graph-games-proto 0.3.2225__py3-none-any.whl → 0.3.2274__py3-none-any.whl

graph_games_proto/__init__.py

@@ -1,3 +1,3 @@
 # __init__.py
 from .main import hello
-from .fns import getpublicgameconfig, alpha0, isterminal, getfinalscores, get_deadlines, get_longest_path_length, get_max_allotted_times, get_legal_actions_for_path, find_player_with_longest_path, calc_player_graph, get_edges, FrozenDek, QValueLearningPolicy, Action2, getnextstate2, isactionlegal2, LegalAction, Fig, RandoPolicy, StaticBoardConfig, autoplay, getpublicstate, generate_cards, PublicState, State, Fig, getprivatescore, get_qvalue_trajectories, PlayerState, initfig, initboardconfig, gettoplay, printstate, getinitialstate, Card, PrivateState, getprivatestate, printaction, json_serializer, getrng, FrozenBoardConfig, initgameconfig, GameConfig
+from .fns import getqproxy0, get_intuited_best_actions, getpublicgameconfig, alpha0, isterminal, getfinalscores, get_deadlines, get_longest_path_length, get_max_allotted_times, get_legal_actions_for_path, find_player_with_longest_path, calc_player_graph, get_edges, FrozenDek, QValueLearningPolicy, Action2, getnextstate2, isactionlegal2, LegalAction, Fig, RandoPolicy, StaticBoardConfig, autoplay, getpublicstate, generate_cards, PublicState, State, Fig, getprivatescore, get_qvalue_trajectories, PlayerState, initfig, initboardconfig, gettoplay, printstate, getinitialstate, Card, PrivateState, getprivatestate, printaction, json_serializer, getrng, FrozenBoardConfig, initgameconfig, GameConfig

graph_games_proto/fns.py

@@ -477,15 +477,14 @@ class FrozenLinkPath(PClass):
         )
 
 
-@dataclass(frozen=True)
-class FrozenPath:
-    num: int
-    link_num: int
-    start_point_uuid: str
-    end_point_uuid: str
-    start_point_num: int
-    end_point_num: int
-    path: FrozenLinkPath
+class FrozenPath(PClass):
+    num = field(type=int)
+    link_num = field(type=int)
+    start_point_uuid = field(type=str)
+    end_point_uuid = field(type=str)
+    start_point_num = field(type=int)
+    end_point_num = field(type=int)
+    path = field(type=FrozenLinkPath)
     def __todict__(self):
         return {
             "num": self.num,
@@ -549,17 +548,6 @@ class FrozenDeckUnit2(PClass):
         )
 
 
-@dataclass(frozen=True)
-class FrozenBoardConfigDataclass:
-    deck_units: List[FrozenDeckUnit2]
-    len_scores: List[FrozenLenScore]
-    links: List[FrozenLink2]
-    clusters: List[FrozenCluster]
-    points: List[FrozenPoint2]
-    board_paths: List[FrozenPath]
-    settings: List[FrozenSetting]
-
-
 class FrozenRoute(PClass):
     num = field(type=int)
     uuid = field(type=str)
@@ -1019,11 +1007,6 @@ def getpublicgameconfig(game_config):
     )
 
 
-class ActionDrawUnit:
-   def __init__(self):
-       pass
-
-
 class PublicPlayer(PClass):
     idx = field(type=int)
     pieces = field(type=list)  # List[Piece]
@@ -1053,7 +1036,7 @@ class Player(PClass):
     idx = field(type=int)
     pieces = field(type=list)  # List[Piece]
     cards = field(type=list)  # List[Card]
-    discard_tray = field(type=list)  # List[Card]
+    discard_tray = field(type=PVector)  # PVector[Card]
     def __todict__(self):
         return {
             "idx": self.idx,
@@ -1164,7 +1147,7 @@ class PrivateState(PClass):
 
