No description
  • Rust 96.8%
  • Go 3.2%
Find a file
Repository files (latest commit first)
Filename Latest commit message Latest commit date
Clouds 61d0b618e6
All checks were successful
Run tests / test-rust (push) Successful in 15s
Merge pull request 'Dagger' (#1) from dagger into latest
Reviewed-on: #1
2026-06-04 11:26:15 +00:00
.forgejo/workflows add dagger pipeline 2026-06-04 12:23:41 +02:00
benches implement mandelbrot function and benchmark move vs mut 2024-05-11 14:55:59 +02:00
examples fix lint 2026-06-04 12:27:56 +02:00
pipelines fix lint 2026-06-04 12:27:56 +02:00
src copy vector generators 2026-01-21 19:41:34 +01:00
.gitignore remove tnar files 2026-01-21 20:12:10 +01:00
.gitmodules add AnT example and AnT submodule 2024-04-20 22:37:19 +02:00
Cargo.lock cargo upgrade 2025-10-23 21:10:58 +02:00
Cargo.toml cargo upgrade 2025-10-23 21:10:58 +02:00
dagger.json add dagger pipeline 2026-06-04 12:23:41 +02:00
LICENSE add gpl 3 2024-07-18 21:58:42 +02:00
README.org update readme 2024-07-03 20:08:11 +02:00

RAnT

A Rust library for numerical simulation of non-linear systems.

Idea

This library aims to be an alternative to other numerical simulators. The tool AnT, that I was using extensively during my masters thesis, is the inspiration for this project.

In contrast to the tool mentioned above, with this library one writes the function of interest in rust code and passes it to the appropriate library functions for simulation and analysis. We hope to achieve performance improvements through compiler optimizations and static linking.

Roadmap [7/10]

  1. Simulate 1D floating-point R -> R functions

    • Cycles and Periods
  2. Scanning parameters 1D
  3. Test performance of generic S -> S functions
  4. Test logistic period results
  5. ND scans
  6. Write result to file
  7. Parallelization
  8. More scanning options

    • initial states (keep from last scan point)
    • multi dimensional for both initial states and parameters
  9. Image generation

    • prototype
    • library
  10. Interactive image generation (zoom and pan)

Design decisions

No implementation of extra library functions for 1D functions

A benchmarking of a generic implementation and a static implementation for 1D functions found no significant difference (58ms vs. 60ms). Suprisingly, the generic implementation was a little faster. Therefore, we will only keep the generic implementation for improved maintenance.

Parallelization

  • Parallel iterators thanks to rayon
  • Each iterator computes one line sequentially at a time