ColumnTransformer for mixed types in ML Python - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the correct class for transforming columns.

ML Python

from sklearn.compose import [1]

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

AStandardScaler

BOneHotEncoder

CColumnTransformer

DSimpleImputer

Attempts:

3 left

💡 Hint

Common Mistakes

Importing a transformer like StandardScaler instead of ColumnTransformer.

Confusing OneHotEncoder with ColumnTransformer.

The ColumnTransformer class is used to apply different transformations to different columns.

2fill in blank

medium

Complete the code to create a ColumnTransformer that applies OneHotEncoder to categorical columns.

ML Python

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

ct = ColumnTransformer(transformers=[('cat', [1](), ['color', 'type'])], remainder='passthrough')

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

AOneHotEncoder

BStandardScaler

CSimpleImputer

DPCA

Attempts:

3 left

💡 Hint

Common Mistakes

Using StandardScaler for categorical data.

Forgetting to import OneHotEncoder.

OneHotEncoder is used to convert categorical columns into one-hot numeric arrays.

3fill in blank

hard

Fix the error in the code to apply StandardScaler to numeric columns.

ML Python

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

ct = ColumnTransformer(transformers=[('num', [1](), ['age', 'income'])], remainder='passthrough')

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

AOneHotEncoder

BStandardScaler

CSimpleImputer

DLabelEncoder

Attempts:

3 left

💡 Hint

Common Mistakes

Using OneHotEncoder for numeric columns.

Using LabelEncoder which is for labels, not features.

StandardScaler scales numeric features to have mean 0 and variance 1.

4fill in blank

hard

Fill both blanks to create a ColumnTransformer that imputes missing values in numeric columns and encodes categorical columns.

ML Python

from sklearn.compose import ColumnTransformer
from sklearn.impute import [1]
from sklearn.preprocessing import [2]

ct = ColumnTransformer(transformers=[('num', [1](), ['age', 'income']), ('cat', [2](), ['gender', 'city'])], remainder='passthrough')

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

ASimpleImputer

BOneHotEncoder

CStandardScaler

DLabelEncoder

Attempts:

3 left

💡 Hint

Common Mistakes

Using StandardScaler instead of SimpleImputer for missing values.

Using LabelEncoder for categorical features in ColumnTransformer.

SimpleImputer fills missing values, and OneHotEncoder encodes categorical features.

5fill in blank

hard

Fill all three blanks to create a pipeline that preprocesses numeric and categorical data, then fits a logistic regression model.

ML Python

from sklearn.compose import ColumnTransformer
from sklearn.pipeline import Pipeline
from sklearn.linear_model import [1]
from sklearn.impute import [2]
from sklearn.preprocessing import [3], StandardScaler

numeric_transformer = Pipeline(steps=[('imputer', [2]()), ('scaler', StandardScaler())])
categorical_transformer = Pipeline(steps=[('encoder', [3](handle_unknown='ignore'))])

ct = ColumnTransformer(transformers=[('num', numeric_transformer, ['age', 'income']), ('cat', categorical_transformer, ['gender', 'city'])])

model = Pipeline(steps=[('preprocessor', ct), ('classifier', [1]())])

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

ALogisticRegression

BSimpleImputer

COneHotEncoder

DRandomForestClassifier

Attempts:

3 left

💡 Hint

Common Mistakes

Using RandomForestClassifier instead of LogisticRegression as the model here.

Forgetting to use SimpleImputer in the numeric pipeline.

Not setting handle_unknown='ignore' in OneHotEncoder.

This pipeline imputes missing numeric values with SimpleImputer, encodes categorical data with OneHotEncoder, and fits a LogisticRegression model.