Feature engineering pipelines in MLOps - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create a simple feature pipeline step using scikit-learn.

MLOps

from sklearn.preprocessing import [1]

scaler = [1]()

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

AStandardScaler

BMinMaxScaler

CPCA

DOneHotEncoder

Attempts:

3 left

💡 Hint

Common Mistakes

Using PCA here instead of a scaler.

Choosing OneHotEncoder which is for categorical data, not scaling.

The StandardScaler is a common feature scaling step in pipelines to standardize features.

2fill in blank

medium

Complete the code to combine two feature transformers into a pipeline.

MLOps

from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler
from sklearn.decomposition import PCA

pipeline = Pipeline(steps=[('scaler', StandardScaler()), ('[1]', PCA(n_components=2))])

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

Apca

Btransformer

Creducer

Dencoder

Attempts:

3 left

💡 Hint

Common Mistakes

Using generic names like 'transformer' which are less descriptive.

Using 'encoder' which is unrelated to PCA.

The name 'pca' is a common step name for the PCA transformer in pipelines.

3fill in blank

hard

Fix the error in the pipeline code by completing the missing method call.

MLOps

from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler
from sklearn.decomposition import PCA

pipeline = Pipeline(steps=[('scaler', StandardScaler()), ('pca', PCA(n_components=2))])
X_transformed = pipeline.[1](X_train)

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

Atransform

Bfit

Cfit_transform

Dpredict

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'fit' alone returns the pipeline itself, not transformed data.

Using 'predict' is for models, not transformers.

The fit_transform method fits the pipeline to the data and then transforms it in one step.

4fill in blank

hard

Fill both blanks to create a ColumnTransformer that applies scaling to numeric and encoding to categorical features.

MLOps

from sklearn.compose import ColumnTransformer
from sklearn.preprocessing import StandardScaler, OneHotEncoder

preprocessor = ColumnTransformer(transformers=[('num', [1], numeric_features), ('cat', [2], categorical_features)])

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

AStandardScaler

BOneHotEncoder

CPCA

DMinMaxScaler

Attempts:

3 left

💡 Hint

Common Mistakes

Using PCA for categorical data.

Mixing up scaler and encoder roles.

Numeric features are scaled with StandardScaler, and categorical features are encoded with OneHotEncoder.

5fill in blank

hard

Fill all three blanks to build a pipeline that preprocesses data and fits a logistic regression model.

MLOps

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

pipeline = Pipeline(steps=[('preprocessor', [1]), ('classifier', LogisticRegression([2]=[3]))])

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

Apreprocessor

Bmax_iter

C100

DStandardScaler

Attempts:

3 left

💡 Hint

Common Mistakes

Passing scaler directly instead of the preprocessor pipeline.

Using wrong parameter names for LogisticRegression.

The pipeline uses a preprocessor step, and LogisticRegression is set with max_iter=100 to control iterations.