Roadmap

Free

A roadmap connecting many of the most important concepts in machine learning, how to learn them, and what tools to use to perform them.

FreeFree tier

Type

Open Source

Visit Website GitHub (40 stars)

About Roadmap

A comprehensive, open-source roadmap created by Daniel Bourke that connects many of the most important concepts in machine learning. It covers five key areas: Machine Learning Problems (identifying ML problems), Machine Learning Process (steps to solve them), Machine Learning Tools (what to use), Machine Learning Mathematics (underlying theory), and Machine Learning Resources (how to learn). Originally designed in 2020, it remains roughly 90% valid for 2023. The roadmap is available as a static image, an interactive online version, and a feature-length video walkthrough. Licensed under MIT and inspired by Daniel Formoso's machine learning mindmaps.

Key Features

Covers five core areas: ML problems, process, tools, mathematics, and resources

Interactive version available for browsing

Feature-length video walkthrough included

Curated list of learning resources and tool recommendations

Inspired by and expands upon Daniel Formoso's ML mindmaps

Open source with MIT license

Pros & Cons

Pros

Comprehensive, well-organized overview of the ML landscape
Free and open source (MIT license)
Includes multiple formats: static image, interactive, and video
Remains largely valid for 2023 despite being created in 2020
Curated resources help avoid information overload

Cons

May not cover the very latest developments (e.g., 2024+ advancements)
Static roadmap; requires additional research for deep dives
No active development or releases since initial creation
Focus is on broad concepts rather than step-by-step tutorials

Best For

Learning machine learning from scratch or as a structured refresherSelecting appropriate tools and frameworks for ML projectsUnderstanding the ML problem-solving pipelineTeaching or mentoring others in machine learning fundamentals

FAQ

Is this roadmap still relevant in 2023?

According to the author, the roadmap is 'still 90% valid for 2023' due to the foundational nature of the concepts covered.

What formats is the roadmap available in?

The roadmap is available as a full PNG image, a Keynote file, an overview PNG, an interactive web version, and a feature-length video walkthrough.

Who created this roadmap?

The roadmap was created by Daniel Bourke (mrdbourke) and was inspired by Daniel Formoso's machine learning mindmaps.

What license is the roadmap released under?

The roadmap is released under the MIT license, allowing free use, modification, and distribution.