Skip to content

AYBU-ParLab/MatGen

 
 

Repository files navigation

MatGen - Sparse Matrix Generator

⚠️ Disclaimer

This repository is a fork of the original project: MatGen---A-Realistic-Sparse-Matrix-Generator by aliemrepmk. All credits for the original implementation go to the original author. This fork may contain modifications, improvements, or additional functionality.

Description

MatGen is a tool for generating, modifying, and analyzing sparse matrices. It supports various generation methods and includes tools for processing and visualization. It can be used in scientific computing, machine learning experiments, or engineering applications where realistic sparse matrices are needed.

Features

  • Matrix Generation: Create sparse matrices with control over size, sparsity, and structure. Includes Fourier, wavelet, bilinear, graph-based, and neural network-based generation options.
  • Matrix Transformation: Resize or reshape matrices while preserving key properties. Batch processing supported.
  • Visualization: View matrices using built-in heatmaps or graph layouts to understand structure.
  • Problem-Solving Modules: Test matrices in simulated numerical problems like heat flow or mechanical systems.
  • Web Interface: Upload, generate, transform, and view matrices through a browser.
  • Parallel Dataset Tools: Matrix downloader, dataset generator, and feature calculator that run in parallel using multi-threading. Supports fast downloading and batch feature extraction.

Contributors

  • Ali Emre Pamuk – Simulations, image processing, frontend/backend
  • Faruk Kaplan – Signal processing, neural networks, optimization, backend
  • Mert Altekin – Simulations, graph processing, frontend/backend
  • Yousif Suhail – Matrix downloader, dataset generator, feature calculator with multithreading and concurrent downloading

About

A Realistic Sparse Matrix Generator Using Signal processing and Image Processing Methods

Resources

Stars

Watchers

Forks

Packages

 
 
 

Contributors

Languages

  • Python 79.4%
  • HTML 12.9%
  • JavaScript 6.8%
  • CSS 0.9%