#1859 - Started giving the red agent some 'intelligence' and a sense of a state. Changed Application.run to .execute.

src/primaite/game/agent/GATE_agents.py

@@ -19,10 +19,10 @@ class GATERLAgent(AbstractGATEAgent):
 
     def __init__(
         self,
-        agent_name: str | None,
-        action_space: ActionManager | None,
-        observation_space: ObservationSpace | None,
-        reward_function: RewardFunction | None,
+        agent_name: Optional[str],
+        action_space: Optional[ActionManager],
+        observation_space: Optional[ObservationSpace],
+        reward_function: Optional[RewardFunction],
     ) -> None:
         super().__init__(agent_name, action_space, observation_space, reward_function)
         self.most_recent_action: ActType

src/primaite/game/agent/interface.py

@@ -109,6 +109,8 @@ class RandomAgent(AbstractScriptedAgent):
         """
         return self.action_space.get_action(self.action_space.space.sample())
 
+class DataManipulationAgent(AbstractScriptedAgent):
+    pass
 
 class AbstractGATEAgent(AbstractAgent):
     """Base class for actors controlled via external messages, such as RL policies."""

src/primaite/game/science.py

@@ -0,0 +1,16 @@
+from random import random
+
+
+def simulate_trial(p_of_success: float):
+    """
+    Simulates the outcome of a single trial in a Bernoulli process.
+
+    This function returns True with a probability 'p_of_success', simulating a success outcome in a single
+    trial of a Bernoulli process. When this function is executed multiple times, the set of outcomes follows
+    a binomial distribution. This is useful in scenarios where one needs to model or simulate events that
+    have two possible outcomes (success or failure) with a fixed probability of success.
+
+    :param p_of_success: The probability of success in a single trial, ranging from 0 to 1.
+    :returns: True if the trial is successful (with probability 'p_of_success'); otherwise, False.
+    """
+    return random() < p_of_success

src/primaite/game/session.py

@@ -60,7 +60,7 @@ class PrimaiteGATEClient(GATEClient):
         return self.parent_session.training_options.rl_algorithm
 
     @property
-    def seed(self) -> int | None:
+    def seed(self) -> Optional[int]:
         """The seed to use for the environment's random number generator."""
         return self.parent_session.training_options.seed
 
@@ -115,7 +115,7 @@ class PrimaiteGATEClient(GATEClient):
         info = {}
         return obs, rew, term, trunc, info
 
-    def reset(self, *, seed: int | None = None, options: dict[str, Any] | None = None) -> Tuple[ObsType, Dict]:
+    def reset(self, *, seed: Optional[int] = None, options: Optional[Dict[str, Any]] = None) -> Tuple[ObsType, Dict]:
         """Reset the environment.
 
         This method is called when the environment is initialized and at the end of each episode.