michaelsdr

dalle-mini

DALL·E Mini - Generate images from a text prompt

Machine-Learning-Collection

A resource for learning about Machine learning & Deep Learning

torchdiffeq

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

Conference-Acceptance-Rate

Acceptance rates for the major AI conferences

picoGPT

An unnecessarily tiny implementation of GPT-2 in NumPy.

score_sde

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

keops

KErnel OPerationS, on CPUs and GPUs, with autodiff and without memory overflows

jaxopt

Hardware accelerated, batchable and differentiable optimizers in JAX.

deq

[NeurIPS'19] Deep Equilibrium Models

ffjord

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

fast-soft-sort

Fast Differentiable Sorting and Ranking

set_transformer

Pytorch implementation of set transformer

ott

Optimal transport tools implemented with the JAX framework, to get differentiable, parallel and jit-able computations.

PCT_Pytorch

Pytorch implementation of PCT: Point Cloud Transformer

benchopt

Making your benchmark of optimization algorithms simple and open

momentumnet

Drop-in replacement for any ResNet with a significantly reduced memory footprint and better representation capabilities

in-context-learning

ViT-CIFAR

PyTorch implementation for Vision Transformer[Dosovitskiy, A.(ICLR'21)] modified to obtain over 90% accuracy FROM SCRATCH on CIFAR-10 with small number of parameters (= 6.3M, originally ViT-B has 86M).

Fixup

A Re-implementation of Fixed-update Initialization

revlib

Simple and efficient RevNet-Library for PyTorch with XLA and DeepSpeed support and parameter offload

difftopk

Differentiable Top-k Classification Learning

torch_ACA

repo for paper: Adaptive Checkpoint Adjoint (ACA) method for gradient estimation in neural ODE

TorchDiffEqPack

CubicEquationSolver

Python Script to find the roots of a Cubic Equation

sinkformers

Transformers with doubly stochastic attention

jacobian_free_backprop

Implicit networks can be trained efficiently and simply by using Jacobian-free Backprop (JFB).

tan

Computationally friendly hyper-parameter search with DP-SGD

jz-hydra-submitit-launcher

A utility to be able to launch jobs on JZ with reasonable defaults

MoCapsNet

Official implementation of the paper Momentum Capsule Networks (MoCapsNet).

benchmark_resnet_classif

Benchopt benchmark for ResNet fitting on a classification task