Adds support for Hindsight Experience Replay (HER) (#299)

* Add Hindsight Experience Replay (HER)

* Minor improvements

setup.py

@@ -10,7 +10,7 @@ setup(name='baselines',
       packages=[package for package in find_packages()
                 if package.startswith('baselines')],
       install_requires=[
-          'gym[mujoco,atari,classic_control]',
+          'gym[mujoco,atari,classic_control,robotics]',
           'scipy',
           'tqdm',
           'joblib',
@@ -19,9 +19,11 @@ setup(name='baselines',
           'progressbar2',
           'mpi4py',
           'cloudpickle',
+          'tensorflow>=1.4.0',
+          'click',
       ],
       description='OpenAI baselines: high quality implementations of reinforcement learning algorithms',
       author='OpenAI',
       url='https://github.com/openai/baselines',
       author_email='gym@openai.com',
-      version='0.1.4')
+      version='0.1.5')