Need help with BayesBoost?
Click the “chat” button below for chat support from the developer who created it, or find similar developers for support.

About the developer

mpearmain
207 Stars 67 Forks Apache License 2.0 33 Commits 0 Opened issues

Description

Bayesian Optimization using xgboost and sklearn API

Services available

!
?

Need anything else?

Contributors list

BayesBoost

Bayesian Optimization using xgboost and sklearn API

Simple test scripts for optimal hyperparameter of xgboost using bayesian optimization.

Original bayesian optimization code is from https://github.com/fmfn/BayesianOptimization and all credit for this work goes to the original author.

To run the examples below you will need to install this package (it is under constant development)

pip install git+https://github.com/fmfn/BayesianOptimization.git

Example 1. is based on the otto dataset from Kaggle, this remains in memory. (https://www.kaggle.com/c/otto-group-product-classification-challenge)

Example 2. is based on Avazu click prediction dataset from Kaggle and requires the 'distributed' version of xgboost. (https://www.kaggle.com/c/avazu-ctr-prediction)

Run

To get this running create a data/otto and data/avazu dir and download the datasets into the respective directories and unzip / untar the files.

Dependencies: * BayesianOptimization (https://github.com/fmfn/BayesianOptimization) * Scipy * Numpy * Scikit-Learn * xgboost (https://github.com/dmlc/xgboost)

References: * http://papers.nips.cc/paper/4522-practical-bayesian-optimization-of-machine-learning-algorithms.pdf * http://arxiv.org/pdf/1012.2599v1.pdf * http://www.gaussianprocess.org/gpml/ * https://www.youtube.com/watch?v=vz3D36VXefI&index=10&list=PLE6Wd9FR--EdyJ5lbFl8UuGjecvVw66F6

We use cookies. If you continue to browse the site, you agree to the use of cookies. For more information on our use of cookies please see our Privacy Policy.