All Algorithms implemented in Python

scikit-learn: machine learning in Python

The simplest, fastest repository for training/finetuning medium-sized GPTs.

The best ChatGPT that $100 can buy.

Machine Learning Engineering Open Book

Official PyTorch Implementation of "Scalable Diffusion Models with Transformers"

Official repo for consistency models.

A modular framework for vision & language multimodal research from Facebook AI Research (FAIR)

Deep Reinforcement Learning for Keras.

Image Restoration Toolbox (PyTorch). Training and testing codes for DPIR, USRNet, DnCNN, FFDNet, SRMD, DPSR, BSRGAN, SwinIR

Simple, minimal implementation of the Mamba SSM in one file of PyTorch.

Automate Scrubbing your Facebook Presence

Schedule-Free Optimization in PyTorch

VideoSys: An easy and efficient system for video generation

A minimal implementation of DeepMind's Genie world model

Cosmos-Predict2.5, the latest version of the Cosmos World Foundation Models (WFMs) family, specialized for simulating and predicting the future state of the world in the form of video.

Fast & Simple repository for pre-training and fine-tuning T5-style models

Cosmos-Transfer2.5, built on top of Cosmos-Predict2.5, produces high-quality world simulations conditioned on multiple spatial control inputs.

A simple and flexible code for Reservoir Computing architectures like Echo State Networks

Post-training scripts and samples for NVIDIA Cosmos ecosystem

Efficient optimizers

Code for coreset selection methods

Code for paper "Multiple Physics Pretraining for Physical Surrogate Models

Large-scale, multi-GPU capable, kernel solver

3D Object Reconstruction project is a workflow that takes a set of stereo images and camera info and outputs a textured mesh (i.e., .OBJ file). The purpose is to translate physical items into the d…

Implementation of the Sliced Wasserstein Autoencoder using PyTorch

Study on the applicability of Direct Feedback Alignment to neural view synthesis, recommender systems, geometric learning, and natural language processing.

Revisiting Efficient Training Algorithms For Transformer-based Language Models (NeurIPS 2023)

codebase for iccv 2025 paper "One Trajectory, One Token: Grounded Video Tokenization via Panoptic Sub-object Trajectory"