Concept drift detection in MLOps - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the library used for concept drift detection.

MLOps

from river import [1]

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

Ametrics

Bdrift

Cdatasets

Dstream

Attempts:

3 left

💡 Hint

Common Mistakes

Importing unrelated modules like datasets or metrics.

The drift module in River library provides tools for concept drift detection.

2fill in blank

medium

Complete the code to create a drift detector object for detecting changes in data streams.

MLOps

detector = drift.[1]()

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

AADWIN

BKMeans

CRandomForest

DPCA

Attempts:

3 left

💡 Hint

Common Mistakes

Choosing clustering or classification algorithms instead of drift detectors.

ADWIN is a popular adaptive windowing method used for concept drift detection.

3fill in blank

hard

Fix the error in the code to correctly update the drift detector with new data points.

MLOps

for value in data_stream:
    detector.[1](value)

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

Aadd

Bfit

Cupdate

Dpredict

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'fit' or 'predict' which are not methods of the detector for updating data.

The update method is used to feed new data points to the drift detector.

4fill in blank

hard

Fill both blanks to check if the detector has detected a change and print a message.

MLOps

if detector.[1]():
    print('Concept drift [2]!')

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

Achange_detected

Bdetected

Coccurred

Dalert

Attempts:

3 left

💡 Hint

Common Mistakes

Using incorrect method names or message words that don't fit the context.

The method change_detected() returns True if drift is detected. The message is 'Concept drift detected!'.

5fill in blank

hard

Fill all three blanks to create a dictionary that maps feature names to their current values only if drift is detected.

MLOps

if detector.change_detected():
    drift_info = [1]: [2] for [3] in features

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

A{feature

Bvalues[feature]

Cfeatures

Dfeature

Attempts:

3 left

💡 Hint

Common Mistakes

Missing braces or incorrect iteration variable names.

The dictionary comprehension syntax requires braces and key-value pairs. Here, keys are feature names and values are their current values.