Metadata-Version: 2.1
Name: estimatePDF
Version: 0.1.0
Summary: Probability Density Function Estimation Library
Home-page: https://github.com/ShanSarkar75/estimatePDF
Author: Shantanu Sarkar
Author-email: Shantanu Sarkar <shantanu75@gmail.com>
License: MIT
Project-URL: Homepage, https://github.com/ShanSarkar75/estimatePDF
Project-URL: Documentation, https://github.com/ShanSarkar75/estimatePDF#readme
Project-URL: Repository, https://github.com/ShanSarkar75/estimatePDF
Project-URL: BugTracker, https://github.com/ShanSarkar75/estimatePDF/issues
Keywords: PDF,probability density,statistics,Python,data science
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Intended Audience :: Science/Research
Classifier: Topic :: Scientific/Engineering :: Mathematics
Requires-Python: >=3.8
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: numpy
Requires-Dist: scipy
Requires-Dist: pandas
Requires-Dist: matplotlib
Requires-Dist: tensorflow

# estimatePDF

**A Python library for Probability Density Function (PDF) estimation using Kernel Density, Histogram Density, and Dual Polynomial Regression.**

---

## Overview

`estimatePDF` estimatePDF is a Python package for probability density function (PDF) estimation and sampling. It provides computationally efficient, GPU-optimized implementations using TensorFlow along with custom polynomial regression methods designed to capture asymmetry in distributions.

---

## Features

- **Density Estimation**
  - SciPy Gaussian KDE
  - TensorFlow-based KDE (for graph execution)
  - Histogram Density Estimation (HDE)

- **Probability Density Functions**
  - Gaussian PDF
  - Asymmetric Laplace PDF & sampling
  - M-Wright functions and variants
  - Inverse Transform Sampling

- **Dual Polynomial Regression (DPR)**
  - Piecewise polynomial PDF approximation
  - Gradient-based threshold detection
  - Fits multimodal or skewed PDFs

---

## Installation

Install from PyPI:

```bash
pip install estimatePDF
