Pipeline components and DAGs in MLOps - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to define a pipeline task using a decorator.

MLOps

@pipeline
 def my_pipeline():
     @[1]
     def preprocess():
         pass

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

Astep

Bcomponent

Ctask

Djob

Attempts:

3 left

💡 Hint

Common Mistakes

Using @step or @job which are not standard task decorators here.

The @task decorator defines a pipeline task in many MLOps frameworks.

2fill in blank

medium

Complete the code to create a Directed Acyclic Graph (DAG) for pipeline execution.

MLOps

with DAG('[1]', schedule_interval='@daily') as dag:
    pass

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

Apipeline_dag

Brun_pipeline

Cdag_pipeline

Dmy_pipeline

Attempts:

3 left

💡 Hint

Common Mistakes

Using generic names like 'pipeline_dag' which may not match the pipeline.

The DAG name is often the pipeline name, here my_pipeline.

3fill in blank

hard

Fix the error in the task dependency definition to ensure correct execution order.

MLOps

preprocess = task1()
train = task2()
train [1] preprocess

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 causing errors.

The >> operator sets train to run after preprocess.

4fill in blank

hard

Fill both blanks to define a pipeline component with input and output.

MLOps

def preprocess(data: [1]) -> [2]:
    # process data
    pass

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

ADataset

BModel

COutput

DInput

Attempts:

3 left

💡 Hint

Common Mistakes

Confusing input/output with data types like Dataset or Model.

The input type is Input and the output type is Output in pipeline components.

5fill in blank

hard

Fill all three blanks to create a pipeline step that runs a training component with parameters.

MLOps

train_step = [1](
    model_name=[2],
    epochs=[3]
)

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

Atrain_component

B'resnet50'

C10

Dtrain_pipeline

Attempts:

3 left

💡 Hint

Common Mistakes

Using pipeline name instead of component for first blank.

Putting number as string for epochs.

The training step calls train_component with model name 'resnet50' and epochs 10.