 
 class FaceupCardStack(PClass):
-    cards = field(type=list)  # List[Card]
+    cards = field(type=list)  # List[str]
     def __todict__(self):
         return {
             "cards": self.cards,
@@ -1177,7 +1160,7 @@ class FaceupCardStack(PClass):
 
 
 class FacedownCardStack(PClass):
-    cards = field(type=list)  # List[Card]
+    cards = field(type=list)  # List[str]
     def __todict__(self):
         return {
             "cards": self.cards,
@@ -1628,15 +1611,15 @@ class LegalAction(PClass):
 
 
 class PublicActionDiscard(PClass):
-    deck_idxs = field(type=list)  # List[int]
+    discard_tray_idxs = field(type=list)  # List[int]
     def __todict__(self):
         return {
-            "deck_idxs": self.deck_idxs,
+            "discard_tray_idxs": self.discard_tray_idxs,
         }
     @staticmethod
     def __fromdict__(d):
         return PublicActionDiscard(
-            deck_idxs=d["deck_idxs"]
+            discard_tray_idxs=d["discard_tray_idxs"]
         )
 
 
@@ -1644,7 +1627,7 @@ class ActionDiscard(PClass):
     discard_tray_idxs = field(type=list)  # List[int]
     def get_public(self, state):
         return PublicActionDiscard(
-            deck_idxs=[0 for card_idx in self.discard_tray_idxs] # TODO: fix this
+            discard_tray_idxs=self.discard_tray_idxs,
         )
     def __todict__(self):
         return {
@@ -1658,15 +1641,15 @@ class ActionDiscard(PClass):
 
 
 class PublicActionKeep(PClass):
-    deck_idxs = field(type=list)  # List[int]
+    discard_tray_idxs = field(type=list)  # List[int]
     def __todict__(self):
         return {
-            "deck_idxs": self.deck_idxs,
+            "discard_tray_idxs": self.discard_tray_idxs,
         }
     @staticmethod
     def __fromdict__(d):
         return PublicActionKeep(
-            deck_idxs=d["deck_idxs"]
+            discard_tray_idxs=d["discard_tray_idxs"]
         )
 
 
@@ -1674,7 +1657,7 @@ class ActionKeep(PClass):
     discard_tray_idxs = field(type=list)  # List[int]
     def get_public(self, state):
         return PublicActionKeep(
-            deck_idxs=[0 for card_idx in self.discard_tray_idxs] # TODO: fix this
+            discard_tray_idxs=self.discard_tray_idxs,
         )
     def __todict__(self):
         return {
@@ -2322,7 +2305,7 @@ def init_state_kernel(game_config, **kwargs):
         starting_decks.append(deck_obj)
 
     default_players = [
-        Player(idx=idx, pieces=[], cards=[], discard_tray=[]) for idx in range(game_config.num_players)
+        Player(idx=idx, pieces=[], cards=[], discard_tray=v()) for idx in range(game_config.num_players)
     ]
 
     return StateKernel(
@@ -2419,9 +2402,8 @@ class QValueLearningPolicy(PClass):
     epsilon = field(type=float, initial=0.1)  # Epsilon for exploration
 
 
-class RandoPolicy:
-   def __init__(self):
-       pass
+class RandoPolicy(PClass):
+    pass
 
 
 # Functions  
@@ -3853,7 +3835,7 @@ def handle_keep_action(kernel, action):
     # Clear the discard tray and add the kept cards back
     player.cards.extend([c for c in kept_cards if c is not None])
 
-    player.discard_tray.clear()
+    player.set(discard_tray=v())
     return kernel.set(decks=kernel.decks, players=kernel.players)
 
 
@@ -3900,7 +3882,7 @@ def handle_discard_action(kernel, action):
     # Clear the discard tray and add the kept cards back
     player.cards.extend([c for c in kept_cards if c is not None])
 
-    player.discard_tray.clear()
+    player.set(discard_tray=v())
     print("****************************** handle_discard_action 10 kept_cards: ", kept_cards)
 
     return kernel.set(decks=kernel.decks, players=kernel.players)
@@ -4595,6 +4577,11 @@ def get_next_player_shuffled_idx(state_kernel, last_action):
 def getnextstate2(s, a, log=False):
     is_legal, reason = isactionlegal2(s, a)
     if not is_legal:
+        print("****************************** Action is not legal: ", json.dumps(a.__todict__(), indent=2))
+        # print("a")
+        # print("json.dumps(s.legal_actions_3): ", json.dumps([la.__todict__() for la in s.legal_actions_3], indent=2))
+        # print("b")
+        # print("\n\n")
         raise ValueError(f"Action is not legal: {a}. Reason: {reason}")
     kernel = s.kernel
     kernel = run_state_action_hooks(kernel, a, AFTER_ACCEPT_ACTION_HOOKS, log)
@@ -4734,7 +4721,7 @@ def get_candidate_decks(ps: PlayerState) -> List[CandidateDeck]:
         spots_to_remove = [card_uuid for card_uuid in public_deck.faceup_spread.spots if card_uuid]
         for card_uuid in spots_to_remove:
             if card_uuid:
-                print("************************************ spots_to_remove: ", spots_to_remove)
+                # print("************************************ spots_to_remove: ", spots_to_remove)
                 candidates.remove(card_uuid)
 
         note_lists[deck_idx].append(f"Removed {len(spots_to_remove)} candidates from deck {public_deck.idx} for faceup spots")
@@ -4746,6 +4733,7 @@ def get_candidate_decks(ps: PlayerState) -> List[CandidateDeck]:
         note_lists[deck_idx].append(f"Removed {len(public_deck.discard_faceup_stack.cards)} candidates from deck {public_deck.idx} for faceup discard")
 
 
+    # print("ps.private.player.cards: ", len(ps.private.player.cards))
     for card_uuid in ps.private.player.cards:
         card = ps.game_config.fig.carduuid2card[card_uuid]
         candidate_lists[card.deck_idx].remove(card_uuid)
@@ -4762,25 +4750,25 @@ def get_candidate_decks(ps: PlayerState) -> List[CandidateDeck]:
 
 
 def imagine_player(p_idx: int, ps: PlayerState, candidate_decks: List[CandidateDeck]) -> Player:
-    print(f"imagine_player {p_idx}")
+    # print(f"imagine_player {p_idx}")
     if ps.private.player.idx == p_idx:
-        print(f"No need to imagine player {p_idx}, returning private player")
+        # print(f"No need to imagine player {p_idx}, returning private player")
         return ps.private.player
-    print(f"len(candidate_decks[0].candidates): ", len(candidate_decks[0].candidates))
-    print(f"public_player.deck_counts: ", ps.public.players[p_idx].deck_counts)
+    # print(f"len(candidate_decks[0].candidates): ", len(candidate_decks[0].candidates))
+    # print(f"public_player.deck_counts: ", ps.public.players[p_idx].deck_counts)
     public_player = ps.public.players[p_idx]
-    print(f"z1")
+    # print(f"z1")
     cards = []
-    print(f"z2")
+    # print(f"z2")
     for deck_idx, count in enumerate(public_player.deck_counts):
-        print(f"deck_idx: {deck_idx}, count: {count}")
+        # print(f"deck_idx: {deck_idx}, count: {count}")
         candidate_deck = candidate_decks[deck_idx]
-        print(f"len(candidate_deck.candidates) before: {len(candidate_deck.candidates)}")
+        # print(f"len(candidate_deck.candidates) before: {len(candidate_deck.candidates)}")
         for _ in range(count):
             cards.append(candidate_deck.candidates.pop())
-        print(f"len(candidate_deck.candidates) after: {len(candidate_deck.candidates)}")
+        # print(f"len(candidate_deck.candidates) after: {len(candidate_deck.candidates)}")
     discard_tray = []
-    print(f"z4")
+    # print(f"z4")
     for deck_idx, count in enumerate(public_player.discard_deck_counts):
         candidate_deck = candidate_decks[deck_idx]
         for _ in range(count):
@@ -4815,7 +4803,7 @@ def imagine_deck(public_deck: PublicDeck, candidate_deck: CandidateDeck) -> Deck
     facedown_stack_cards = [
         candidate_deck.candidates.pop() for _ in range(public_deck.facedown_stack.num_cards)
     ]
-    print(f"len(facedown_stack_cards) pops for deck_idx [{public_deck.idx}]: ", len(facedown_stack_cards))
+    # print(f"len(facedown_stack_cards) pops for deck_idx [{public_deck.idx}]: ", len(facedown_stack_cards))
 
