gryf/coach
1
0
Fork 0
mirror of https://github.com/gryf/coach.git synced 2025-12-17 11:10:20 +01:00
Code Issues Projects Releases Wiki Activity

Commit Graph

  • 9f92064e67 cleanup graph_manager:act Zach Dwiel 2018-10-01 17:01:03 -04:00
  • b5305bd075 update dockerfile Zach Dwiel 2018-09-28 16:46:15 -04:00
  • 950f261201 extract method all_presets Zach Dwiel 2018-09-26 15:59:14 -04:00
  • ed3a3b39be add comments Zach Dwiel 2018-09-26 15:58:15 -04:00
  • 04038c9f40 improve integration test output format Zach Dwiel 2018-09-26 15:54:57 -04:00
  • 1c238b4c60 Added data store backend. (#17) Balaji Subramaniam 2018-10-04 09:45:59 -07:00
  • 6b2de6ba6d Adding initial interface for backend and redis pubsub (#19) Ajay Deshpande 2018-10-03 15:07:48 -07:00
  • a54ef2757f ignore deprecation warnings in test logging Zach Dwiel 2018-09-25 16:15:24 -04:00
  • acc7f70de3 enumerate each preset as its own test Zach Dwiel 2018-09-25 16:14:47 -04:00
  • 1e83a27bee update dockerfile and makefile Zach Dwiel 2018-09-25 16:14:14 -04:00
  • 67faa80ea0 allow custom number of training steps Zach Dwiel 2018-09-21 15:49:06 -04:00
  • d69332efd4 fixed bug in training worker Zach Dwiel 2018-09-20 11:26:17 -04:00
  • cd733b2404 add support for running kubernetes orchestrator from behind proxy Zach Dwiel 2018-09-20 11:25:45 -04:00
  • ad4d2c3053 add make stop_kubernetes Zach Dwiel 2018-09-20 11:25:03 -04:00
  • 5e85a0f972 use the number of heat up steps specified in schedule parameters Zach Dwiel 2018-09-19 13:10:31 -04:00
  • 98850464cc Adding nfs pv, pvc, waiting for memory to be full Ajay Deshpande 2018-09-19 09:03:11 -07:00
  • 13d81f65b9 add redis options to training worker Zach Dwiel 2018-09-19 11:13:20 -04:00
  • 04f32a0f02 add heatup step to training worker Zach Dwiel 2018-09-18 19:55:09 +00:00
  • 7c1f0dce4f include registry in image name Zach Dwiel 2018-09-18 14:36:17 +00:00
  • 0812a94fbd first pass at kubernetes Zach Dwiel 2018-09-17 22:31:17 +00:00
  • 3328b25549 reenable redis; better error message Zach Dwiel 2018-09-17 19:50:03 +00:00
  • 009cf670f3 fix simple typos; temporarily disable redis in rollout worker Zach Dwiel 2018-09-17 18:33:20 +00:00
  • f5b7122d56 weight for checkpoint before trying to start rollout worker Zach Dwiel 2018-09-15 00:55:50 +00:00
  • 4352d6735d add training worker Zach Dwiel 2018-09-15 00:23:16 +00:00
  • 28926bf2a4 Changing parameters Ajay Deshpande 2018-09-14 16:22:13 -07:00
  • c2991819b4 Adding right arguments to the agent Ajay Deshpande 2018-09-14 16:17:34 -07:00
  • ad7f031031 Adding dockerfile Ajay Deshpande 2018-09-14 16:10:00 -07:00
  • ce9838a7d6 Adding kubernetes orchestrator for rollouts, adding requirements for incremental docker builds Ajay Deshpande 2018-09-14 15:58:57 -07:00
  • 6541bc76b9 working checkpoints Zach Dwiel 2018-09-14 20:59:29 +00:00
  • 433bc3e27b standardizing variable access Zach Dwiel 2018-09-13 19:35:28 +00:00
  • e34b9ae9cf allow specifying preset as a commandline parameter to rollout worker Zach Dwiel 2018-09-13 19:35:01 +00:00
  • 3714d8ec80 extract functions display_all_presets_and_exit, expand_preset Zach Dwiel 2018-09-13 19:34:32 +00:00
  • 21f8ca3978 Removing comments and pytests Ajay Deshpande 2018-09-12 20:30:09 -07:00
  • 5a54f67a63 Adding distributed experience replay Ajay Deshpande 2018-09-12 20:21:49 -07:00
