jbegaint

tuning_playbook

A playbook for systematically maximizing the performance of deep learning models.

llama2.c

Inference Llama 2 in one file of pure C

mlx

MLX: An array framework for Apple silicon

rye

a Hassle-Free Python Experience

mistral-src

Reference implementation of Mistral AI 7B v0.1 model.

dinov2

PyTorch code and models for the DINOv2 self-supervised learning method.

ml-ferret

pip-tools

A set of tools to keep your pinned Python dependencies fresh.

mlx-examples

Examples in the MLX framework

ijepa

Official codebase for I-JEPA, the Image-based Joint-Embedding Predictive Architecture. First outlined in the CVPR paper, "Self-supervised learning from images with a joint-embedding predictive architecture."

python-diskcache

Python disk-backed cache (Django-compatible). Faster than Redis and Memcached. Pure-Python.

ml-fastvit

This repository contains the official implementation of the research paper, "FastViT: A Fast Hybrid Vision Transformer using Structural Reparameterization" ICCV 2023

cramming

Cramming the training of a (BERT-type) language model into limited compute.

setuptools_scm

the blessed package to manage your versions by scm tags

hiera

Hiera: A fast, powerful, and simple hierarchical vision transformer.

tensordict

TensorDict is a pytorch dedicated tensor container.

StyleDrop-PyTorch

Unoffical implement for [StyleDrop](https://arxiv.org/abs/2306.00983)

fastdvdnet

FastDVDnet: A Very Fast Deep Video Denoising algorithm

mlx-data

Efficient framework-agnostic data loading

dropout

Code release for "Dropout Reduces Underfitting"

unearth

A utility to fetch and download python packages

LIC_TCM

nccl-fastsocket

NCCL Fast Socket is a transport layer plugin to improve NCCL collective communication performance on Google Cloud.

constriction

Entropy coders for research and production in Python and Rust.

torchfix

TorchFix - a linter for PyTorch-using code with autofix support

Linear-Autoregressive-Similarity-Index

Code for "The Unreasonable Effectiveness of Linear Prediction as a Perceptual Metric"

CompressAI-Trainer

Training platform for End-to-End compression models, losses and metrics defined in Compressai

2023_ISMRM_Efficient_Operator_Educational

Materials for the "Efficient Forward Models for Image Reconstruction" educational talk at the 2023 ISMRM Annual Meeting.

Lookahead

This project compares Geoffrey Hinton's new Lookahead optimizer with other conventional optimizers like AdamW, SGD, RMSProp, etc. This is a work in progress

compression-server