Removing outliers from a pandas dataframe


Keywords
CLI, OUTLIER, Data, outlier removal
Install
pip install outlierpack-NG==0.0.5

Documentation

OUTLIER REMOVAL

PROJECT 2, UCS633 - Data Analysis and Visualization
Nikhil Gupta  
COE17
Roll number: 101703371

Output is the number of rows removed from the input dataset. The remaining rows of the dataset are streamed to a new csv file whose name is required as an input.

The no of rows removed: 5

Note the outlier removal is performed using IQR method.

Installation

pip install outlierpack_NG

Note the name has an underscore not a hyphen. If installation gives error or package is not found after installing, install as sudo.

Recommended - test it out in a virtual environment.

To use via command line

Outcli myData.csv newData.csv

First argument after outcli is the input csv filename from which the dataset is extracted. The second argument is for storing the final dataset after processing.

To use in .py script

from outlib.models import r_outliers
r_outliers('myData.csv', 'newData.csv')

Can email me for any issues or suggestions