shangzhihao

shangzhihao / HyperStar2

Public

Easy distribution fitting of correlated data

10
0
100% credibility
Found Feb 23, 2026 at 10 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Java
AI Summary

HyperStar2 is a desktop application for loading time-series interval data from text files, visualizing it with interactive charts, fitting phase-type statistical models, and exporting the results.

How It Works

1
🔍 Discover HyperStar2

You find this handy desktop tool for analyzing your time-series data patterns, like intervals between events.

2
🚀 Start the app

You run the program easily, and a clean window opens ready for your data.

3
📁 Load your data

Pick a simple text file with one number per line, like your sample timings, and watch it load smoothly.

4
📊 View your charts

See colorful graphs of your data's histogram, cumulative view, or correlations to understand its shape.

5
🔧 Fit a model

Choose from easy options like basic curves or advanced mixtures, click fit, and let it crunch the numbers quickly.

6
✨ See the magic overlay

Your graphs update with smooth fitted lines matching your data perfectly, so you can spot how well it fits.

💾 Save your results

Export the model details to a file, ready to use in your work or reports, feeling accomplished.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 10 to 10 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is HyperStar2?

HyperStar2 is a JavaFX desktop app for easy distribution fitting of correlated time-series interval data to phase-type models like Exponential, Erlang, Hyper-Erlang, and Markov Arrival Processes. Load samples from a text file (one number per line), trim by percentage or range, and visualize histograms, CDFs, and autocorrelation plots with overlaid fitted curves. Run via Gradle (`./gradlew run`), interact by adding peak markers on PDFs, and export matrices like alpha/Q or D0/D1 for simulations.

Why is it gaining traction?

It streamlines the easy distribution process for correlated data in a clickable GUI, skipping custom scripts for k-means clustering or moment matching. The hook is instant visual feedback on fits, plus PSO optimization for MAPs, making it faster than pure libraries for exploratory work. Ties into 31150 papers on phase-types, offering an easy GitHub project for quick contributions like sample datasets.

Who should use this?

Queueing theorists fitting inter-arrival times from traces, performance modelers building MAPs for network simulations, and stochastic process researchers needing Hyper-Erlang overlays on empirical plots. Ideal for academics citing the HyperStar2 paper or devs prototyping easy distribution NC/WMS models without deep coding.

Verdict

Solid niche tool for phase-type fitting, but with 10 stars and 1.0% credibility score, it's immature—docs are README-only, no tests visible. Try for quick correlated fits if JavaFX fits your stack; otherwise, await more polish or fork for easy GitHub badges.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.