[P] Open-source ML homeworks with auto-tests - fundamental algorithms from first principles

2026-03-21 · Source: Machine Learning · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering · Depth: Novice, quick

Summary

A set of open-source machine learning and deep learning homework assignments, designed for a course at Skoltech, is now publicly available on GitHub. These assignments focus on building fundamental algorithms from first principles, guiding students through step-by-step problems rather than presenting a blank page. A key feature is the integration of automated test-based grading, which provides immediate feedback to students and reduces the grading load on teaching staff. The repository includes notebooks with tasks, helper scripts, auto-tests, grading scripts to prevent accidental misuse of files, and pre-generated data for tests. The content is released under a permissive license, encouraging reuse and adaptation.

Key takeaway

For educators designing machine learning courses, consider adopting an auto-tested, first-principles approach to homework. This method can significantly improve student comprehension of core algorithms and streamline your grading process, allowing students to learn from their mistakes efficiently and independently.

Key insights

Building fundamental ML algorithms from scratch with automated testing enhances deep understanding and provides immediate feedback.

Principles

• Deep understanding requires building from scratch
• Automated testing provides immediate feedback
• Step-by-step guidance prevents learning stalls

Method

Design ML homeworks with starter templates and test suites, guiding students through problems and using automated tests for immediate feedback and grading.

In practice

• Use auto-tests for immediate student feedback
• Provide grading scripts for transparency
• Publish under a permissive license

Topics

• Machine Learning Education
• Fundamental Algorithms
• Automated Grading
• Open-source ML
• Educational Resources

Code references

• fxlrnrpt/sktech_ml_homeworks_2026

Best for: AI Student, Machine Learning Engineer, AI Engineer

Related on AIssential

• 5 Ways to Use Cross-Validation to Improve Time Series Models — Machine Learning Mastery provides comprehensive resources across core machine learning concepts and practical applications.
• Public Machine Learning Solver Framework for Novices in the Machine Learning Domain — A new public framework guides ML novices through problem-solving by recommending complete, expert-informed pipelines.
• You Already Know Machine Learning and Just Call It Something Else — Machine learning models are functions whose parameters are automatically tuned, making ML concepts accessible to software engineers.
• Single AI classes colleges are offering worth it? — Many university-branded single AI courses are overpriced prompt engineering tutorials; seek those teaching underlying ML concepts or practical applications.
• Prodigy: A new tool for radically efficient machine teaching — Prodigy is a tool designed to make machine teaching and data annotation radically more efficient, addressing a key bottleneck in AI development.
• Introduction to Machine Learning — Machine Learning is algorithms improving performance at a task through experience.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Machine Learning.