champion and meets quality standards.

MLOps

if challenger_metrics['accuracy'] [1] champion_metrics['accuracy'] and challenger_metrics['f1_score'] [2] 0.75:
    model_registry.[3](challenger_model, name='champion')

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

A>

B>=

Cregister

Dsave

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'save' instead of 'register' may not update the champion label.

Using '>=' in the first blank would allow equal accuracy, not strictly better.

The code checks if the challenger accuracy is greater than the champion's, the F1 score is at least 0.75, then registers the challenger as the new champion.