ggwp 0.0.47

Prepare Fast, Analyze Faster

ggwp: Prepare Fast, Analyze Faster

What is ggwp?

ggwp is a Python package for fast and easy data analytics.
It aims to make a data model that can be applied for some use cases
such as customer analytics. The data model created by ggwp is designed
directly from my personal experiences, which will be more and more data models
in the future. In addition, ggwp now has some new features for
logging, modeling and evaluating your models, which of these can speed up your workflow
FOR REAL!!

Main Features

Here are current features available in ggwp

from ggwp.EzDataModel import *

Example codes ->

• DataModel: prepare your raw data for a general data model
• RFMT: create RFMT dataset
• Cohort: create Cohort dataset
• CustomerMovement: create Customer Movement dataset
• BasketEvolving: create Basket Evolving dataset

from ggwp.EzModeling import *

Example codes ->

• Check: check your data quality and more
• Log: log your data, its status and your remark, so you know what you've done for each stage
• Evaluation: evaluate your model using traditional (R2, RMSE, F-1, ACC, ROC) and practical metrics (Cost&Benefit, Lift)
• BenchMark: benchmark your models' performances giving you some intuition

from ggwp.EzPipeline import *

Example codes ->

• GroupImputer: impute your value with multiple subsets (as you desire)
• ConvertVariables: convert all of your columns in one command
• ColumnAssignment: adding a new column in your pipeline is now available
• OneHotTransformer: the brand-new One-Hot-Encoding is now at your service
• DataFrameTransformer: write your own function and use it along with your pipeline seamlessly

Where to get ggwp

The source code is currently hosted at GitHub:
https://github.com/datanooblol/ggwp

The latest version is available at
Python Package Index (PyPI)

# pip  
pip install ggwp  

Dependencies

• numpy
• pandas
• sklearn
• xgboost

Disclaimer

ggwp is now in devoping phase. Hence, if you experience any inconveniences, please be patient...