Highly optimized search for similar multisets

Python implementation of Ising model in 2D and 3D

A python library for fast and optimized VLSI Computer-Aided-Design algorithms

Fundamentals of Accelerated Computing with CUDA Python

Convert Python class to Numba aware class, using StructRef.

Course Project - Advanced Methods in Satellite Image Processing - Spring Semester 2023 - Indian Institute of Technology Bombay

Materials for ARC's cluster club session on accelerating scientific python codes

Advanced statistical analysis suite for ultra-high-energy cosmic rays. Uses optimized Monte Carlo methods to detect clustering around extragalactic sources. Focus: NGC 5128 (Centaurus A) and AGN correlation searches.

Benchmark testing the difference in computation performance with and without Numba JIT in python and with C++

An optimized tool in Python for simulating eutrophication, nutrient cycle and transport and phytoplankton dynamics, which is compatible with parameter optimization tools.

A high-performance Quantum QCAD engine that uses Numba-JIT acceleration to map the "Sweet Spot" where algorithmic complexity meets physical noise. Unlike bloated enterprise kits, it provides a transparent, low-level look at how T_1 and T_2 decay shift the optimal depth of NISQ algorithms.

Seam Carving for Content Aware Image Resizing Project - Computer Vision Lab

A comprehensive collection of examples demonstrating every major feature of Numba, from JIT compilation to CUDA programming and performance optimization.

Autonomous, multi-threaded options trading engine. Bypasses the Python GIL via the Actor Model for lock-free data ingestion and low-latency execution on the Zerodha Kite API.

This repository contains Python code for numerical simulation, data analysis, and figures associated with the Herho et al. (2024) manuscript.

Hardware-accelerated, low-overhead middleware bridge linking Genetec Security Center live video streams with the Edwards FireWorks incident platform. Employs multi-core parallel CPU routines (Numba NJIT) and NVIDIA GPU kernels (CUDA) to process unmanaged memory pixel matrices for real-time visual fire and smoke detection, backed by thread-safe, val

High-performance, multi-GPU computer vision pipeline for Genetec streams using NVIDIA DeepStream, Triton, and Numba JIT to track personnel, evaluate weapon threats, and perform in-VRAM OCR analytics.