Dealing with outliers
Project description
py_outliers_utils
Overview
As data rarely comes ready to be used and analyzed for machine learning right away, this package aims to help speed up the process of cleaning and doing initial exploratory data analysis specific to outliers. The package focuses on the tasks of identifying univariate outliers, providing summary of outliers like count, range of outliers, visualize them and giving functionality to remove them from data.
Installation
$ pip install py_outliers_utils
Functions
The three functions contained in this package are as follows:
outlier_identifier
: A function to identify outliers in the dataset and provide their summary as an outputvisualize_outliers
: A function to generate the eda plots highlighting outliers providing additional functionality to visualize themtrim_outliers
: A function to generate outlier free dataset by imputing them with mean, median or trim entire row with outlier from dataset.
Our Place in the Python Ecosystem
While Python packages with similar functionalities exist, this package aims to provide summary, visualization of outliers in a single package with an additional functionality to generate outlier-free dataset. Few packages with similar functionality are as follows:
Usage
To import the package and check the version:
import py_outliers_utils
print(py_outliers_utils.__version__)
py_outliers_utils
can be used to deal with the outliers in a dataset and plot the distribution of the dataset. More information can be found in the documentation.
Contributing
This package is authored by Karanpreet Kaur, Linhan Cai, Qingqing Song as part of the course project in DSCI-524 (UBC-MDS program). You can see the list of all contributors in the contributors tab.
We welcome and recognize all contributions. If you wish to participate, please review our Contributing guidelines
License
py_outliers_utils
is licensed under the terms of the MIT license.
Credits
py_outliers_utils
was created with cookiecutter
and the py-pkgs-cookiecutter
template.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for py_outliers_utils-1.2.9-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 6c3af37836a10b9b21d82973b0e07d735ab569801a74cfa5f4d6e1226bad7101 |
|
MD5 | 898294d0ffa52064bfec8e504955d662 |
|
BLAKE2b-256 | b90fa2fce9a71528f819c4f67783c81f26b53daaa39968fb271cfcb33dcc03e4 |