ML Debugging Exercises

A collection of practical machine learning debugging scenarios illustrating common failure modes, incorrect evaluation practices, and corrected implementations.

Purpose

This repository demonstrates machine learning engineering workflows for:

• identifying model failure modes
• diagnosing data leakage
• handling class imbalance
• correcting evaluation mistakes
• comparing broken and fixed pipelines

It is designed as a public demonstration repository using standard datasets and simplified workflows. Proprietary systems and internal datasets are not included.

Exercise Areas

• Data Leakage
• Class Imbalance
• Wrong Evaluation Metric

Repository Structure

data_leakage/
    broken_pipeline.py
    fixed_pipeline.py
    explanation.md

class_imbalance/
    broken_training.py
    fixed_training.py
    explanation.md

wrong_metric/
    broken_evaluation.py
    fixed_evaluation.py
    explanation.md

Quick Start

Install dependencies:

pip install -r requirements.txt

Run an example:

python data_leakage/broken_pipeline.py
python data_leakage/fixed_pipeline.py
python class_imbalance/broken_training.py
python wrong_metric/fixed_evaluation.py

Setup

Install dependencies:

pip install -r requirements.txt

Notes

This repository contains demonstration implementations intended to illustrate practical machine learning debugging patterns using public datasets.