Ajay Deshpande
33dc29ee99
Uploading checkpoint if crd provided ( #191 )
...
* Uploading checkpoint if crd provided
* Changing the calculation of total steps because of a recent change in core_types
Fixes #195
2019-04-26 12:27:33 -07:00
Ajay Deshpande
2c1a9dbf20
Adding framework for multinode tests ( #149 )
...
* Currently runs CartPole_ClippedPPO and Mujoco_ClippedPPO with inverted_pendulum level.
2019-02-26 13:53:12 -08:00
Gal Novik
fc6604c09c
added missing license headers
2018-11-27 22:43:40 +02:00
Sina Afrooze
5332013bd1
Implement frame-work agnostic rollout and training workers ( #137 )
...
* Added checkpoint state file to coach checkpointing.
* Removed TF specific code from rollout_worker, training_worker, and s3_data_store
2018-11-23 18:05:44 -08:00
Ajay Deshpande
4a6c404070
Adding worker logs and plumbed task_parameters to distributed coach ( #130 )
2018-11-23 15:35:11 -08:00
Balaji Subramaniam
101c55d37d
Handle both Environment Steps and Episodes on the subscriber side. ( #99 )
2018-11-15 14:42:21 -08:00
Ajay Deshpande
fde73ced13
Simulating the act on the trainer. ( #65 )
...
* Remove the use of daemon threads for Redis subscribe.
* Emulate act and observe on trainer side to update internal vars.
2018-11-15 08:38:58 -08:00
Ajay Deshpande
875d6ef017
Adding target reward and target sucess ( #58 )
...
* Adding target reward
* Adding target successs
* Addressing comments
* Using custom_reward_threshold and target_success_rate
* Adding exit message
* Moving success rate to environment
* Making target_success_rate optional
2018-11-12 15:03:43 -08:00
Balaji Subramaniam
7e7006305a
Integrate coach.py params with distributed Coach. ( #42 )
...
* Integrate coach.py params with distributed Coach.
* Minor improvements
- Use enums instead of constants.
- Reduce code duplication.
- Ask experiment name with timeout.
2018-11-05 09:33:30 -08:00
zach dwiel
3e5e5475de
update training worker
2018-10-23 19:58:17 -04:00
Zach Dwiel
9804b033a2
rename save_checkpoint_dir -> checkpoint_save_dir
2018-10-23 17:10:58 -04:00
Zach Dwiel
7382a142bb
remove unused steps parameter from GraphManager.train
2018-10-23 16:57:06 -04:00
Balaji Subramaniam
ca9015d8b1
Make NFS work end-to-end.
2018-10-23 16:55:37 -04:00
Ajay Deshpande
fb1039fcb5
Checkpoint and evaluation optimizations
2018-10-23 16:55:37 -04:00
Ajay Deshpande
b285a02023
Adding parameteres, checking transitions before training
2018-10-23 16:55:37 -04:00
Ajay Deshpande
0f46877d7e
Adding steps and waiting for new checkpoint
2018-10-23 16:55:37 -04:00
Ajay Deshpande
0e121c5762
Ignoring redis sub if testing
2018-10-23 16:55:37 -04:00
Ajay Deshpande
a7f5442015
Adding should_train helper and should_train in graph_manager
2018-10-23 16:54:43 -04:00
Balaji Subramaniam
844a5af831
Make distributed coach work end-to-end.
...
- With data store, memory backend and orchestrator interfaces.
2018-10-23 16:54:43 -04:00
Ajay Deshpande
6b2de6ba6d
Adding initial interface for backend and redis pubsub ( #19 )
...
* Adding initial interface for backend and redis pubsub
* Addressing comments, adding super in all memories
* Removing distributed experience replay
2018-10-23 16:51:48 -04:00
Zach Dwiel
67faa80ea0
allow custom number of training steps
2018-10-23 16:51:48 -04:00
Zach Dwiel
d69332efd4
fixed bug in training worker
2018-10-23 16:51:48 -04:00
Zach Dwiel
5e85a0f972
use the number of heat up steps specified in schedule parameters
2018-10-23 16:51:48 -04:00
Ajay Deshpande
98850464cc
Adding nfs pv, pvc, waiting for memory to be full
2018-10-23 16:50:48 -04:00
Zach Dwiel
13d81f65b9
add redis options to training worker
2018-10-23 16:47:46 -04:00
Zach Dwiel
04f32a0f02
add heatup step to training worker
2018-10-23 16:47:46 -04:00
Zach Dwiel
4352d6735d
add training worker
2018-10-23 16:47:46 -04:00
