Aaron Lou

tvm. Terminal Velocity Matching

nano-vllm. Nano vLLM

Liger-Kernel. Efficient Triton Kernels for LLM Training

ring-flash-attention. Ring attention implementation with flash attention

open_lm. A repository for research on medium sized language models.

LLaVA. [NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.

DiS. Scalable Diffusion Models with State Space Backbone

mamba. Mamba SSM architecture

U-ViT. A PyTorch implementation of the paper "All are Worth Words: A ViT Backbone for Diffusion Models".

audiocraft. Audiocraft is a library for audio processing and generation with deep learning. It features the state-of-the-art EnCodec audio compressor / tokenizer, along with MusicGen, a simple and controllable music generation LM with textual and melodic conditioning.

Riemannian-Residual-Neural-Networks. [NeurIPS 2023] Riemannian Residual Neural Networks (https://arxiv.org/abs/2006.10254)

DNA-Diffusion. 🧬 Generative modeling of regulatory DNA sequences with diffusion probabilistic models 💨

magvit2-pytorch. Implementation of MagViT2 Tokenizer in Pytorch

consistencydecoder. Consistency Distilled Diff VAE

VQ-Diffusion. Official implementation of VQ-Diffusion

RWKV-LM. RWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RNN and transformer - great performance, linear time, constant space (no kv-cache), fast training, infinite ctx_len, and free sentence embedding.

gpt-2-output-dataset. Dataset of GPT-2 outputs for research in detection, biases, and more

mauve. Package to compute Mauve, a similarity score between neural text and human text. Install with `pip install mauve-text`.

multinomial_diffusion. Python

vdm. Jupyter Notebook

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

openwebtext. Open clone of OpenAI's unreleased WebText dataset scraper. This version uses pushshift.io files instead of the API for speed.

the-pile. Python

flash-attention. Fast and memory-efficient exact attention

gpt-2-Pytorch. Simple Text-Generator with OpenAI gpt-2 Pytorch Implementation

CUDALibrarySamples. CUDA Library Samples

ringattention. Large Context Attention

DiT. Official PyTorch Implementation of "Scalable Diffusion Models with Transformers"

reparam-discrete-diffusion. Reparameterized Discrete Diffusion Models for Text Generation

plaid. Python

score_sde_pytorch. PyTorch implementation for Score-Based Generative Modeling through Stochastic Differential Equations (ICLR 2021, Oral)

l2hmc-qcd. Application of the L2HMC algorithm to simulations in lattice QCD.

cider. PyTorch implementation of CIDER (How to exploit hyperspherical embeddings for out-of-distribution detection), ICLR 2023

riemannian-score-sde. Score-based generative models for compact manifolds

esm. Evolutionary Scale Modeling (esm): Pretrained language models for proteins

DiffDock. Implementation of DiffDock: Diffusion Steps, Twists, and Turns for Molecular Docking

torsional-diffusion. Implementation of Torsional Diffusion for Molecular Conformer Generation (NeurIPS 2022)

se3_diffusion. Implementation for SE(3) diffusion model with application to protein backbone generation

emgithub. Embed a file from GitHub repository just like GitHub Gist.

Stanford-LaTeX-Poster-Template. Stanford LaTeX poster template

ddim. Denoising Diffusion Implicit Models

pytorch-fid. Compute FID scores with PyTorch.

Awesome-Diffusion-Models. A collection of resources and papers on Diffusion Models

llama. Inference code for Llama models

al-folio. A beautiful, simple, clean, and responsive Jekyll theme for academics

ffcv. FFCV: Fast Forward Computer Vision (and other ML workloads!)

Diffusion-LM. Diffusion-LM

diffusion. Denoising Diffusion Probabilistic Models

denoising-diffusion-pytorch. Implementation of Denoising Diffusion Probabilistic Model in Pytorch

PDEBench. PDEBench: An Extensive Benchmark for Scientific Machine Learning

edm. Elucidating the Design Space of Diffusion-Based Generative Models (EDM)

guided-diffusion. Python

Equivariant-Manifold-Flows. [NeurIPS 2021] Equivariant Manifold Flows (https://arxiv.org/abs/2107.08596)

slurm_scripts. Some slurm scripts I find useful.

score_flow. Official code for "Maximum Likelihood Training of Score-Based Diffusion Models", NeurIPS 2021 (spotlight)

mega. Sequence modeling with Mega.

chebpy. A Python implementation of Chebfun

chebfun. Chebfun: numerical computing with functions.

