FastML Documentation

Introduction

Welcome to the FastML documentation. FastML is an MLOps boilerplate designed for rapid development and deployment of machine learning projects. It offers a comprehensive suite of scripts and modules for data handling, preprocessing, modeling, and deployment, streamlining the journey from idea to production.

About This Documentation

This documentation provides a detailed overview of each module and function within FastML. It is organized to help users quickly understand how to utilize the various components of the library to build and deploy machine learning models efficiently.

What You Will See

In this documentation, you will find:

1. Configuration Files: Explanation of configuration settings that control the behavior of different modules.
2. Data Handling: Instructions and examples on how to use connectors for various data sources such as CSV, SQL, MongoDB, and more.
3. Data Preprocessing: Methods for preparing data across different formats including tabular data, text, images, time series, and graphs.
4. Modeling: Detailed guides on setting up and training models for various tasks such as regression, classification, clustering, NLP, and using OpenAI models.
5. Deployment: Steps to containerize models, manage model versions, and serve models for production use.
6. CI/CD: Integration with CI/CD pipelines using GitHub Actions to automate the deployment process.
7. Monitoring with Weights & Biases (WandB): Integration with WandB for experiment tracking, model monitoring, and logging.

Each section provides code examples and markdown documentation to make it easy for users to understand and implement the functionality in their own projects.

We hope this documentation helps you make the most out of FastML and accelerates your machine learning development process.