Practice Questions — Feature Engineering and Feature Selection

import pandas as pd
from sklearn.datasets import load_iris

iris = load_iris()
df = pd.DataFrame(iris.data, columns=iris.feature_names)
corr_matrix = df.corr()

print("Feature pairs with |correlation| > 0.8:")
for i in range(len(corr_matrix.columns)):
    for j in range(i+1, len(corr_matrix.columns)):
        if abs(corr_matrix.iloc[i, j]) > 0.8:
            print(f"  {corr_matrix.columns[i]} <-> {corr_matrix.columns[j]}: {corr_matrix.iloc[i, j]:.4f}")

from sklearn.feature_selection import SelectKBest, mutual_info_regression
from sklearn.datasets import fetch_california_housing
import numpy as np

housing = fetch_california_housing()
X, y = housing.data, housing.target

selector = SelectKBest(mutual_info_regression, k=4)
selector.fit(X, y)

scores = selector.scores_
selected_mask = selector.get_support()
selected_features = [housing.feature_names[i] for i in range(len(selected_mask)) if selected_mask[i]]

print("MI scores per feature:")
for name, score in sorted(zip(housing.feature_names, scores), key=lambda x: -x[1]):
    marker = " <-- selected" if name in selected_features else ""
    print(f"  {name}: {score:.4f}{marker}")
print(f"\nSelected: {selected_features}")

from sklearn.feature_selection import RFECV
from sklearn.ensemble import RandomForestClassifier
from sklearn.datasets import load_wine
from sklearn.model_selection import StratifiedKFold

wine = load_wine()
X, y = wine.data, wine.target

rfecv = RFECV(
    estimator=RandomForestClassifier(n_estimators=100, random_state=42),
    step=1,
    cv=StratifiedKFold(5),
    scoring='accuracy',
    min_features_to_select=1
)
rfecv.fit(X, y)

print(f"Optimal features: {rfecv.n_features_}")
print(f"Selected features:")
for name, selected in zip(wine.feature_names, rfecv.support_):
    if selected:
        print(f"  {name}")
print(f"\nBest CV accuracy: {rfecv.cv_results_['mean_test_score'].max():.4f}")

import pandas as pd
import numpy as np

def extract_datetime_features(df, col):
    df = df.copy()
    dt = pd.to_datetime(df[col])
    df['hour'] = dt.dt.hour
    df['day_of_week'] = dt.dt.dayofweek
    df['month'] = dt.dt.month
    df['is_weekend'] = (dt.dt.dayofweek >= 5).astype(int)
    df['quarter'] = dt.dt.quarter
    df['hour_sin'] = np.sin(2 * np.pi * dt.dt.hour / 24).round(4)
    df['hour_cos'] = np.cos(2 * np.pi * dt.dt.hour / 24).round(4)
    return df

df = pd.DataFrame({'timestamp': ['2026-01-15 09:30', '2026-06-20 22:00']})
result = extract_datetime_features(df, 'timestamp')
print(result.drop('timestamp', axis=1).to_string())

import pandas as pd
import numpy as np

df = pd.DataFrame({
    'length': [30, 50, 20, 80],
    'width': [20, 30, 15, 40],
    'height': [15, 25, 10, 35]
})

df['volume'] = df['length'] * df['width'] * df['height']
df['surface_area'] = 2 * (df['length']*df['width'] + df['width']*df['height'] + df['length']*df['height'])
df['aspect_ratio_lw'] = df['length'] / df['width']
df['max_dimension'] = df[['length', 'width', 'height']].max(axis=1)
df['is_cubic'] = ((df['length'] == df['width']) & (df['width'] == df['height'])).astype(int)

print(df[['volume', 'surface_area', 'aspect_ratio_lw', 'max_dimension', 'is_cubic']].to_string())

import pandas as pd

def add_text_features(df):
    df = df.copy()
    df['word_count'] = df['text'].str.split().str.len()
    df['avg_word_length'] = df['text'].apply(
        lambda x: sum(len(w) for w in x.split()) / len(x.split())
    ).round(2)
    df['has_exclamation'] = df['text'].str.contains('!').astype(int)
    df['uppercase_ratio'] = df['text'].apply(
        lambda x: sum(1 for c in x if c.isupper()) / len(x)
    ).round(4)
    return df

# Test
df = pd.DataFrame({'text': ['Hello World!', 'this is a test', 'URGENT: Read NOW']})
result = add_text_features(df)
print(result[['word_count', 'avg_word_length', 'has_exclamation', 'uppercase_ratio']].to_string())