TS2Kit. Differentiable Spherical Harmonic Transforms in PyTorch

zeroth-order-Riemann-Hess-code. code for “On Sharp Stochastic Zeroth Order Hessian Estimatorsover Riemannian Manifolds”

grasp_diffusion. Pytorch implementation of diffusion models on Lie Groups for 6D grasp pose generation https://sites.google.com/view/se3dif/home

score_sde. Official code for Score-Based Generative Modeling through Stochastic Differential Equations (ICLR 2021, Oral)

ncsnv2. The official PyTorch implementation for NCSNv2 (NeurIPS 2020)

ncsn. Noise Conditional Score Networks (NeurIPS 2019, Oral)

pixel-cnn-pp. Pytorch Implementation of OpenAI's PixelCNN++

moser_flow. Python

www_personal. Personal Webpage

s4. Structured state space sequence models

awesome-normalizing-flows. Awesome resources on normalizing flows.

rectangular-flows. Code for the NeurIPS'21 paper "Rectangular Flows for Manifold Learning"

awesome-phd-advice. Collection of advice for prospective and current PhD students

PHDimGeneralization. Official implementation of "Intrinsic Dimension, Persistent Homology and Generalization in Neural Networks", NeurIPS 2021.

align_uniform. Open source code for paper "Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hypersphere" ICML 2020

signatory. Differentiable computations of the signature and logsignature transforms, on both CPU and GPU. (ICLR 2021)

PINNs. Physics Informed Deep Learning: Data-driven Solutions and Discovery of Nonlinear Partial Differential Equations

Awesome-Hyperbolic-NeuralNetworks. Papers and Codes for the deep learning in hyperbolic space

hyperbolic_nn_plusplus. Official PyTorch implementation of Hyperbolic Neural Networks++

orthopy. :triangular_ruler: Orthogonal polynomials in all shapes and sizes.

plottify. A simple matplotlib wrapper that automatically scales font sizes, scatter point sizes, and line widths to be proportional to figure size.

zeus. ⚡️ zeus: Lightning Fast MCMC ⚡️

PointFlow. PointFlow : 3D Point Cloud Generation with Continuous Normalizing Flows

torch-cephes. Cephes Mathematical Functions library wrapped for Torch

OrthNet. TensorFlow, PyTorch and Numpy layers for generating Orthogonal Polynomials

mujoco. Multi-Joint dynamics with Contact. A general purpose physics simulator.

lie-transformer. LieTransformer: Equivariant Self-Attention for Lie Groups

lie-transformer-pytorch. Implementation of Lie Transformer, Equivariant Self-Attention, in Pytorch

rcpm. Riemannian Convex Potential Maps

meshlab. The open source mesh processing system

torchph. The essence of my research, distilled for reusability. Enjoy 🥃!

pointnet.pytorch. pytorch implementation for "PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation" https://arxiv.org/abs/1612.00593

POT. POT : Python Optimal Transport

torchtyping. Type annotations and dynamic checking for a tensor's shape, dtype, names, etc.

Non-Homophily-Benchmarks. [WWW 2021 GLB] New Benchmarks for Learning on Non-Homophilous Graphs

geomstats. Computations and statistics on manifolds with geometric structures.

applications. Examples of more involved applications using Geomstats

awesome-equivariant-network. Paper list for equivariant neural network

Must-read-papers-and-continuous-tracking-on-Graph-Neural-Network-GNN-progress. Papers on Graph neural network(GNN)

CorrectAndSmooth. [ICLR 2021] Combining Label Propagation and Simple Models Out-performs Graph Neural Networks (https://arxiv.org/abs/2010.13993)

csGraduateFellowships. A curated list of fellowships for graduate students in Computer Science and related fields.

PyHessian. PyHessian is a Pytorch library for second-order based analysis and training of Neural Networks

ffjord. code for "FFJORD: Free-form Continuous Dynamics for Scalable Reversible Generative Models".

umap. Uniform Manifold Approximation and Projection

geometric_ml. This repository contains code for applying Riemannian geometry in machine learning.

geotorch. Constrained optimization toolkit for PyTorch

HyperbolicNF. ICML 2020 Paper: Latent Variable Modelling with Hyperbolic Normalizing Flows

rvae. Python

Differentiable-Frechet-Mean. [ICML 2020] Differentiating through the Fréchet Mean (https://arxiv.org/abs/2003.00335).

rs4a. Randomized Smoothing of All Shapes and Sizes (ICML 2020).

torchsde. Differentiable SDE solvers with GPU support and efficient sensitivity analysis.

pytorch-lightning. Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.

ripser. Ripser: efficient computation of Vietoris–Rips persistence barcodes

gudhi-devel. The GUDHI library is a generic open source C++ library, with a Python interface, for Topological Data Analysis (TDA) and Higher Dimensional Geometry Understanding.

