fix(ReforceXY): reset last reward shaping on neutral self loop

When the environment is reset, ensure that the last reward shaping value

Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org>

diff --git a/ReforceXY/user_data/freqaimodels/ReforceXY.py b/ReforceXY/user_data/freqaimodels/ReforceXY.py
index dbde6c63e607128e35c698dcc00d54a2f2b97dee..db83ef140cf6c1a45bbeff2a8a498836ed20805f 100644 (file)
--- a/ReforceXY/user_data/freqaimodels/ReforceXY.py
+++ b/ReforceXY/user_data/freqaimodels/ReforceXY.py
@@ -2351,6 +2351,7 @@ class MyRLEnv(Base5ActionRLEnv):
             self._last_next_potential = float(self._last_potential)
             self._last_entry_additive = 0.0
             self._last_exit_additive = 0.0
+            self._last_reward_shaping = 0.0
             return base_reward
 
     def _set_observation_space(self) -> None: