DVC (Data Version Control) basics in MLOps - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to initialize a new DVC project in the current directory.

MLOps

dvc [1]

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

Ainit

Bstart

Ccreate

Dsetup

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'dvc start' instead of 'dvc init'.

Trying 'dvc create' which is not a valid command.

The dvc init command initializes a new DVC project by creating necessary files and folders.

2fill in blank

medium

Complete the code to add a data file named 'data.csv' to DVC tracking.

MLOps

dvc [1] data.csv

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

Apush

Btrack

Ccommit

Dadd

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'dvc track' which is not a valid DVC command.

Using 'dvc push' which uploads data but does not track it.

The dvc add command tells DVC to track the specified data file or directory.

3fill in blank

hard

Fix the error in the command to push data to remote storage.

MLOps

dvc [1] origin

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

Apull

Bclone

Cpush

Dfetch

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'dvc pull' which downloads data instead of uploading.

Using 'dvc clone' which is for git repositories.

The dvc push command uploads tracked data files to the remote storage named 'origin'.

4fill in blank

hard

Fill both blanks to create a DVC pipeline stage named 'train' that runs 'python train.py'.

MLOps

dvc run --name [1] -- [2] train.py

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

Atrain

Bpython

Cbash

Dexecute

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'bash' instead of 'python' to run the Python script.

Omitting the stage name or using a wrong name.

The dvc run command creates a pipeline stage. The --name option names the stage, and the command after -- is the script to run.

5fill in blank

hard

Fill all three blanks to create a DVC pipeline stage named 'evaluate' that runs 'python eval.py' and depends on 'model.pkl'.

MLOps

dvc run --name [1] --deps [2] -- [3] eval.py

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

Aevaluate

Bmodel.pkl

Cpython

Dtrain

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'train' as the stage name instead of 'evaluate'.

Using 'bash' or other interpreters instead of 'python'.

This command creates a pipeline stage named 'evaluate' that depends on the file 'model.pkl' and runs the script 'eval.py' using Python.