Crate temporal_attractor_studio 

Temporal Attractor Studio - Real FTLE and Lyapunov exponent calculation

This crate provides a complete implementation for calculating Finite-Time Lyapunov Exponents (FTLE) and estimating the largest Lyapunov exponent from trajectory data or time series using:

• Delay embedding for univariate time series
• Theiler window exclusion to avoid temporal neighbors
• VP-tree for efficient nearest-neighbor search
• Parallel slope calculation over early divergences

Examples:

use temporal_attractor_studio::prelude::*;

fn main() -> Result<(), Box<dyn std::error::Error>> {
    // For multivariate state data
    let trajectory = vec![
        vec![1.0, 2.0, 3.0],
        vec![1.1, 2.1, 3.1],
        vec![1.2, 2.2, 3.2],
        vec![1.3, 2.3, 3.3],
        vec![1.4, 2.4, 3.4],
    ];

    let result = estimate_lyapunov(&trajectory, 0.01, 12, 20, 4000, 1e-12)?;
    println!("Lyapunov exponent: {:.6}", result.lambda);
    println!("Doubling time: {:.3} time units", result.doubling_time);

    // For univariate time series with delay embedding
    let series = vec![1.0, 2.1, 1.9, 3.2, 2.8, 4.1, 3.7, 5.2];
    let embedded = delay_embed(&series, 3, 1)?;
    let result = estimate_lyapunov_default(&embedded)?;
    println!("Embedded series Lyapunov exponent: {:.6}", result.lambda);

    Ok(())
}

Re-exports

pub use ftle::*;

Modules

attractor

Attractor engine for temporal dynamics

echo_state

Echo-state network module for temporal prediction

ftle

Core FTLE module implementing real algorithms from lyapfit research FTLE and Lyapunov exponent calculator via nearest-neighbor divergence.

prelude

Prelude module for easy imports

time_expansion_bridge

Time expansion bridge for consciousness integration

Enums

TemporalStudioError

Central error type for Temporal Attractor Studio

Functions

init

Initialize the framework with logging

Type Aliases

StudioResult

Result type for the studio