End of Distribution

end_of_distribution is a Python library designed for statistical outlier detection, focusing on identifying values at the "end of the distribution" in datasets. It provides methods for detecting outliers using both Z-score and Interquartile Range (IQR), commonly used techniques in data preprocessing and analysis.

Features

• Z-Score Outliers: Identify outliers based on their distance from the mean.
• IQR Outliers: Detects outliers based on the interquartile range, robust against extreme values.

Installation

After cloning the repository:

git clone https://github.com/adityayadav0111/end_of_distribution.git
cd end_of_distribution

To install directly if available on PyPI:

pip install end_of_distribution

Usage

Import the functions from end_of_distribution to detect outliers in your data.

from lib.end_of_distribution import z_score_outliers, iqr_outliers

# Example data
data = [1, 2, 3, 4, 100, 5]

# Detect outliers with Z-score
z_outliers = z_score_outliers(data, threshold=3)
print("Z-score Outliers:", z_outliers)

# Detect outliers with IQR
iqr_outliers = iqr_outliers(data)
print("IQR Outliers:", iqr_outliers)

API Reference

• z_score_outliers(data, threshold=3)

  • Parameters:
    • data (list): List of numerical values to check for outliers.
    • threshold (float): Z-score threshold for identifying outliers. Default is 3.
  • Returns: A list with True/False for outliers in the dataset.

• iqr_outliers(data)

  • Parameters:
    • data (list): List of numerical values to check for outliers.
  • Returns: A list with True/False indicating outliers.

Contributing

We welcome contributions! Here’s how to get started:

1. Fork the repository and clone it locally.
2. Create a new branch for your contribution.
3. Make your changes and commit.
4. Push your changes and open a pull request.

License

This project is licensed under the MIT License - see the LICENSE file for details.