MLSys-NYU-2022
Open-source teaching materials for a practical Machine Learning in Finance course, focusing on industry tools and real-world use cases.
What is MLSys-NYU-2022?
MLSys-NYU-2022 is a collection of open-source educational materials from the Machine Learning in Finance course at NYU Tandon. It provides slides, scripts, and assignments that introduce ML concepts through finance-specific use cases like fraud detection and recommender systems. The course focuses on practical skills using industry tools such as Metaflow and Streamlit to build production-ready ML systems.
Machine learning students, practitioners, and educators seeking a finance-oriented, hands-on curriculum that emphasizes real-world tools and MLOps best practices. It's ideal for those transitioning from theory to applied ML in financial contexts.
Developers choose this for its practical, tool-driven approach that bridges academic learning and industry needs. Unlike purely theoretical resources, it offers structured, reproducible materials with a focus on finance applications and modern MLOps workflows.
Overview
Slides, scripts and materials for the Machine Learning in Finance Course at NYU Tandon, 2022
Use Cases
Best For
- Learning machine learning with finance-specific case studies
- Understanding MLOps tools like Metaflow and Comet in practice
- Building reproducible ML pipelines for educational purposes
- Exploring fraud detection and recommender systems implementations
- Teaching a practical ML course with industry-standard tooling
- Prototyping ML web apps with Streamlit and Flask
Not Ideal For
- Teams needing up-to-date MLOps tools and practices, as the materials are from 2022 and may not reflect current industry standards
- Learners seeking a deep theoretical foundation in machine learning, since the course prioritizes practical application over complex mathematical concepts
- Developers looking for production-ready, optimized code, as the scripts are educational and may require significant modification for deployment
- Self-studiers without access to instructor guidance, given the README's emphasis on class participation and the incomplete standalone experience
Pros & Cons
Pros
Covers real-world ML applications like fraud detection and recommender systems within financial contexts, providing practical case studies not commonly found in generic courses.
Integrates modern tools such as Metaflow, Streamlit, and Comet with hands-on examples and sandbox access, bridging the gap between academic learning and professional workflows.
Organized by week with self-contained folders, READMEs, and reproducible requirements.txt files, making it easy to navigate and replicate the coursework step-by-step.
Emphasizes experiment tracking, pipeline deployment, and evaluation methods using tools like Comet and Flask, preparing users for building reliable ML systems in production.
Cons
Based on the 2022 edition, and the README explicitly states the course changes yearly, so some tools, libraries, or best practices may be outdated compared to current standards.
Lacks live lectures and instructor interaction, which are highlighted as crucial in the disclaimer, reducing its effectiveness for self-learners without supplementary support.
Requires configuration of multiple external platforms like Metaflow and Comet, which can be challenging without the provided sandbox accounts or institutional access.
The modular structure leads to repeated code, such as training loops across weeks, which might not be efficient for learning or code reuse in real projects.
Frequently Asked Questions
Related Projects
AI Expert RoadmapRoadmap to becoming an Artificial Intelligence Expert in 2022
QuantResearchQuantitative analysis, strategies and backtests
Hands-on Train and Deploy MLTrain and Deploy an ML REST API to predict crypto prices, in 10 steps
Advanced-Deep-TradingMostly experiments based on "Advances in financial machine learning" book
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.