Why pipelines ensure reproducibility in ML Python - Test Your Understanding

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create a pipeline that standardizes data and fits a model.

ML Python

from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import LogisticRegression

pipeline = Pipeline(steps=[('scaler', StandardScaler()), ('model', [1])])
pipeline.fit(X_train, y_train)

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

AKMeans()

BStandardScaler()

CRandomForestClassifier()

DLogisticRegression()

Attempts:

3 left

💡 Hint

Common Mistakes

Using StandardScaler() as the model step instead of LogisticRegression()

Forgetting to include a model in the pipeline

The pipeline needs a model step, so LogisticRegression() is the correct choice.

2fill in blank

medium

Complete the code to apply the pipeline to transform test data and predict labels.

ML Python

y_pred = pipeline.[1](X_test)

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

Afit

Bpredict

Cfit_transform

Dtransform

Attempts:

3 left

💡 Hint

Common Mistakes

Using transform instead of predict

Calling fit on test data

To get predictions from the pipeline, use the predict method.

3fill in blank

hard

Fix the error in the pipeline creation by selecting the correct import for the pipeline class.

ML Python

from sklearn.[1] import Pipeline

pipeline = Pipeline(steps=[('scaler', StandardScaler()), ('model', LogisticRegression())])

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

Apipeline

Bpipelines

Cpipeline_module

Dpipeline_class

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'pipelines' instead of 'pipeline' in the import

Trying to import Pipeline from a non-existent module

The correct import is from sklearn.pipeline import Pipeline.

4fill in blank

hard

Fill both blanks to create a pipeline that scales data and fits a decision tree model.

ML Python

from sklearn.pipeline import Pipeline
from sklearn.preprocessing import [1]
from sklearn.tree import [2]

pipeline = Pipeline(steps=[('scaler', StandardScaler()), ('model', DecisionTreeClassifier())])

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

AStandardScaler

BMinMaxScaler

CDecisionTreeClassifier

DRandomForestClassifier

Attempts:

3 left

💡 Hint

Common Mistakes

Mixing scaler and model imports

Using RandomForestClassifier instead of DecisionTreeClassifier

StandardScaler is used for scaling, and DecisionTreeClassifier is the model.

5fill in blank

hard

Fill all three blanks to create a pipeline, fit it, and get the accuracy score.

ML Python

from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler
from sklearn.svm import SVC
from sklearn.metrics import accuracy_score

pipeline = Pipeline(steps=[('scaler', [1]), ('model', [2])])
pipeline.fit(X_train, y_train)
y_pred = pipeline.[3](X_test)
score = accuracy_score(y_test, y_pred)

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

AStandardScaler()

BSVC()

Cpredict

Dfit

Attempts:

3 left

💡 Hint

Common Mistakes

Using fit instead of predict to get predictions

Forgetting parentheses when creating scaler or model instances

StandardScaler() scales data, SVC() is the model, and predict gets predictions.