Local changes

.. _installation:

************

Installation

============

************

We recommend that you install ``deepchem`` with ``conda``. ::

  $ conda install -c omnia deepchem 

Conda is a cross-platform package manager built especially for scientific

python. It will install ``deepchem`` along with all dependencies from a

pre-compiled binary. If you don't have Python or the ``conda`` package

manager, we recommend starting with the `Anaconda Scientific Python

distribution <https://store.continuum.io/cshop/anaconda/>`_, which comes

pre-packaged with many of the core scientific python packages that deepchem 

uses (see below).

Supported Platforms

===================

Currently, we test and run ``deepchem`` with Python 2.7 on

- x86-64 Ubuntu 12.04 LTS, 14.04 LTS

- x86-64 CentOS 5.x, 6.x

From Source

-----------

===========

``deepchem`` is a python package which leans heavily on the python

scientific-computing ecosystem.  Since the current version of ``deepchem``

only supports installation from source, we need to install prerequisite

packages frrom various sources.

Anaconda Python

~~~~~~~~~~~~~~~

---------------

To facilitate ease-of-install, we strongly recommend that users install

Anaconda python. Anaconda will install a number of python scientific

`installation instructions <http://docs.continuum.io/anaconda/install#linux-install>`_.

Cheminformatics Packages

~~~~~~~~~~~~~~~~~~~~~~~~

------------------------

``deepchem`` uses `rdkit <http://www.rdkit.org/docs/Install.html>`_  and

`openbabel <https://github.com/openbabel/openbabel>`_ to facilitate

    $ conda install -c omnia openbabel

Machine Learning Packages

~~~~~~~~~~~~~~~~~~~~~~~~~

``deepchem`` uses a variety of backend machine learning

packages. Anaconda python will install ``scikit-learn`` and ``theano`` by

default. ``scikit-learn`` wraps a variety of standard machine learning

models in a convenient python API, while ``theano`` is a tensor

manipulation library.  Many models in ``deepchem`` use [keras](keras.io), a

convenient deep-learning wrapper for Theano. ``deepchem`` depends on additions

to ``keras`` which have not been merged into the master ``keras`` source,

so we need to install a custom version of ``keras`` from source.

-------------------------

``deepchem`` uses a variety of backend machine learning packages. Anaconda

python will install ``scikit-learn`` by default. ``scikit-learn`` wraps a

variety of standard machine learning models in a convenient python API.

``theano`` is a tensor manipulation library.  Many models in ``deepchem``

use [keras](keras.io), a convenient deep-learning wrapper for Theano.

``deepchem`` requires recent additions to ``theano`` and ``keras`` not yet

included in the latest releases, so we recommend installing from ``conda``.

.. code-block:: bash

    $ git clone https://github.com/pandegroup/keras.git

    $ cd keras/

    $ python setup.py install

    $ conda install -c omnia theano 

    $ conda install -c omnia keras 

Installing ``deepchem`` from source

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

-----------------------------------

.. code-block:: bash

    $ git clone https://github.com/pandegroup/deepchem.git

    $ cd deepchem/

    $ python setup.py develop

Testing the Installation

=========================

Running the tests verifies that everything is working. The test

suite uses `nose <https://nose.readthedocs.org/en/latest/>`_, which can be

installed via ``pip``. ::

  pip install nose

Then execute the command ::

  nosetests -v deepchem