Skip to main content

PunditKit: A GUI for Scikit-Learn Models

Project description


Simplify. Visualise. Learn.


PunditKit is a free toolkit for exploratory data analysis and modelling of tabular data - such as a database or a spreadsheet - with a simple interface, visual diagnostics, and the choice of a large number of different models.

Included Features:

  • Machine learning models using scikit-learn
  • Interpretable explanations of predictions using lime
  • Exploratory data summaries for checking datasets
  • Model diagnostics for evaluating the effectiveness of different models
  • Feature importance: identify key drivers of the response
  • Partial dependence plots: understand relationships with a particular feature.

PunditKit is under active development. The goal is to develop an opinionated modelling framework with best practice modelling and visualisation techniques. If you encounter any problems or have any feature requests, please consider raising an issue.

Installation (via pip)

PunditKit is developed using Python. First download and install a Python 3.x distribution such as Anaconda

PunditKit can then be installed using pip from the command line (if pip is added to PATH during installation) or using Anaconda prompt.

pip install punditkit

This adds the punditkit command.

Modelling a dataset

Suppose you have a file called iris.csv that you would like to model.

To use punditkit on the dataset, run:

punditkit iris.csv

Currently only Comma Separated Values (CSV) datasets are supported. Datasets within Excel spreadsheets can be analysed using punditkit, use Save As CSV from within Excel to convert to the right format.

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

punditkit-0.0.2.tar.gz (16.9 kB view hashes)

Uploaded Source

Built Distribution

punditkit-0.0.2-py3-none-any.whl (32.7 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page