     imagined = Deck(
         idx=public_deck.idx,
@@ -4871,25 +4859,25 @@ def imagine_state(ps: PlayerState) -> State:
 
     candidate_decks = get_candidate_decks(ps)
 
-    print_public_deck_stats(ps.public.decks[0])
+    # print_public_deck_stats(ps.public.decks[0])
     
-    print("\n")
-    print("************************************ candidate_decks[1].notes a: ", candidate_decks[1].notes)
-    print("************************************ len(candidate_decks[1].candidates) a: ", len(candidate_decks[1].candidates))
-    print("\n")
+    # print("\n")
+    # print("************************************ candidate_decks[1].notes a: ", "\n".join(candidate_decks[1].notes))
+    # print("************************************ len(candidate_decks[1].candidates) a: ", len(candidate_decks[1].candidates))
+    # print("\n")
     
     imagined_decks = imagine_decks(ps, candidate_decks)
 
-    print("\n")
-    print("************************************ len(candidate_decks[1].candidates) b: ", len(candidate_decks[1].candidates))
-    print("player deck[1] counts: ", [public_player.deck_counts[1] for public_player in ps.public.players])
-    print("\n")
+    # print("\n")
+    # print("************************************ len(candidate_decks[1].candidates) b: ", len(candidate_decks[1].candidates))
+    # print("player deck[1] counts: ", [public_player.deck_counts[1] for public_player in ps.public.players])
+    # print("\n")
 
     imagined_players = imagine_players(ps, candidate_decks)
     
-    print("\n")
-    print("************************************ len(candidate_decks[1].candidates) c: ", len(candidate_decks[1].candidates))
-    print("\n")
+    # print("\n")
+    # print("************************************ len(candidate_decks[1].candidates) c: ", len(candidate_decks[1].candidates))
+    # print("\n")
 
     imagined_kernel = init_state_kernel(
         GameConfig(
@@ -4991,7 +4979,7 @@ def getpublicdeck(s, d):
     return PublicDeck(
         idx=d.idx,
         uuid=d.uuid,
-        all_cards=d.all_cards,
+        all_cards=copy.deepcopy(d.all_cards),
         faceup_stack = copy.deepcopy(d.faceup_stack) if d.faceup_stack else None,
         faceup_spread = copy.deepcopy(d.faceup_spread) if d.faceup_spread else None,
         facedown_stack = PublicFacedownCardStack(num_cards=len(d.facedown_stack.cards)) if d.facedown_stack else None,
@@ -5551,10 +5539,6 @@ def getscorecodes(f):
     return score_codes
 
 
-# Implementing the following Julia function:
-# function getvalidspotnums(s::State)
-#     filter(n -> !isnothing(s.faceup_spots[n]), 1:length(s.faceup_spots))
-# end
 def getvalidspotnums(s):
     return [n for n in range(1, len(s.faceup_spread.spots) + 1) if s.faceup_spread.spots[n-1] is not None]
 
@@ -5782,7 +5766,7 @@ def get_intuited_best_actions(ps: PlayerState):
     ]
     if not possible_actions:
         return None
-    return possible_actions[:8]
+    return possible_actions[:16]
 
 
 def get_spread(q_values, p_idx):
@@ -5810,9 +5794,8 @@ def dynamics(s, a):
     return next_s, rewards
 
