Causal depthwise conv1d in CUDA, with a PyTorch interface

C++ extensions in PyTorch

A nearly complete collection of prefix sum algorithms implemented in CUDA, D3D12, Unity and WGPU. Theoretically portable to all wave/warp/subgroup sizes.

The AdEMAMix Optimizer: Better, Faster, Older.

STOI loss function in PyTorch

a python script for geocoding boulder and roped problems

PyTorch implementation of the Perceptual Evaluation of Speech Quality for wideband audio

Fast and memory-efficient exact attention

Mamba SSM architecture

Awesome Papers related to Mamba.

Code for exploring Based models from "Simple linear attention language models balance the recall-throughput tradeoff"

A lightweight spatio-temporal network for online eye tracking with event camera

Pythonic and efficient simulation of spin systems

Simple, minimal implementation of the Mamba SSM in one pytorch file. Using logcumsumexp (Heisen sequence).

An algorithm for generating hard RBM instances

Code implementing "Efficient Parallelization of a Ubiquitious Sequential Computation" (Heinsen, 2023)

tutorial for writing custom pytorch cpp+cuda kernel, applied on volume rendering (NeRF)

Fast Discounted Cumulative Sums in PyTorch

Development repository for the Triton language and compiler

A chatbot for recruiters to query information about a job applicant