Machine Learning Documentation & Guides

By Brannon Dorsey

This is a draft release. This documentation is under active development, so expect imperfections and upcoming content changes. If you notice something, please create an issue on GitHub, or better yet, submit a pull request with a fix.

This website is a reference for all things machine learning at Branger_Briz. Since November 2016, our research team has been exploring the current state-of-the art in machine learning and this website serves as the product of much of that research. The goal of this resource is to provide a central repository for developers, engineers, and programmers of all types to learn about machine learning so that it can be used in current and future projects.

• What is Machine Learning (start here)
• General Purpose Algorithms
• Data is Key
• Machine Learning Models
• The Machine Learning Pipeline
• Types of Learning
• Performance Measures
• Features & Design Matrices

Guides & Examples

A miscellaneous collection of guides, code snippets, and examples on a range of practical ML topics.

• Linear Regression
• Neural Networks and Deep Learning
• Normalization and Preprocessing
• Regularization
• Probability Distributions
• Setup an ML GPU development environment
• Tensorflow.js

Twitterbot Guide

An in-depth tutorial where we build an ML pipeline for generating your own Twitter bots that sound like real Twitter users. We'll show you how to use character embeddings, recurrent neural networks, hyperparameter search, and transfer learning, using Keras and Tensorflow.js. We recommend starting at Part 1.

• Part 0: Model Architecture
• Part 1: Twitter Data Preparation
• Part 2: Model Training and Iteration
• Part 3: Model Inference and Deployment
• Part 4: Transfer Learning, Fine-tuning, and user personalization

Projects

At Branger_Briz, we've made used machine learning in several projects, both research and commercial.

• Muse Music Videos
• Miami Children's Museum Masterpiece Me and Interactive Mural installations
• Sportsmanias: a sports news aggregation platform
• Midi-rnn: Using machine learning to create new melodies

Other Resources

In addition to these resources, I recommend checking out Google's awesome Machine Learning Crash Course, a free self-study course on the basics of ML. They've also compiled an ML Glossary of terms that will be helpful to you as you try and make sense of data science jargon.

Machine Learning Mastery's blog also has hundreds of helpful short blog posts and code snippets on tons of topics in Machine Learning. I've found myself using several of those guides to learn about various topics in ML.

Return to the main page.

All source code in this document is licensed under the GPL v3 or any later version. All non-source code text is licensed under a CC-BY-SA 4.0 international license. You are free to copy, remix, build upon, and distribute this work in any format for any purpose under those terms. A copy of this website is available on GitHub.