pyenfra

Python Package for Environmental Fractal Analysis and Chaos Estimation

Official repository: github: pyenfra

Description

pyenfra is a Python library for fractal analysis, modeling and chaos estimation in time-series with the emphasis on environmental datasets. Package contains a suite of fractal and chaos metrics: - Hurst exponent for rescaled range, - Detrended Fluctuation Analysis (DFA), - Multifractality by Generalized Hurst Slope, - Wavelet Transform Modulus Maxima (with varying bands and modulus methods), - Sample Entropy, - Recurrence Quantification Analysis (RQA), - Lyapunov exponents.

Installation

pyenfra can be installed using pip or locally by downloading package copy.

pip install: python pip install pyenfra local:

use repositiory to obtain package copy.

Usage

Below are a couple examples of package usage.

Please refer to examples.py for extended, detailed examples and computation workflows.

```python import numpy as np import matplotlib.pyplot as plt

import pyenfra

Generate White noise sample data

ts_white = np.random.RandomState(0).randn(2000)

Example: Compute Hurst exponent

hvalue = pyenfra.functions.hurst(tswhite, num=30, minn=10, minsegments=10)

Example: Interpret Hurst

print(pyenfra.interpreters.interprethurst(tswhite, useconfidenceinterval=False))

Example: Plot Hurst climacogram for AR(1)

axhurst = pyenfra.plotting.plothurst(tswhite, num=30, minn=10, minsegments=10, figsize=(5,4), scatterkwargs={'color':'C0'}, linekwargs={'color':'C1'}) axhurst.figure.suptitle("Climacogram: AR(1) Persistent Process") plt.show()

Example: Compute Lyapunov Exponent

lyapval, divergence, times = pyenfra.functions.lyapunov(tswhite, dim=3, tau=1, fs=1.0, maxiter=200, theiler=1) print(f"Estimated Lyapunov exponent (logistic r=3.99): {lyapval:.4f}") print("Interpretation:", pyenfra.interpreters.interpretlyapunov(lyapval)) ```

Roadmap

Future works on the package include: - [ ] HOST model integration. - [ ] On demand functions.

Contributing

Pull requests are welcome.

For major changes, please open an issue first to discuss implementation or changes.

Acknowledgment

This work was supported by NOAA grant NA19NOS4730207. Funding agency had no impact on work structure or findings.

License

This package is available under MIT license.