 Richard LiawandRobert Nishihara
|
94f32db5e6
|
A3C Polishing (#385)
* number
* gym doesn't have versioning
* Benchmarks
* visualization
* formatting
* small fix for tensorboard
* first pass removing universe dependency
* code
* results polish
* removed extra line
* removed universe dependency
* doc
* remove gym versioning stuff
* changes as suggested
* nit
|
2017-04-11 22:51:52 -07:00 |
|
|
Richard LiawandRobert Nishihara
|
b463d9e5c7
|
Initial A3C Example - PongDeterministic-v3 (#331)
* Initializing A3C code
* Modifications for Ray usage
* cleanup
* removing universe dependency
* fixes (not yet working
* hack
* documentation
* Cleanup
* Preliminary Portion
Make sure to change when merging
* RL part
* Cleaning up Driver and Worker code
* Updating driver code
* instructions...
* fixed
* Minor changes.
* Fixing cmake issues
* ray instruction
* updating port to new universe
* Fix for env.configure
* redundant commands
* Revert scipy.misc -> cv2 and raise exception for wrong gym version.
|
2017-03-11 00:57:53 -08:00 |
|