fixes #146

A better (more involved) solution might be to check for a TensorFlow installation manually in setup.py and deal with that accordingly.

Force shebang lines to Python 3

fix misspellings

Docstring fix

Call ppo2 and not ppo1 in ppo2 README.md

This is a Python 3-only library. A shebang with `#!/usr/bin/env python`
will launch python2 on many systems which do not have python3
installed. Setting the shebang to `#!/usr/bin/env python3` will show a
useful error on systems without Python 3.

some logger changes

Fixes frame stacking in A2C and ACKTR for multi-channel observations

Fixes the NaN issues in ACKTR + bug in run_mujoco.py

Use standardized advantages in trpo.

Fix command lines

Deprecated VARIABLES -> GLOBAL_VARIABLES.

Fix relative links in README.md

Adding links to source files

- removed vf clipping in pposgd - that was severely degrading performance on mujoco because it didn’t account for scale of returns
- switched adam epsilon in pposgd_simple
- brought back no-ops in atari wrapper (oops)
- added readmes
- revamped run_X_benchmark scripts to have standard form
- cleaned up DDPG a little, removed deprecated SimpleMonitor and non-idiomatic usage of logger

* Export param noise

* Update documentation

* Final finishing touches

ppo and trpo