move RLBootstrapDataset into a own class with its own test file

snakes/BestBattleSnake.py

@@ -2,14 +2,15 @@ from collections.abc import Iterator
 from collections import deque
 from typing import Any, cast
 from time import perf_counter
-from pathlib import Path
-import random, json, os
+import random, os
+
+from server.dataset.RLBootstrapDataset import RLBootstrapDataset
 
 from snakes.TemplateSnake import TemplateSnake
 from server.GameBoard import GameBoard
 
 class BestBattleSnake(TemplateSnake):
-  VERSION = "2.6.0"
+  VERSION = "2.6.1"
   Point = tuple[int, int]
   Coord = dict[str, int]
   SnakeState = dict[str, Any]
@@ -40,11 +41,7 @@ class BestBattleSnake(TemplateSnake):
     self.previous_hazards = set()
     self.duel_style = self._get_duel_style()
     self.timeout_buffer_ms = self._get_timeout_buffer_ms()
-    self.rl_bootstrap_enabled = self._env_bool("RL_BOOTSTRAP_ENABLED", default=False)
-    self.rl_min_base_rows = self._env_int("RL_MIN_BASE_ROWS", default=5000)
-    self.rl_base_dataset_path = Path(os.getenv("RL_BASE_DATASET", "data/dataset/best_moves.jsonl"))
-    self.rl_bootstrap_path = Path(os.getenv("RL_BOOTSTRAP_OUTPUT", "data/dataset/rl_bootstrap.jsonl"))
-    self.rl_needs_more_data = False
+    self.rl_bootstrap = RLBootstrapDataset()
     self.future_planning_depth = max(1, min(4, self._env_int("BATTLE_FUTURE_PLANNING_DEPTH", default=2)))
     self.future_planning_branch = max(1, min(3, self._env_int("BATTLE_FUTURE_PLANNING_BRANCH", default=2)))
     self.future_planning_min_time_ms = max(25, self._env_int("BATTLE_FUTURE_PLANNING_MIN_MS", default=70))
@@ -88,12 +85,6 @@ class BestBattleSnake(TemplateSnake):
     except ValueError:
       return 120
 
-  def _env_bool(self, name:str, default:bool=False) -> bool:
-    value = os.getenv(name)
-    if value is None:
-      return default
-    return value.lower() in {"1", "true", "yes", "on"}
-
   def _env_int(self, name:str, default:int) -> int:
     value = os.getenv(name)
     if value is None:
@@ -103,46 +94,6 @@ class BestBattleSnake(TemplateSnake):
     except ValueError:
       return default
 
-  def _count_jsonl_rows(self, path:Path) -> int:
-    if not path.exists() or not path.is_file():
-      return 0
-    count = 0
-    with path.open("r", encoding="utf-8") as handle:
-      for line in handle:
-        if line.strip():
-          count += 1
-    return count
-
-  def _refresh_rl_bootstrap_state(self):
-    if not self.rl_bootstrap_enabled:
-      self.rl_needs_more_data = False
-      return
-    base_rows = self._count_jsonl_rows(self.rl_base_dataset_path)
-    self.rl_needs_more_data = base_rows < self.rl_min_base_rows
-
-  def _record_rl_bootstrap_sample(self, game_data:GameBoard, move:str, safe_moves:MoveMap, reason:str, scores:dict[str, float]|None=None):
-    if not self.rl_bootstrap_enabled or not self.rl_needs_more_data:
-      return
-
-    try:
-      self.rl_bootstrap_path.parent.mkdir(parents=True, exist_ok=True)
-      row = {
-        "source": "best_battlesnake_bootstrap",
-        "game_id": getattr(game_data, "id", None),
-        "turn": game_data.get_turn(),
-        "move": move,
-        "safe_moves": list(safe_moves.keys()),
-        "reason": reason,
-        "game_board": game_data.get_game_board_as_dict(),
-      }
-      if scores:
-        row["scores"] = {k: round(v, 5) for k, v in scores.items()}
-
-      with self.rl_bootstrap_path.open("a", encoding="utf-8") as handle:
-        handle.write(json.dumps(row, ensure_ascii=False) + "\n")
-    except Exception:
-      return
-
   def choose_move(self, game_data:GameBoard) -> str:
     """Pick the next move from a Battlesnake move request.
 
@@ -162,7 +113,7 @@ class BestBattleSnake(TemplateSnake):
       self.last_move = None
       self.previous_hazards = set()
       self.last_game_id = game_id
-      self._refresh_rl_bootstrap_state()
+      self.rl_bootstrap.refresh_state()
 
     my_snake = cast(dict[str, Any], game_data.get_my_snake())
     my_head = my_snake["head"]
@@ -215,7 +166,7 @@ class BestBattleSnake(TemplateSnake):
           "reason": "no_safe_moves",
         }
       )
-      self._record_rl_bootstrap_sample(game_data, fallback, safe_moves, "no_safe_moves")
+      self.rl_bootstrap.record_sample(game_data, fallback, safe_moves, "no_safe_moves")
       self.previous_hazards = set(hazard_set)
       return fallback
 
@@ -246,7 +197,7 @@ class BestBattleSnake(TemplateSnake):
       self.recent_heads.append(current_head_point)
       self.last_move = best_move
       self.add_to_history({"turn": turn, "move": best_move, "scores": scores})
-      self._record_rl_bootstrap_sample(game_data, best_move, safe_moves, "constrictor", scores)
+      self.rl_bootstrap.record_sample(game_data, best_move, safe_moves, "constrictor", scores)
       self.previous_hazards = set(hazard_set)
       return best_move
 
@@ -270,7 +221,7 @@ class BestBattleSnake(TemplateSnake):
       self.recent_heads.append(current_head_point)
       self.last_move = best_move
       self.add_to_history({"turn": turn, "move": best_move, "scores": scores})
-      self._record_rl_bootstrap_sample(game_data, best_move, safe_moves, "duel", scores)
+      self.rl_bootstrap.record_sample(game_data, best_move, safe_moves, "duel", scores)
       self.previous_hazards = set(hazard_set)
       return best_move
 
@@ -418,7 +369,7 @@ class BestBattleSnake(TemplateSnake):
       self.recent_heads.append(current_head_point)
       self.last_move = quick_move
       self.add_to_history({"turn": turn, "move": quick_move, "reason": "timeout_budget"})
-      self._record_rl_bootstrap_sample(game_data, quick_move, safe_moves, "timeout_budget")
+      self.rl_bootstrap.record_sample(game_data, quick_move, safe_moves, "timeout_budget")
       self.previous_hazards = set(hazard_set)
       return quick_move
 
@@ -462,7 +413,7 @@ class BestBattleSnake(TemplateSnake):
     self.recent_heads.append(current_head_point)
     self.last_move = best_move
     self.add_to_history({"turn": turn, "move": best_move, "scores": scores})
-    self._record_rl_bootstrap_sample(game_data, best_move, safe_moves, "multi", scores)
+    self.rl_bootstrap.record_sample(game_data, best_move, safe_moves, "multi", scores)
     self.previous_hazards = set(hazard_set)
     return best_move