polytrend 1.0.9

PolyTrend is a regression tool that fits polynomial curves to noisy data.

PolyTrend

PolyTrend is a Python package aimed at facilitating polynomial trend fitting, visualization, and extrapolation. It offers a comprehensive set of functionalities to analyze and interpret data using polynomial regression techniques. Below, we provide an overview of the package along with additional formatting and explanations relevant for PyPI.

Introduction

PolyTrend is designed to approximate and plot a polynomial function onto given data, thereby aiding in the analysis of trends and patterns within datasets. Its development contributes to various fields including interpolation, polynomial regression, and approximation theory.

Key Functionalities

PolyTrend offers the following key functionalities:

• polyplot(): This method plots the polynomial fit based on specified degrees of the polynomial and the provided data.
• polyfind(): This method calculates the best-fit polynomial function by evaluating different polynomial degrees and selecting the one with the lowest Bayesian Information Criterion (BIC) score.
• polygraph(): This method visualizes the polynomial function, the known data points, and any extrapolated data points if provided.

Usage

Users can utilize PolyTrend to perform the following tasks:

1. Data Analysis: Analyze trends and patterns within datasets using polynomial regression techniques.
2. Visualization: Visualize polynomial fits alongside original data points to gain insights into the relationship between variables.
3. Extrapolation: Extrapolate future values based on the fitted polynomial function, enabling predictive modeling tasks.

Dependencies

PolyTrend relies on the following libraries for its computations and visualizations:

• NumPy
• pandas
• Matplotlib
• scikit-learn

Additional Resources

For further details on polynomial regression, refer to this wiki.

Installation

PolyTrend can be installed via pip:

pip install polytrend

Example Usage

import random
import polytrend as pt

# sample input
degrees = [1, 2, 3]
data = [
    (float(x), float(0.5 * x**2 - 2 * x + 1 + random.uniform(-1000, 1000)))
    for x in range(0, 100)
]

# Fit polynomial and visualize
pt.polyplot(degrees, data, extrapolate_data=[112, 140])

Feedback and Contributions

Feedback and contributions to PolyTrend are welcomed and encouraged. Please feel free to submit any issues or pull requests via the GitHub repository.

License

PolyTrend is licensed under the GNU GPL License. See the LICENSE file for details.