• Windows users: pip installation may not work on some Windows environments, and it may cause unexpected errors.

  Installation from pip on Windows is therefore currently disabled for further investigation; please install from Github instead.

• If you want to run XGBoost process in parallel using the fork backend for joblib/multiprocessing, you must build XGBoost without support for OpenMP by make no_omp=1. Otherwise, use the forkserver (in Python 3.4) or spawn backend. See the sklearn_parallel.py demo.

Since this package contains C++ source code, pip needs a C++ compiler from the system to compile the source code on-the-fly.

On macOS, gcc@5 is required as later versions remove support for OpenMP. See here for more info.

Please install gcc@5 from Homebrew:

brew install gcc@5

After installing gcc@5, set it as your compiler:

export CC = gcc-5
export CXX = g++-5

Please install gcc:

sudo apt-get install build-essential      # Ubuntu/Debian
sudo yum groupinstall 'Development Tools' # CentOS/RHEL

For a stable version, install using pip:

pip install xgboost

For an up-to-date version, install from Github:

• Run ./build.sh in the root of the repo.

• Make sure you have setuptools installed: pip install setuptools

• Install with cd python-package; python setup.py install from the root of the repo

• For Windows users, please use the Visual Studio project file under the Windows folder. See also the installation tutorial from Kaggle Otto Forum.

• Add MinGW to the system PATH in Windows if you are using the latest version of xgboost which requires compilation:

  python
  import os
  os.environ['PATH'] = os.environ['PATH'] + ';C:\\Program Files\\mingw-w64\\x86_64-5.3.0-posix-seh-rt_v4-rev0\\mingw64\\bin'
  

• Refer also to the walk through example in demo folder.
• See also the example scripts for Kaggle Higgs Challenge, including speedtest script on this dataset.