There are many high-quality machine learning resources on the internet, but the field is complex and growing so quickly it requires a resource just to navigate those resources. That's what this guide is.

I'm going through many machine learning resources to create the best technical guide to machine learning for anyone wanting to learn. I'll clarify these resources based on the topics they cover, the prerequisites they require, the time investment, and their overall quality to create the best step-by-step guide to machine learning.

I'll keep this updated as machine learning evolves and new resources emerge. I'll also add code tutorials, written guides, and videos where I think it's necessary. A less-technical guide will be coming soon for those wanting to understand machine learning without building ML systems.

This guide is free and will always be free. You can support it (for free) by:

• Following me on X so I can share this info with more people
• Star this repo and share this road map on your socials
• Join my machine learning network: Society's Backend

Things to Know Before You Begin:

• Take your time. Machine learning is a complex field that is growing quickly. Don't feel like you need to know everything and don't rush learning.
• The primary resources are free. I believe knowledge should be free. This guide includes the best free resources (with a section below on paid) because I think you should explore your free options before diving into something that costs. You'll likely find the same knowledge you would have for free and you'll have a better understanding of what paid resources are worth it once you need them.
• We're still figuring AI out. AI has implications industries other than just technology. This makes it even more impactful but also more complicated. While AI research is advancing, understanding the implications of AI is evolving (think ethical, economic, social, etc.). It's no longer enough to just understand the technical details of machine learning, you should also have an understanding of what it means.
• Your greatest learning resource is people. I've always said this and its especially true for AI. Very few people have a good grasp on all AI topics. The best way to learn about AI is from people who know about it and are sharing their knowledge. I've included a list of people to follow and newsletters to subscribe to for continued learning.
• Everyone will need a baseline understanding of AI. Everyone will use machine learning and should understand the tools they use. That's why I'm going to create a less-technical guide to machine learning as well. I've written more in-depth about this here.

You will need a base knowledge of Python. Model training, data preprocessing, and most machine learning libraries use Python. The best starter course for Python (and programming in general): https://cs50.harvard.edu/python/2022/.

Other alternatives are Google's Python Class and Kaggle's Python Course.

Anytime you're working with code you should know how to use version control. Version control is used to manage, save, maintain, and collaborate on code while tracking changes to the code over time. GitHub (this website) is one of the most popular platforms for doing this using a version control technology called Git. It's 100% free and a must for anyone building software.

Learn how to use Git and learn how to use GitHub. You don't need to run through the entirety of these guides as only the basic are necessary to get started; however, running all the way through each is time well spent.

Don't forget to follow me on GitHub and star this repo once your GitHub account is set up.

These are the best technical resources I've found for learning about machine learning. This is a tentative ordering for how to go through them. In January 2024, I'll be going through these courses to add details about topics covered, time investment, the order to go through them, and more.

Machine learning is a quickly evolving field. These are the resources I use to stay updated on developments and products.

X is the best place for any sort of tech news, updates, and learning. I've curated a list of people to follow to learn more about ML for you to use as a starting point. You can follow the list and pin it on your feed. I suggest following all the people within this list as this will curate your 'For You' and 'Following Feeds' and other features such as 'Top Articles'.

I've included people with differing viewpoints on AI because I think taking both into account is the only way to learn. Most things said aren't the absolute truth. Use them to learn and form your own opinion.

Another great resources are newsletters. Here are high-quality newsletters that won't spam your inbox:

• Ahead of AI
• AI Made Simple
• AI Supremacy
• Society's Backend
• The Batch

I'll be adding more newsletter and more detail about each over time.

Beyond my college degree, I haven't paid for any machine learning courses so I can't speak to the quality of any paid resources. These are resources I've heard good things about. Most have a trial period and then cost money. Again, use free resources first.

• Google/Google DeepMind
• OpenAI
• Stability AI
• Midjourney
• Meta
• Scale AI
• Fast AI

An overview of machine learning tools and how they're useful. More on this coming soon.

• HuggingFace
• GitHub
• Python
• PyTorch
• Tensorflow
• Keras
• Google Colab
• Giskard: a testing framework for ML models
• Google AI Studio
• MergeKit
• HuggingFace Tasks

• Model-Based Machine Learning
• Dive into Deep Learning

A list of important machine learning topics. More on this coming soon including definitions and links to learn more.

• Reinforcement Learning
• Deep Learning
• Supervised Learning
• Unsupervised Learning
• Transfer Learning
• Fine-tuning
• Semi-supervised Learning
• Reinforcement Learning w/Human Feedback (RLHF)
• Online learning
• Retrieval-Augmented Generation (RAG)
• Low-Rank Adaptation of Large Language Models (LORA)
• Mixture of Experts (MoE)

This road map will always be free and will always focus on free resources. Support it by:

• Following me on X
• Star this repo and share this roadmap on your socials
• Join my machine learning network: Society's Backend

Last Updated January 2024

In order to comply with X Shopping's terms of service, this road map is also available in paperback. It'll cost as I'll have to send a print out of it to you. You can contact me for this on X.