A Python library for efficient feature ranking and selection on sparse data sets.

A simplistic and efficient pure-python neural network library from Phys Whiz.

A small sample of a volumetric Lagrangian Coherent Structure code for GPUs and CPUs leveraging Numba.

A simple problem to compare execution speed between Numba and C++

Plasma Particle Dynamics (PPDyn), a python code to simulate plasma particles using Molecular Dynamics Algorithm. Numba JIT compiler for Python has been implemented for faster performance. Detailed documentation can be found at https://ppdyn.readthedocs.io/.

Modern Portfolio Theory (MPT) and Monte Carlo simulations to optimize and backtest a portfolio of various financial assets

Single neuronal models encoded in python language numba speed up. Note model waveforms are returned in Neo Format and model currents and voltages are represented with quantities units.

A customized sparse solver wrapper with Numba compatibility

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

QuantHedge-MM implements advanced computational methods for pricing and hedging options in markets with stochastic regime shifts. Built for quants and researchers, it extends Black-Scholes to Markov-modulated models.

A high-performance, parallelized order book simulator with trace-level performance introspection using magic-trace. Designed for systems engineers and quant-minded developers.

This repository contains an advanced tutorial on optimizing Python code for machine learning applications, focusing on processing large amounts of data efficiently. It covers three powerful libraries: Numba, NumPy, and Polars.

Building upon original repo, trying to implement encoder-decoder transformer using CUDA

This repository contains a Python script that performs portfolio risk analysis on selected assets using Value at Risk (VaR) and Expected Shortfall (ES) measures. The script uses Numba to optimize the risk calculations, downloads historical data from Yahoo Finance with yfinance, and visualizes the portfolio's return distribution

This project implements a CNN from scratch using NumPy and Numba to recognize hand signs representing digits (0-9) from the Sing Language MNIST dataset.

A 3D procedurally generated maze game, based off of 2D ray casts. Built in python using limited libraries.

Simulation of plant growth in 2D. Accelerated with Numba

This is a tutorial about Numba-CUDA

Performance Comparison for Conway's Game of Life

A near-zero overhead AVL Tree for Python that bypasses object-heavy limits using JIT compilation and manual memory management.