facebookresearch / dietgpu
GPU implementation of a fast generalized ANS (asymmetric numeral system) entropy encoder and decoder, with extensions for lossless compression of numerical and other data types in HPC/ML applications.
☆309Updated this week
Related projects: ⓘ
- A library to analyze PyTorch traces.☆270Updated 2 weeks ago
- A GPU performance profiling tool for PyTorch models☆493Updated 3 years ago
- A tensor-aware point-to-point communication primitive for machine learning☆247Updated last year
- ☆248Updated this week
- Implementation of a Transformer, but completely in Triton☆242Updated 2 years ago
- An open-source efficient deep learning framework/compiler, written in python.☆646Updated 3 weeks ago
- A library of GPU kernels for sparse matrix operations.☆240Updated 3 years ago
- DiffQ performs differentiable quantization using pseudo quantization noise. It can automatically tune the number of bits used per weight …☆229Updated last year
- PyTorch RFCs (experimental)☆120Updated 3 weeks ago
- Training neural networks in TensorFlow 2.0 with 5x less memory☆127Updated 2 years ago
- torch::deploy (multipy for non-torch uses) is a system that lets you get around the GIL problem by running multiple Python interpreters i…☆173Updated 3 months ago
- Lightweight and Parallel Deep Learning Framework☆261Updated last year
- Fast Block Sparse Matrices for Pytorch☆545Updated 3 years ago
- Block-sparse primitives for PyTorch☆147Updated 3 years ago
- A Fusion Code Generator for NVIDIA GPUs (commonly known as "nvFuser")☆250Updated this week
- A collection of tools for neural compression enthusiasts.☆499Updated this week
- Nod.ai 🦈 version of 👻 . You probably want to start at https://github.com/nod-ai/shark for the product and the upstream IREE repository …☆107Updated this week
- This repository contains the experimental PyTorch native float8 training UX☆210Updated last month
- ☆234Updated last month
- A CPU+GPU Profiling library that provides access to timeline traces and hardware performance counters.☆692Updated last week
- A performant, memory-efficient checkpointing library for PyTorch applications, designed with large, complex distributed workloads in mind…☆145Updated this week
- [Prototype] Tools for the concurrent manipulation of variably sized Tensors.☆252Updated last year
- A Python-level JIT compiler designed to make unmodified PyTorch programs faster.☆995Updated 5 months ago
- Torch Distributed Experimental☆115Updated last month
- Customized matrix multiplication kernels☆53Updated 2 years ago
- jax-triton contains integrations between JAX and OpenAI Triton☆328Updated this week
- Butterfly matrix multiplication in PyTorch☆160Updated 11 months ago
- MONeT framework for reducing memory consumption of DNN training☆172Updated 3 years ago
- Python bindings for NVTX☆66Updated last year
- Pax is a Jax-based machine learning framework for training large scale models. Pax allows for advanced and fully configurable experimenta…☆446Updated this week