Reusable Machine Learning Codebase 🔬

This project has reusable code and components for machine learning tasks. To contribute, raise a PR.

Currently supports python 3.7.6

Key Features

• Built on top of Hydra
• Automatic logging for reroducible-results
• Faster training using JAX

More features coming soon

Roadmap 🚧

01. Read Data

• tabular data

  • pandas supported csv, tsv and other files
  • postgesql

• folders

  • class separated folders for supervised learning
    • images
    • text
  • conversion to coco format

02. Featurisation

• sklearn
  • n-grams
  • tfidf
  • trucated svd

03. Supervised Learning

• sklearn
  • classificaton

04. Unsupervised Learning

Coming soon!