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

create per environment Dockerfiles. (#70)

Browse Source 
* create per environment Dockerfiles.

Adjust CI setup to better parallelize runs.
Fix a couple of issues in golden and trace tests.
Update a few of the docs.

* bugfix in mmc agent.

Also install kubectl for CI, update badge branch.

* remove integration test parallelism.
This commit is contained in:
Scott Leishman
2018-11-14 07:40:22 -08:00
committed by GitHub
parent a849c17e46
commit 524f8436a2
20 changed files with 448 additions and 139 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
rl_coach/agents/mmc_agent.py
View File

@@ -64,7 +64,7 @@ class MixedMonteCarloAgent(ValueOptimizationAgent):
one_step_target = batch.rewards()[i] + \
(1.0 - batch.game_overs()[i]) * self.ap.algorithm.discount * \
q_st_plus_1[i][selected_actions[i]]
monte_carlo_target = total_returns()[i]
monte_carlo_target = total_returns[i]
TD_targets[i, batch.actions()[i]] = (1 - self.mixing_rate) * one_step_target + \
self.mixing_rate * monte_carlo_target