A framework for fast and interactive conducting machine learning experiments on tabular data


Keywords
Machine, Learning, Parallel, Computing, Feature, Engineering, categorical-features, data-science, ensemble-learning, feature-engineering, feature-extraction, feature-selection, kaggle, machine-learning, notebook, parallel-computing, pipeline, python, reproducibility, tutorial
License
MIT
Install
pip install kts==0.2.58

Documentation

KTS logo

PyPI version Docs CI Codecov CodeFactor

An interactive environment for modular feature engineering, experiment tracking, feature selection and stacking.

Install KTS with pip install kts. Compatible with Python 3.6+.

Modular Feature Engineering




Define features as independent blocks to organize your projects.

Source Code Tracking




Track source code of every feature and experiment to make each of them reproducible.

Parallel Computing and Caching




Compute independent features in parallel. Cache them to avoid repeated computations.

Experiment Tracking




Track your progress with local leaderboards.

Feature Selection




Compute feature importances and select features from any experiment
with experiment.feature_importances() and experiment.select().

Interactivity and Rich Reports




Monitor the progress of everything going on in KTS with our interactive reports.
From model fitting to computing feature importances.


Getting Started

Titanic Tutorial

Start exploring KTS with tutorial based on Titanic dataset. Run notebooks interactively in Binder or just read them in NBViewer.

1. Feature Engineering

nbviewer Binder

2. Modelling

nbviewer Binder

3. Stacking

nbviewer Binder

Documentation

Check out docs.kts.ai for a more detailed description of KTS features and interfaces

Inline Docs

Most of our functions and classes have rich docstrings. Read them right in your notebook, without interruption.


Acknowledgements

MVP of the project was designed and implemented by the team of Mikhail Andronov, Roman Gorb and Nikita Konodyuk under the mentorship of Alexander Avdyushenko during a project practice held by Yandex and Higher School of Economics on 1-14 February 2019 at Educational Center «Sirius».