Genderly
Description
A library to classify an indian name as Male or Female. Mainly focused around making the user onboarding smooth for eCommerce merchants by pre-filling the gender for their users during a signup.
Installation
pip install genderly
Usage
The library comes with a pre trained model which can be used as follows
import genderly genderly.decide_gender(['raj'])
or
import genderly genderly.decide_gender(df['first_name'])
Note : It is recommended that the first name being passed to the model is atleat 3 letter long. The model wouldnot work with last names or surnames. It has been trained on ly to work with first names.
The function decide_gender() taked Pandas Series object and lists as input.
In case one wants to rerun the training
import genderly genderly.start_training()
The model has been trained on the basis of open Indian names database. One can add more data to the csv file which can be found at genderly/data/list_NameGender.csv
Dependencies python 2.7 pandas numpy scikit-learn