Weights and Biases overview in MLOps - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to initialize a Weights and Biases run.

MLOps

import wandb
wandb.[1](project="my-project")

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

Ainit

Blaunch

Crun

Dstart

Attempts:

3 left

💡 Hint

Common Mistakes

Using wandb.start() instead of wandb.init()

Using wandb.run() which does not exist

The correct function to initialize a W&B run is wandb.init(). It sets up the tracking for your experiment.

2fill in blank

medium

Complete the code to log a metric value to Weights and Biases.

MLOps

wandb.log({"accuracy": [1])

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

Aaccuracy_value

B0.95

C"accuracy"

Dlog_accuracy

Attempts:

3 left

💡 Hint

Common Mistakes

Passing the metric name as a string instead of the value

Passing a variable name that is undefined

You log a metric by passing a dictionary with the metric name and its numeric value. Here, 0.95 is a valid accuracy value.

3fill in blank

hard

Fix the error in the code to correctly finish a Weights and Biases run.

MLOps

wandb.[1]()

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

Aclose

Bstop

Cend

Dfinish

Attempts:

3 left

💡 Hint

Common Mistakes

Using wandb.stop() which does not exist

Using wandb.close() which is incorrect

The correct method to finish a W&B run is wandb.finish(). It ends the tracking session properly.

4fill in blank

hard

Fill both blanks to create a W&B config dictionary with learning rate and epochs.

MLOps

config = {"learning_rate": [1], "epochs": [2]

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

A0.01

B10

C100

D0.1

Attempts:

3 left

💡 Hint

Common Mistakes

Using a large number for learning rate

Using a decimal for epochs

The learning rate is usually a small decimal like 0.01, and epochs is a positive integer like 10.

5fill in blank

hard

Fill all three blanks to log a model artifact with name, type, and description.

MLOps

artifact = wandb.Artifact(name=[1], type=[2], description=[3])

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

A"model_v1"

B"model"

C"First version of the model"

D"dataset"

Attempts:

3 left

💡 Hint

Common Mistakes

Using wrong type like 'dataset' for a model

Not using quotes around strings

The artifact name is a string like "model_v1", the type is "model", and the description is a string explaining the artifact.