A platform independent standalone library that plays Lottie Animation.

lottie renderer for imgui based on rlottie library

A personal experimental C++ Syntax 2 -> Syntax 1 compiler

Bonsai GPU tree code

Combustion engine simulator that generates realistic audio.

Graphic notes on Gilbert Strang's "Linear Algebra for Everyone"

A latent text-to-image diffusion model

An unofficial Unity port of the MobileNeRF viewer

God's actual programming language.

Python code to parse Microsoft PDB files

FauxPilot - an open-source alternative to GitHub Copilot server

Simple C and Halide implementations of the BLAS3 function SGEMM (real single precision general matrix_matrix multiplication).

This is a failed attempt to port Recfilter to latest Halide from 2020. Maybe someone else can figure this out?

A very basic photo editor as a side project begun from my interest in image processing. Utilizes Halide and OpenGL as the main frameworks.

fast, high peformance image processing library for golang

UofT CSC2222 project: using halide to acceleration deep learning neural network

This is an implementation of a BLAS (Level 1) routine called "ISAMAX" which finds the index of the first element having maximum absolute value, in both Halide and C.

This is an implementation of a BLAS (Basic Linear Algebra Subprograms) routine from the second level which is the general matrix-vector multiplication "xGEMV",

An image processing toolkit

Thrust, CUB, TBB, AVX2, AVX-512, CUDA, OpenCL, OpenMP, SyCL - all it takes to sum a lot of numbers fast!

This is an implementation of a BLAS2 (Basic Linear Algebra Subprograms) routine called **xGEMV** or **General Matrix-Vector Multiplication**, in both C and Halide.

This is an implementation of a BLAS1 (Basic Linear Algebra Subprograms) routine called **xSCAL** which scales a vector by a constant, in both C and Halide.

PatchMatch & Poisson Image Editing Implemented in C++, OpenMP, CUDA & Halide

Optimized Parallel RSA Image Encryption using Halide and Cuda.

High throughput rsa encryption/decryption using Halide scheduling.

A 4x8 CGRA (Coarse Grained Reconfigurable Array) used to accelerate applications written in Halide

A very simple standalone demo of how to use Halide to compute a spatially varying convolution