pinemosquito / xgboost

XGBoost on Mac OS

XGBoost (eXtreme Gradient Boosting) is an advanced implementation of gradient boosting trees.

It is a machine learning algorithm that is very successful at tackling Kaggle's competition.

I was not able to install XGBoost on my Mac until I came across Jean Francois Puget's blog by chance.

Installation

go to root directory, type git clone command

git clone --recursive https://github.com/dmlc/xgboost

This downloads XGBoost code into xgboost directory

Compile for multi-threading

By default, Mac OS use cc/cc++ compilers, which do not support open MP option as required by XGBoost multi-threading. We need to specify how to compile

Change directory to XGBoost

cd xgboost

First copy this file so that any local changes will be ignored by git

cp make/config.mk

Open make/config.mk and modify

    vi make/config.mk
    Uncomment export CC = gcc
    Uncomment export CXX = g++

shift zz to save and exit

After lot of trials and errors, I found that I was using gcc 7.1.0. So I need to edit make/config.mk

    export CC = gcc-7
    export CXX = g++-7

where 7 refers to version of gcc

Build XGBoost

    cd xgboost; cp make/config.mk ./config.mk; make -j4

Running XGBoost

Once the build is finished, we can use XGBoost with its command line. I am using Python, hence I performed this final step. You may need to enter the admin password to execute it.

    cd python-package; sudo python setup.py install