https://github.com/khushi130404/outlier_exterminator
Outlier_Exterminator is a Python tool for detecting and treating outliers using IQR, Z-Score, and Percentile methods. It supports trimming, capping, and Winsorization, demonstrated in a Jupyter Notebook.
https://github.com/khushi130404/outlier_exterminator
capping iqr machine-learning numpy percentile-method trimming winsorization z-score
Last synced: about 1 month ago
JSON representation
Outlier_Exterminator is a Python tool for detecting and treating outliers using IQR, Z-Score, and Percentile methods. It supports trimming, capping, and Winsorization, demonstrated in a Jupyter Notebook.
- Host: GitHub
- URL: https://github.com/khushi130404/outlier_exterminator
- Owner: Khushi130404
- License: mit
- Created: 2024-12-22T05:06:38.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2024-12-27T03:54:01.000Z (over 1 year ago)
- Last Synced: 2025-09-03T18:56:36.566Z (10 months ago)
- Topics: capping, iqr, machine-learning, numpy, percentile-method, trimming, winsorization, z-score
- Language: Jupyter Notebook
- Homepage:
- Size: 455 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
- Security: SECURITY.md
Awesome Lists containing this project
README
# Outlier_Exterminator
Outlier_Exterminator is a Python-based tool designed to handle and process outliers in datasets. It provides implementations of the Interquartile Range (IQR), Z-Score, and Percentile methods to detect and treat outliers. The tool supports various outlier treatment techniques, including trimming, capping, and Winsorization.
This project is developed and demonstrated in a Jupyter Notebook for ease of use and understanding.
## Features
1. Outlier Detection Methods :-
- IQR Method: Detects outliers using the Interquartile Range.
- Z-Score Method: Identifies outliers based on the standard deviation.
- Percentile Method: Flags outliers based on specified percentile thresholds.
2. Outlier Treatment Techniques :-
- Trimming: Removes detected outliers from the dataset.
- Capping: Replaces outliers with the nearest valid value within bounds.
- Winsorization: Replaces extreme values with predefined percentiles.
3. Customizable Parameters :-
- Configure thresholds for Z-scores.
- Set percentile ranges for trimming and Winsorization.
- Define the IQR multiplier for flexible outlier detection.
4. Visualization :-
- Generate boxplots and histograms to visualize the effects of outlier treatment.
## Required Libraries :-
- pandas
- numpy
- matplotlib
- seaborn