V0.1

.github/workflows/unittests.yml → .github/workflows/test.yml

@@ -1,5 +1,5 @@
 
-name: unittest
+name: test
 
 on:
 #   workflow_dispatch:

maro/rl/shaping/k_step_experience_shaper.py

@@ -58,4 +58,4 @@ def __call__(self, trajectory, snapshot_list):
             agent_exp[KStepExperienceKeys.NEXT_ACTION.value].appendleft(next_transition["action"])
             agent_exp[KStepExperienceKeys.DISCOUNT.value].appendleft(self._reward_decay ** (min(self._steps, len(trajectory)-1-i)))
 
-        return experiences
+        return dict(experiences)