Apache Airflow for ML orchestration in MLOps - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the Airflow DAG class.

MLOps

from airflow import [1]

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

AVariable

BTaskInstance

CBaseOperator

DDAG

Attempts:

3 left

💡 Hint

Common Mistakes

Importing TaskInstance instead of DAG

Using BaseOperator which is for tasks, not the DAG itself

The DAG class is the core component to define workflows in Airflow.

2fill in blank

medium

Complete the code to define the default arguments for the DAG.

MLOps

default_args = {
    'owner': 'airflow',
    'start_date': [1]
}

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

Adatetime(2023, 1, 1)

Bdatetime.now()

Cdatetime.today()

Ddatetime.utcnow()

Attempts:

3 left

💡 Hint

Common Mistakes

Using datetime.now() which changes every time and causes errors

Using datetime.today() which is similar to now and not recommended

The start_date should be a fixed datetime object, not a dynamic call like datetime.now().

3fill in blank

hard

Fix the error in the DAG instantiation by completing the missing argument.

MLOps

dag = DAG('ml_pipeline', default_args=default_args, schedule_interval=[1])

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

A'@daily'

Bdaily

CNone()

DNone

Attempts:

3 left

💡 Hint

Common Mistakes

Using bare word daily without quotes

Using None() which is not valid syntax

The schedule_interval expects a string like '@daily' or a cron expression, not a bare word or None() call.

4fill in blank

hard

Fill both blanks to create a PythonOperator task that runs a function named train_model.

MLOps

train_task = PythonOperator(task_id='train', python_callable=[1], dag=[2])

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

Atrain_model

Btrain

Cdag

Ddefault_args

Attempts:

3 left

💡 Hint

Common Mistakes

Using task_id as python_callable

Passing default_args instead of dag object

The python_callable argument takes the function name train_model, and the dag argument links the task to the DAG object.

5fill in blank

hard

Fill both blanks to set task dependencies: preprocess_task runs before train_task, which runs before evaluate_task.

MLOps

preprocess_task[1]train_task[2]evaluate_task

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

A >>

B <<

C &

D +

Attempts:

3 left

💡 Hint

Common Mistakes

Using '<<' which reverses the order

Using '&' or '+' which are not valid for dependencies

The >> operator in Airflow sets the order of tasks, meaning the left task runs before the right task.