this is a python module which is intended for data scientist and statistician to visualize features and understand relationships between different features without writing lots of codes.
This module is like an aggregator which combines multiple functionalities of different modules and brings them together to keep you out of trouble
installation
pip install dumbledore
requirements
this module depends on all most of scientific libraries like pandas , matplotlib , seaborn, numpy etc. all of which you must have already installed if you are a data scientist or statistician but don't worry if you haven't the pip command will automatically install all the dependencies.
examples
the vis_feature() function helps you understand and visualize between different types of variables
- for a continuous feature and categorical target variable
import dumbledore as db
import pandas as pd
...
# load data
...
db.basics.vis_feature((data , 'gender' , 'tenure', 'gist_earth_r' , target_continous=True , jitter=0.3)
- when your feature and target both are continous
db.basics.vis_feature((data , 'tenure' , 'MonthlyCharges', 'gist_earth_r' , continous=True , target_continous=True )
similarly you can feed in categorical target or categorical feature you have to keep in mind the two arguements
continous : True if feature is continous, False if feature categorical
target_continous : True if target variable is continous , False if target variable is categorical
note : this module is still under development and has been released as alpha. so if you get any bug please open a issue without hesitation. or if you want a new feature please make a feature request
THIS PRODUCT IS DISTRIBUTED UNDER GNU GPLv3 WHICH CAN BE FOUND IN THE LICENSE FILE
developed by : imzain448
meet me on
- Gmail : ahmadzain.448@gmail.com