Pinned Repositories
attention-gym
Helpful tools and examples for working with flex-attention
botorch
Bayesian optimization in PyTorch
captum
Model interpretability and understanding for PyTorch
data
A PyTorch repo for data loading and utilities to be shared by the PyTorch domain libraries.
gpt-fast
Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.
opacus
Training PyTorch models with differential privacy
segment-anything-fast
A batched offline inference oriented version of segment-anything
tnt
A lightweight library for PyTorch training tools and utilities
torchrec
Pytorch domain library for recommendation systems
torchtune
PyTorch native post-training library
Meta PyTorch's Repositories
meta-pytorch/torcheval
A library that contains a rich collection of performant PyTorch model metrics, a simple interface to create new metrics, a toolkit to facilitate metric computation in distributed training and tools for PyTorch model evaluations.
meta-pytorch/tlparse
TORCH_LOGS parser for PT2
meta-pytorch/torchsnapshot
A performant, memory-efficient checkpointing library for PyTorch applications, designed with large, complex distributed workloads in mind.
meta-pytorch/test-infra
meta-pytorch/AccelKernels
The accel_kernels repository is under pytorch-labs, which houses experimental GPU and accelerator kernels implemented in Triton or other DSLs, for SoTA algorithms.
meta-pytorch/multipy
torch::deploy (multipy for non-torch uses) is a system that lets you get around the GIL problem by running multiple Python interpreters in a single C++ process.
meta-pytorch/FACTO
Framework for Algorithmic Correctness Testing of Operators
meta-pytorch/torchfix
TorchFix - a linter for PyTorch-using code with autofix support
meta-pytorch/segment-anything-fast
A batched offline inference oriented version of segment-anything
meta-pytorch/gpt-fast
Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.
meta-pytorch/applied-ai
Applied AI experiments and examples for PyTorch
meta-pytorch/LeanRL
LeanRL is a fork of CleanRL, where selected PyTorch scripts optimized for performance using compile and cudagraphs.
meta-pytorch/triton-cpu
An experimental CPU backend for Triton (https//github.com/openai/triton)
meta-pytorch/popcorn-kernels
For building the world's largest dataset of GPU kernels.
meta-pytorch/tosa_serialization_lib-mirror
This is a mirror of https//git.mlplatform.org/tosa/serialization_lib.git/
meta-pytorch/ethos-u-core-driver-mirror
This is a mirror of https//git.mlplatform.org/ml/ethos-u/ethos-u-core-driver.git/
meta-pytorch/superblock
A block oriented training approach for inference time optimization.
meta-pytorch/float8_experimental
This repository contains the experimental PyTorch native float8 training UX
meta-pytorch/workshops
This is a repository for all workshop related materials.
meta-pytorch/hydra-torch
Configuration classes enabling type-safe PyTorch configuration for Hydra apps