#917 - started working on the Agent abstract classes and sub-classes

src/primaite/agents/agent_abc.py

@@ -1,14 +1,20 @@
 from abc import ABC, abstractmethod
-from typing import Optional
+from typing import Optional, Final, Dict, Any
 
+from primaite import getLogger
+from primaite.config.training_config import TrainingConfig
 from primaite.environment.primaite_env import Primaite
 
+_LOGGER = getLogger(__name__)
+
 
 class AgentABC(ABC):
 
     @abstractmethod
     def __init__(self, env: Primaite):
         self._env: Primaite = env
+        self._training_config: Final[TrainingConfig] = self._env.training_config
+        self._lay_down_config: Dict[str, Any] = self._env.lay_down_config
         self._agent = None
 
     @abstractmethod
@@ -33,4 +39,35 @@ class AgentABC(ABC):
 
     @abstractmethod
     def export(self):
-        pass
+        pass
+
+
+class DeterministicAgentABC(AgentABC):
+    @abstractmethod
+    def __init__(self, env: Primaite):
+        self._env: Primaite = env
+        self._agent = None
+
+    @abstractmethod
+    def _setup(self):
+        pass
+
+    def learn(self, time_steps: Optional[int], episodes: Optional[int]):
+        pass
+        _LOGGER.warning("Deterministic agents cannot learn")
+
+    @abstractmethod
+    def evaluate(self, time_steps: Optional[int], episodes: Optional[int]):
+        pass
+
+    @abstractmethod
+    def load(self):
+        pass
+
+    @abstractmethod
+    def save(self):
+        pass
+
+    @abstractmethod
+    def export(self):
+        pass

src/primaite/agents/sb3.py

@@ -1,28 +1,35 @@
-# from typing import Optional
-#
-# from primaite.agents.agent_abc import AgentABC
-# from primaite.environment.primaite_env import Primaite
-#
-#
-# class SB3PPO(AgentABC):
-#     def __init__(self, env: Primaite):
-#         super().__init__(env)
-#
-#     def _setup(self):
-#         if self._env.training_config
-#         pass
-#
-#     def learn(self, time_steps: Optional[int], episodes: Optional[int]):
-#         pass
-#
-#     def evaluate(self, time_steps: Optional[int], episodes: Optional[int]):
-#         pass
-#
-#     def load(self):
-#         pass
-#
-#     def save(self):
-#         pass
-#
-#     def export(self):
-#         pass
+from typing import Optional
+
+from stable_baselines3 import PPO
+
+from primaite.agents.agent_abc import AgentABC
+from primaite.environment.primaite_env import Primaite
+from stable_baselines3.ppo import MlpPolicy as PPOMlp
+
+class SB3PPO(AgentABC):
+    def __init__(self, env: Primaite):
+        super().__init__(env)
+
+    def _setup(self):
+        self._agent = PPO(
+            PPOMlp,
+            self._env,
+            verbose=0,
+            n_steps=self._training_config.num_steps
+        )
+
+
+    def learn(self, time_steps: Optional[int], episodes: Optional[int]):
+        pass
+
+    def evaluate(self, time_steps: Optional[int], episodes: Optional[int]):
+        pass
+
+    def load(self):
+        pass
+
+    def save(self):
+        pass
+
+    def export(self):
+        pass