  • 747000647f add dockerfile Zach Dwiel 2018-09-12 19:58:26 +00:00
  • bc664c4169 add the first pass of rollout_worker.py Zach Dwiel 2018-09-12 19:53:04 +00:00
  • 61ed6b8ce4 add better defaults to TaskParameters Zach Dwiel 2018-09-12 19:51:40 +00:00
  • 5758c2f23e typo; increased detail in comment Zach Dwiel 2018-09-07 16:06:05 -04:00
  • a1295d16b3 first pass that transition collection interface Zach Dwiel 2018-09-07 15:40:45 -04:00
  • dc77c54ad9 add to gitignore Zach Dwiel 2018-09-07 15:40:08 -04:00
  • 9f1f9e5ab4 replace ExperienceReplay._num_transitions with len(ExperienceReplay.transitions) Zach Dwiel 2018-09-07 14:56:43 -04:00
  • cccfe88f9b remove unused method: update_last_transition_info Zach Dwiel 2018-09-07 14:50:50 -04:00
  • fb21251157 add horizontal scaling document Zach Dwiel 2018-09-07 11:44:28 -04:00
  • 5a8da90d32 bug-fix for dumping movies (+ small refactoring and rename 'VideoDumpMethod -> 'VideoDumpFilter') Gal Leibovich 2018-10-21 17:29:10 +03:00
  • 364168490f checkpointing fix Shadi Endrawis 2018-10-07 20:06:08 +03:00
  • 5c4f9d58dd renamed quick start guide tutorial Gal Novik 2018-10-03 18:15:29 +03:00
  • f7990d4003 trace tests update Shadi Endrawis 2018-10-02 17:55:16 +03:00
  • 51726a5b80 network_imporvements branch merge Shadi Endrawis 2018-10-02 13:41:46 +03:00
  • 72ea933384 bug-fix for clipped_ppo not logging several signals + small cleanup Gal Leibovich 2018-10-02 14:22:37 +03:00
  • 73cc6e39d0 bug fix for clipped ppo for discrete controls itaicaspi-intel 2018-09-18 10:37:42 +03:00
  • abaa58b559 human agent will exit when human control not supported by environment; jupyter notebooks fixes Gal Novik 2018-09-17 15:59:00 +03:00
  • bb76c5c726 CARLA cleanups + calculating the distance to goal itaicaspi-intel 2018-09-16 16:37:04 +03:00
  • 6797824892 bug fixes in the CARLA dataset downloader and extractor itaicaspi-intel 2018-09-16 14:27:22 +03:00
  • 23a9f00e28 fix for human control itaicaspi-intel 2018-09-16 12:43:15 +03:00
  • cf892463e2 updated CARLA to allow using actions of size 3 + automatic downloading of the CARLA imitation dataset itaicaspi-intel 2018-09-16 12:07:11 +03:00
  • d3c8a5d7c1 remove some accidentaly committed files itaicaspi-intel 2018-09-14 18:22:04 +03:00
  • f8d3574b8c updated CARLA to allow the usage of predefined experiment suites itaicaspi-intel 2018-09-14 18:07:24 +03:00
  • e8a2b679d1 using the CoRL2017 experiment suite for CARLA_CIL itaicaspi-intel 2018-09-13 16:59:22 +03:00
  • 06c969951e adding docker and jenkins files itaicaspi-intel 2018-09-04 16:43:52 +03:00
  • d3f97cd93b initial CIL implementation (WIP) itaicaspi-intel 2018-09-13 15:29:29 +03:00
  • 99649c1626 progress bar update itaicaspi-intel 2018-09-13 15:03:24 +03:00
  • 607ef17431 added a simple progress bar implementation itaicaspi-intel 2018-09-13 14:21:38 +03:00
  • fa79d8d365 Carla updates itaicaspi-intel 2018-09-13 11:30:38 +03:00
  • fa4895f840 new traces itaicaspi-intel 2018-09-12 15:29:42 +03:00
  • 673911ff7f very minor cleanup Zach Dwiel 2018-09-11 13:42:50 -04:00
  • a16d724963 removing some of the presets from the trace tests + more robust replay buffer loading itaicaspi-intel 2018-09-12 15:25:13 +03:00
  • 171fe97a3a imitation related bug fixes itaicaspi-intel 2018-09-12 14:54:33 +03:00
