============ Installation ============ Requirements ============ Python 3.5 or newer, NumPy and SciPy packages, and a working c-compiler for building extensions are needed for installing RLScore. For example `Anaconda `_ is a Python distribution, available for Windows, OS X and Linux, that includes the required dependencies. For Windows, you need to also install `Microsoft Visual C++ Compiler for Python `_. Installation ============ For bleeding edge versions, we recommend cloning the git repository: `https://github.com/aatapa/RLScore `_, and install using setupy.py, for example: global installation :: python setup.py install installing to home directory :: python setup.py install --home= compile extensions inside source folder (you need to add RLScore folder to PYTHONPATH to use it) :: python setup.py build_ext --inplace Alternatively, it is possible to install the latest pip indexed release: pip install rlscore