[bug-fix] Use correct agent_ids for demo loader (#3464)

Unity-Technologies · Feb 18, 2020 · 596c364 · 596c364
1 parent a5626aa
commit 596c364
Show file tree

Hide file tree

Showing 2 changed files with 5 additions and 3 deletions.
diff --git a/com.unity.ml-agents/CHANGELOG.md b/com.unity.ml-agents/CHANGELOG.md
@@ -21,6 +21,7 @@ and this project adheres to [Semantic Versioning](http://semver.org/spec/v2.0.0.
 
 ### Bugfixes
 - Fixed an issue which caused self-play training sessions to consume a lot of memory. (#3451)
+- Fixed an IndexError when using GAIL or behavioral cloning with demonstrations recorded with 0.14.0 or later (#3464)
 
 ## [0.14.0-preview] - 2020-02-13
 

diff --git a/ml-agents/mlagents/trainers/demo_loader.py b/ml-agents/mlagents/trainers/demo_loader.py
@@ -51,9 +51,10 @@ def make_demo_buffer(
             previous_action = np.array(
                 pair_infos[idx - 1].action_info.vector_actions, dtype=np.float32
             )
-        agent_id = current_step_info.agent_id[0]
-        current_agent_step_info = current_step_info.get_agent_step_result(agent_id)
-        next_agent_step_info = next_step_info.get_agent_step_result(agent_id)
+        curr_agent_id = current_step_info.agent_id[0]
+        current_agent_step_info = current_step_info.get_agent_step_result(curr_agent_id)
+        next_agent_id = next_step_info.agent_id[0]
+        next_agent_step_info = next_step_info.get_agent_step_result(next_agent_id)
 
         demo_raw_buffer["done"].append(next_agent_step_info.done)
         demo_raw_buffer["rewards"].append(next_agent_step_info.reward)