  • a9bd1047c4 load and save function for non-episodic replay buffers + carla improvements + network bug fixes itaicaspi-intel 2018-09-06 16:46:57 +03:00
  • d59a700248 updated benchmarks for pong and breakout for dueling ddqn with PER Itai Caspi 2018-09-06 14:05:38 +03:00
  • 08a557bfd1 updated the benchmarks for space invaders with dueling ddqn variants Gal Leibovich 2018-09-06 12:13:38 +03:00
  • 72a1d9d426 Itaicaspi/episode reset refactoring (#105) Itai Caspi 2018-09-04 15:07:54 +03:00
  • 7086492127 parallel trace tests fix Shadi Endrawis 2018-09-03 20:47:10 +03:00
  • 2c62a40466 bug fix in dueling network + revert to TF 1.6 for CPU due to requirements compatibility issues itaicaspi-intel 2018-09-02 13:38:16 +03:00
  • 3a399d1361 Tensorflow 1.10 and python 3.6 (#104) Itai Caspi 2018-09-02 10:12:00 +03:00
  • 5aca3a5ed1 Update README.md Gal Leibovich 2018-08-30 23:33:44 +03:00
  • 55c3034f4d Update README.md Itai Caspi 2018-08-30 23:25:10 +03:00
  • e5526b98f8 Update README.md Itai Caspi 2018-08-30 22:58:37 +03:00
  • d862a3be83 rainbow dqn hyper-parameter updates Gal Leibovich 2018-08-30 20:41:33 +03:00
  • 07db625987 Running trace tests in parallel + other small fixes Shadi Endrawis 2018-08-30 19:34:52 +03:00
  • ebe574e463 add missing hidden layer in rainbow_q_head Gal Leibovich 2018-08-30 19:34:27 +03:00
  • ea294de7fd adding dueling support for rainbow dqn (now only missing n-step) Gal Leibovich 2018-08-30 18:15:59 +03:00
  • d2623c0eee bug-fix in dueling dqn Gal Leibovich 2018-08-30 18:02:20 +03:00
  • bbe7ac3338 Rainbow DQN agent (WIP - still missing dueling and n-step) + adding support for Prioritized ER for C51 Gal Leibovich 2018-08-30 15:11:51 +03:00
  • fd2f4b0852 bug fix in HRL HER memory + some small improvements itaicaspi-intel 2018-08-29 14:36:18 +03:00
  • 1aa2ab0590 parameter noise exploration - using Noisy Nets Gal Leibovich 2018-08-27 18:19:01 +03:00
  • 658b437079 removing datasets + imports optimization itaicaspi-intel 2018-08-19 14:16:01 +03:00
  • d826382b11 removing test from Doom_Health_Supreme_DFP + relaxing time limit on reward tests Gal Leibovich 2018-08-26 18:42:26 +03:00
  • 2021490caa small adjustment to golden tests + fixes for Doom_Health_DFP and Doom_Health_Supreme_DFP Gal Leibovich 2018-08-23 15:59:00 +03:00
  • 3fd0bf4f0f Update README.md Itai Caspi 2018-08-26 12:09:46 +03:00
  • 9bb7bd2e9c bug-fix in local_batch_run_coach and rename to run_multiple_seeds Gal Leibovich 2018-08-23 14:38:42 +03:00
  • a4471389a4 brightened starcraft.gif Gal Leibovich 2018-08-20 13:50:09 +03:00
  • 904570000a Update README.md Gal Leibovich 2018-08-20 12:04:29 +03:00
  • 5e275e9795 update starcraft gif Gal Leibovich 2018-08-20 10:58:16 +03:00
  • 3abb6cd415 Trace tests update Shadi Endrawis 2018-08-20 13:01:17 +03:00
  • c1f428666e bug-fix for checkpointing for single-worker algorithms Gal Leibovich 2018-08-19 20:17:15 +03:00
  • 9f599f38cf Update README.md Itai Caspi 2018-08-19 13:09:06 +03:00
  • c5165cd7d6 benchmarks and pip package updates Itai Caspi 2018-08-19 14:23:20 +03:00
  • 23d2945bf8 Update README.md Gal Leibovich 2018-08-19 11:02:45 +03:00
  • 0be4a42701 updates needed for the pip package Itai Caspi 2018-08-19 10:39:03 +03:00
  • e2e8143b94 additional benchmarks for dqn and a3c Itai Caspi 2018-08-18 15:21:50 +03:00