Evaluate OutlierHandler Class for Code Quality and Testing
Overview
This question evaluates a candidate's competence in machine-learning data preprocessing, scikit-learn-style transformer API design, unit testing, and code quality assessment including PEP-8 and maintainability concerns.
Capital One
Jul 12, 2025, 6:59 PM
Data Scientist
Onsite
Machine Learning
52
0
Code Review: OutlierHandler and Imputer Classes
Context
You are given a Python module that implements one OutlierHandler class and three Imputer classes for preprocessing tabular data. The classes appear to be intended for use in machine-learning pipelines (e.g., scikit-learn style), but the code has mixed style and testing coverage.
Assumptions to make the question self-contained:
- OutlierHandler detects outliers per feature using a rule such as IQR capping (Q1 − 1.5×IQR, Q3 + 1.5×IQR) or z-score thresholds and caps or replaces outliers during transform.
- Each Imputer learns a statistic on fit (e.g., mean/median/mode/constant) and fills missing values during transform.
Tasks
- Provide a high-level summary of what the OutlierHandler class does.
- Explain why separating fit and transform into two methods is beneficial in this context.
- Identify coding-style and maintainability issues in the file (naming, docstrings, magic numbers, etc.).
- Propose the single most critical unit test you would add for OutlierHandler.
-
For the three imputation classes, describe their overall purpose and identify at least two style problems (e.g., use of
from numpy import *).
Hints: Relate your answers to the scikit-learn transformer API, unit-testing best practices, and PEP-8.
Solution
Login required
Show