TopologyLayer. A Topology Layer for Machine Learning : Persistent Homology + Features for PyTorch

cptq.github.io. HTML

WGAN-TS. Implementation of WGAN-TS

Neural-Manifold-Ordinary-Differential-Equations. [NeurIPS 2020] Neural Manifold Ordinary Differential Equations (https://arxiv.org/abs/2006.10254)

dnn-topology. Computing topological cavities of deep neural networks

blackbox-backprop. Torch modules that wrap blackbox combinatorial solvers according to the method presented in "Differentiating Blackbox Combinatorial Solvers"

relie. Experiments for the AISTATS publication on Reparameterizing Distributions over Lie Groups

manifold-flow. Manifold-learning flows (ℳ-flows)

higher. higher is a pytorch library allowing users to obtain higher order gradients over losses spanning training loops rather than individual training steps.

hgnn. Hyperbolic Graph Neural Networks

mean-field-theory-deep-learning. paper lists and information on mean-field theory of deep learning

expm. Differentiable and numerically stable implementation of the matrix exponential

Grassmann.jl. ⟨Grassmann-Clifford-Hodge⟩ differential geometric algebra

cp-book. Book Code for Competitive Programming

Manopt.jl. 🏔️Optimization on Riemannian Manifolds in Julia

torchdiffeq. Differentiable ODE solvers with full GPU support and O(1)-memory backpropagation.

matrix-sqrt. matrix square root and its gradient

hyrnn. Hyperbolic Neural Networks, pytorch

matrix-manifolds. Source code for the "Computationally Tractable Riemannian Manifolds for Graph Embeddings" paper

ODE-DL. Paper List For Linking ODE and Deep Learning

pymanopt. Python toolbox for optimization on Riemannian manifolds with support for automatic differentiation

hgcn. Hyperbolic Graph Convolutional Networks in PyTorch.

gnn_in_neurips_2019. A comprehensive collection of GNN works in NeurIPS 2019.

HyperbolicTiling_Compression. Official implementation (compression part) for paper: Numerically Accurate Hyperbolic Embeddings Using Tiling-Based Models

geoopt. Riemannian Adaptive Optimization Methods with pytorch optim

knockknock. 🚪✊Knock Knock: Get notified when your training ends with only two additional lines of code

GNNPapers. Must-read papers on graph neural networks (GNN)

graph-neural-networks-roadmap.

ega. amateur translation project of Grothendieck's EGA.

geodeb. Debug and visualize computational geometry problems

manifold_mixup. Code for reproducing Manifold Mixup results (ICML 2019)

Intermediate-Level-Attack. [ICCV 2019] Enhancing Adversarial Example Transferability with an Intermediate Level Attack (https://arxiv.org/abs/1907.10823)

pytorch_geometric. Graph Neural Network Library for PyTorch

TurningWeaknessIntoStrength. Official implementation for paper: A New Defense Against Adversarial Images: Turning a Weakness into a Strength

Dataset-REPAIR. REPresentAtion bIas Removal (REPAIR) of datasets

expRNN. Optimization with orthogonal constraints and on general manifolds

awesome-neural-ode. A collection of resources regarding the interplay between differential equations, deep learning, dynamical systems, control and numerical methods.

liegroups. Python implementation of SO2, SE2, SO3, and SE3 matrix Lie groups using numpy or pytorch

hyperbolics. Hyperbolic Embeddings

runelite. Open source Old School RuneScape client

hyperbolic_nn. Source code for the paper "Hyperbolic Neural Networks", https://arxiv.org/abs/1805.09112

poincare-embeddings. PyTorch implementation of the NIPS-17 paper "Poincaré Embeddings for Learning Hierarchical Representations"

cleverhans. An adversarial example library for constructing attacks, building defenses, and benchmarking both

dopamine. Dopamine is a research framework for fast prototyping of reinforcement learning algorithms.

baselines. OpenAI Baselines: high-quality implementations of reinforcement learning algorithms

openml-python. OpenML's Python API for a World of Data and More 💫

ray. Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

MARL-Papers. Paper list of multi-agent reinforcement learning (MARL)

rlgraph. RLgraph: Modular computation graphs for deep reinforcement learning

darts. Differentiable architecture search for convolutional and recurrent networks

s2cnn. Spherical CNNs

lie_learn. Computations involving Lie groups and harmonic analysis

tensorforce. Tensorforce: a TensorFlow library for applied reinforcement learning

tensorflow. An Open Source Machine Learning Framework for Everyone

TEMPO. Transforming Existing Music to Produce Originals

Kaggle-FA17. Python