update of api docstrings across coach and tutorials [WIP] (#91)

* updating the documentation website
* adding the built docs
* update of api docstrings across coach and tutorials 0-2
* added some missing api documentation
* New Sphinx based documentation

docs/_sources/features/algorithms.rst.txt

@@ -0,0 +1,10 @@
+Algorithms
+==========
+
+Coach supports many state-of-the-art reinforcement learning algorithms, which are separated into three main classes -
+value optimization, policy optimization and imitation learning.
+A detailed description of those algorithms may be found in the `agents <../components/agents/index.html>`_ section.
+
+.. image:: /_static/img/algorithms.png
+   :width: 600px
+   :align: center

docs/_sources/features/benchmarks.rst.txt

@@ -0,0 +1,22 @@
+Benchmarks
+==========
+
+Reinforcement learning is a developing field, and so far it has been particularly difficult to reproduce some of the
+results published in the original papers. Some reasons for this are:
+
+* Reinforcement learning algorithms are notoriously known as having an unstable learning process.
+  The data the neural networks trains on is dynamic, and depends on the random seed defined for the environment.
+
+* Reinforcement learning algorithms have many moving parts. For some environments and agents, there are many
+  "tricks" which are needed to get the exact behavior the paper authors had seen. Also, there are **a lot** of
+  hyper-parameters to set.
+
+In order for a reinforcement learning implementation to be useful for research or for data science, it must be
+shown that it achieves the expected behavior. For this reason, we collected a set of benchmark results from most
+of the algorithms implemented in Coach. The algorithms were tested on a subset of the same environments that were
+used in the original papers, and with multiple seed for each environment.
+Additionally, Coach uses some strict testing mechanisms to try and make sure the results we show for these
+benchmarks stay intact as Coach continues to develop.
+
+To see the benchmark results, please visit the
+`following GitHub page <https://github.com/NervanaSystems/coach/tree/master/benchmarks>`_.

docs/_sources/features/environments.rst.txt

@@ -0,0 +1,31 @@
+Environments
+============
+
+Coach supports a large number of environments which can be solved using reinforcement learning.
+To find a detailed documentation of the environments API, see the `environments section <../components/environments/index.html>`_.
+The supported environments are:
+
+* `DeepMind Control Suite <https://github.com/deepmind/dm_control>`_  - a set of reinforcement learning environments
+  powered by the MuJoCo physics engine.
+
+* `Blizzard Starcraft II <https://github.com/deepmind/pysc2>`_  - a popular strategy game which was wrapped with a
+  python interface by DeepMind.
+
+* `ViZDoom <http://vizdoom.cs.put.edu.pl/>`_  - a Doom-based AI research platform for reinforcement learning
+  from raw visual information.
+
+* `CARLA <https://github.com/carla-simulator/carla>`_  - an open-source simulator for autonomous driving research.
+
+* `OpenAI Gym <https://gym.openai.com/>`_  - a library which consists of a set of environments, from games to robotics.
+  Additionally, it can be extended using the API defined by the authors.
+
+  In Coach, we support all the native environments in Gym, along with several extensions such as:
+
+  * `Roboschool <https://github.com/openai/roboschool>`_  - a set of environments powered by the PyBullet engine,
+    that offer a free alternative to MuJoCo.
+
+  * `Gym Extensions <https://github.com/Breakend/gym-extensions>`_  - a set of environments that extends Gym for
+    auxiliary tasks (multitask learning, transfer learning, inverse reinforcement learning, etc.)
+
+  * `PyBullet <https://github.com/bulletphysics/bullet3/tree/master/examples/pybullet>`_  - a physics engine that
+    includes a set of robotics environments.

docs/_sources/features/index.rst.txt

@@ -0,0 +1,10 @@
+Features
+========
+
+.. toctree::
+   :maxdepth: 1
+   :caption: Features
+
+   algorithms
+   environments
+   benchmarks