 
-def alpha0(ps: PlayerState):
+def alpha0(ps: PlayerState, td=3):
     # print("************************************ ps.public.decks[0].facedown_stack 8: ", ps.public.decks[0].facedown_stack)
-    td = 3
     legal_actions = ps.private.legal_actions_3
     if not legal_actions:
         return None
@@ -5830,7 +5813,7 @@ def get_max_spread_idx(q_proxies, p_idx):
 
 @dispatch(PlayerState, Action2, int)
 def getqproxy0(ps: PlayerState, a: LegalAction, td: int):
-    print("~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ getqproxy0 td: ", td)
+    # print("~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ getqproxy0 td: ", td)
 
     def qproxybase():
         next_s, rewards = imagine_dynamics(ps, a)

{graph_games_proto-0.3.2225.dist-info → graph_games_proto-0.3.2274.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: graph_games_proto
-Version: 0.3.2225
+Version: 0.3.2274
 Requires-Dist: multipledispatch==1.0.0
 Requires-Dist: pyrsistent==0.20.0
 Requires-Dist: numpy==2.2.4

graph_games_proto-0.3.2274.dist-info/RECORD

@@ -0,0 +1,9 @@
+graph_games_proto/__init__.py,sha256=3XawPcYk9Zk-Hj4M8eNBdGzBccKG8dWj8ZZGRkWnmA4,735
+graph_games_proto/all_types.py,sha256=IpbwftEcHS5Ewz-saFNk0lO9FvcbuHG36odRTayCXUk,54911
+graph_games_proto/fns.py,sha256=YsvqISYFmrU1AVc6geWcxxNwB74-Hqd-LPhnh-oej-8,211037
+graph_games_proto/main.py,sha256=fj2U7KcwrpZtuUhjOX5yVxY18LZvvsxDFYZ_S5mxe04,145
+graph_games_proto/state.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+graph_games_proto-0.3.2274.dist-info/METADATA,sha256=UTUbt6Q5D-LRqREehgGanaXHHqbxVFppCkZhqASLFdw,188
+graph_games_proto-0.3.2274.dist-info/WHEEL,sha256=CmyFI0kx5cdEMTLiONQRbGQwjIoR1aIYB7eCAQ4KPJ0,91
+graph_games_proto-0.3.2274.dist-info/top_level.txt,sha256=-4QSrBMf_MM4BGsr2QXBpqDx8c8k_OPnzGyFjqjakes,18
+graph_games_proto-0.3.2274.dist-info/RECORD,,

graph_games_proto-0.3.2225.dist-info/RECORD

@@ -1,9 +0,0 @@
-graph_games_proto/__init__.py,sha256=BcRdYe0yElSo3vQ5eqhecM1QP-NXaV-mQtcBH6KUnN0,696
-graph_games_proto/all_types.py,sha256=IpbwftEcHS5Ewz-saFNk0lO9FvcbuHG36odRTayCXUk,54911
-graph_games_proto/fns.py,sha256=UgSKX2RtHOY0LwQwV3RE_h0AUE-vGKCiw5nSz25oZF8,211032
-graph_games_proto/main.py,sha256=fj2U7KcwrpZtuUhjOX5yVxY18LZvvsxDFYZ_S5mxe04,145
-graph_games_proto/state.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-graph_games_proto-0.3.2225.dist-info/METADATA,sha256=yq0W9HamGVrJZy9eeEyIe9FwF1qdnKW2HrsuuL2DTm0,188
-graph_games_proto-0.3.2225.dist-info/WHEEL,sha256=CmyFI0kx5cdEMTLiONQRbGQwjIoR1aIYB7eCAQ4KPJ0,91
-graph_games_proto-0.3.2225.dist-info/top_level.txt,sha256=-4QSrBMf_MM4BGsr2QXBpqDx8c8k_OPnzGyFjqjakes,18
-graph_games_proto-0.3.2225.dist-info/RECORD,,