Resources

We value reproducible and open science.

Software tools

  • Context-Adaptive ML:

    • ContextualizedML : An SKLearn-style toolbox for estimating and analyzing models, distributions, and functions with context-specific parameters
  • FastLM Tools for large-scale, distributed, parallel, sparsity in LMs. Includes:

    • FastCache: Fast Caching for Diffusion Transformer Through Learnable Linear Approximation
    • PiKV: KV Cache Management System for MoE
    • GraphSnapShot: Caching Local Structure for Fast Graph Learning
  • Explainable Boosting Machines (EBM):

    • TalkToEBM - A Natural Language Interface to Explainable Boosting Machines (EBMs)
    • EBM Utils - Utilities for EBM
    • GAM Purification - Tools for purifying additive models with interaction effects into identifiable representations.
  • Academic Utils:

Advice

Compilations

Templates


Internal resources