Skip to content

Latest commit

 

History

History
135 lines (107 loc) · 4.6 KB

README.md

File metadata and controls

135 lines (107 loc) · 4.6 KB

PyMo

A framework in Mojo for AI/ML/DL applications and other domains.

An extended framework for Mojo:

  • The Mojo🔥programming language
  • Mojo is built by Modular Inc. and the Mojo community.

Note

  • Work in Progress - Proof of Concept
  • Currently running on Mojo v25.1.0, Magic v0.7.0

Overview

PyMo is a framework designed to leverage the capabilities of the Mojo programming language. The goal is to provide a seamless transition for Python developers into the realm of AI, Machine Learning, Deep Learning (AI/ML/DL), and other domains - with the performance enhancements offered by Mojo. Since Mojo is designed as a systems programming language, its capabilities go beyond AI/ML/DL. This extends to PyMo as well - it can be leveraged in other programming languages and domains. See docs for more information.

Project Structure

├── init.🔥
├── data/
├── docs/
├── examples/
├── magic.lock
├── mojoproject.toml
├── pymo/
├── test/
└── test_pymo.mojo

Installation & Setup

To install and set up PyMo (on Linux), follow these steps:

Install Magic Build System & Package Manager

  • See Magic documentation for installation instructions: Magic docs.

Clone the repository

git clone git@github.com:rcghpge/pymo.git
cd pymo

Initialize PyMo environment with Magic

  • Add and install packages and librairies (optional) 
    # Initialize PyMo
magic shell

# Install and add packages and libraries
magic add xgboost

Run Test(s)

# Workflow/Development environment
magic run test
mojo test

# Initialize debugging session
mojo test --debug

# Initialize REPL environment
mojo repl

Initialize a Jupyter Notebook Environment (Optional)

magic run jupyter notebook

Format Project's Code (Optional)

magic run format

Build & Ship Package(s) and Module(s) in Mojo (Optional)

magic run build

Features

Interoperability with Python:

PyMo showcases Mojo's potential to become a superset of Python, focusing on enhancing Python's AI/ML/DL capabilities while providing a glimpse into Mojo's language and domain-agnostic potential.

Integration with Popular Libraries:

Here's how PyMo interacts with some well-known Python libraries and frameworks:

  • scikit-learn: For machine learning models.
  • pandas: For data manipulation and analysis.
  • numpy: For numerical computing.
  • seaborn: For statistical data visualization.
  • matplotlib: For plotting visualizations.
  • and more.

Currently proofing, testing, and benchmarking frameworks and libraries.

Getting Started

Check out examples/ and test/ folders for sample code demonstrating how to use PyMo with various AI/ML/DL tasks and initial framework draft of PyMo.

MAX

  • check out MAX - an AI inference platform built by Modular
  • PyMo can be leveraged in this framework.

Python

  • If you are new to programming, check out PyPi - The Python Package Index. A repository of software for the Python programming language and also see Python documentation. Python is an awesome programming language to get started with in programming.

Machine Learning University

  • MLU-Explain - Visual explanations of core machine learning concepts

Contributing

Contributions are welcome! Here's how you can contribute:

  • If you are interested to build out this project. Feel free to contact me. See contact below.
  • Join the discussion. See repository's Discussions board.
  • Report bugs or request features by opening an issue.
  • Fix bugs or implement features by opening a pull request.
  • Please ensure your code adheres to the project's and Mojo's coding standards before submission.

License

The PyMo project is primarily licensed under the MIT License.

Additionally, PyMo also incorporates components, each governed by its respective license:

  • LLVM: Components utilized from LLVM are licensed under the Apache License v2.0 with LLVM Exceptions. See the LLVM License.

  • MAX and Mojo: Usage and distribution are licensed under the MAX & Mojo Community License.

Acknowledgements

  • The Mojo and Python community.
  • The open source community.

Contact

For any inquiries or further information, you can find my email in contacts here.