Repository hosting code used to reproduce results in "Actions Speak Louder than Words: Trillion-Parameter Sequential Transducers for Generative Recommendations" (https://arxiv.org/abs/2402.17152).

CUDA Templates for Linear Algebra Subroutines

What would you do with 1000 H100s...

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

AITemplate is a Python framework which renders neural network into high performance CUDA/HIP C++ code. Specialized for FP16 TensorCore (NVIDIA GPU) and MatrixCore (AMD GPU) inference.

RLMeta is a light-weight flexible framework for Distributed Reinforcement Learning Research.

Graph Neural Network Library for PyTorch

Open deep learning compiler stack for cpu, gpu and specialized accelerators

😎 Awesome lists about all kinds of interesting topics

The Vim plugin shortlist

A collection of ZSH frameworks, plugins, themes and tutorials.

Scoreboard for ONNX Backend Compatibility

Cheatsheets for web development - devhints.io

Modularized configuration for a NixOS system

Visualizer for neural network, deep learning and machine learning models

Convert TensorFlow, Keras, Tensorflow.js and Tflite models to ONNX

Convolution visualizations

Tensorflow Backend for ONNX

Tutorials for creating and using ONNX models

ONNX to Core ML Converter

TorchBench is a collection of open source benchmarks used to evaluate PyTorch performance.

A collection of pre-trained, state-of-the-art models in the ONNX format

Caffe2 is a lightweight, modular, and scalable deep learning framework.

Open standard for machine learning interoperability

Tensors and Dynamic neural networks in Python with strong GPU acceleration