Fix Feature Engineering Bug

Debuggingeasy~20 min

Objective

Fix three data engineering bugs that cause model performance to degrade in production.

Background

Your model performs well in offline evaluation but fails in production. After investigation, you discover the feature engineering pipeline has three bugs: target leakage in feature construction, incorrect datetime feature extraction, and missing null value handling that causes downstream NaN errors.

Requirements

1.Remove target leakage: the label column must not be used as a feature
2.Fix datetime feature extraction to use day_of_week instead of raw day
3.Add null handling for numeric columns before feature computation

Evaluation (100 points)

No target leakage

The label column must not appear in the feature matrix

30pt

Correct day-of-week extraction

Datetime features should use dayofweek (0-6) not day (1-31)

25pt

Null values handled

handle_nulls should fill or drop NaN values

25pt

Derived features handle division edge cases

price_per_unit should not produce NaN/Inf from null inputs

20pt

Hints

Select a file to start editing

Terminal

AI Assistant50K tokens left

Ask me about the code, bugs, or concepts.
I'll guide you in plain English, no code output.
Budget: 50K tokens per lab