gym-csle-stopping-game 0.6.4__py3-none-any.whl → 0.6.5__py3-none-any.whl

gym_csle_stopping_game/__version__.py

@@ -1 +1 @@
-__version__ = '0.6.4'
+__version__ = '0.6.5'

gym_csle_stopping_game/dao/stopping_game_config.py

@@ -14,7 +14,7 @@ class StoppingGameConfig(SimulationEnvInputConfig):
                  T: npt.NDArray[Any], O: npt.NDArray[np.int_], Z: npt.NDArray[Any],
                  R: npt.NDArray[Any], S: npt.NDArray[np.int_], A1: npt.NDArray[np.int_],
                  A2: npt.NDArray[np.int_], L: int, R_INT: int, R_COST: int, R_SLA: int, R_ST: int,
-                 b1: npt.NDArray[np.float_],
+                 b1: npt.NDArray[np.float64],
                  save_dir: str, checkpoint_traces_freq: int, gamma: float = 1, compute_beliefs: bool = True,
                  save_trace: bool = True) -> None:
         """

gym_csle_stopping_game/dao/stopping_game_state.py

@@ -10,7 +10,7 @@ class StoppingGameState(JSONSerializable):
     Represents the state of the optimal stopping game
     """
 
-    def __init__(self, b1: npt.NDArray[np.float_], L: int) -> None:
+    def __init__(self, b1: npt.NDArray[np.float64], L: int) -> None:
         """
         Intializes the state
 

gym_csle_stopping_game/envs/stopping_game_mdp_attacker_env.py

@@ -48,7 +48,7 @@ class StoppingGameMdpAttackerEnv(BaseEnv):
         self.reset()
         super().__init__()
 
-    def step(self, pi2: Union[npt.NDArray[Any], int, float, np.int_, np.float_]) \
+    def step(self, pi2: Union[npt.NDArray[Any], int, float, np.int_, np.float64]) \
             -> Tuple[npt.NDArray[Any], int, bool, bool, Dict[str, Any]]:
         """
         Takes a step in the environment by executing the given action

gym_csle_stopping_game/util/stopping_game_util.py

@@ -11,7 +11,7 @@ class StoppingGameUtil:
     """
 
     @staticmethod
-    def b1() -> npt.NDArray[np.float_]:
+    def b1() -> npt.NDArray[np.float64]:
         """
         Gets the initial belief
 
@@ -233,7 +233,7 @@ class StoppingGameUtil:
         return int(np.random.choice(np.arange(0, len(S)), p=state_probs))
 
     @staticmethod
-    def sample_initial_state(b1: npt.NDArray[np.float_]) -> int:
+    def sample_initial_state(b1: npt.NDArray[np.float64]) -> int:
         """
         Samples the initial state
 
@@ -264,7 +264,7 @@ class StoppingGameUtil:
         return int(o)
 
     @staticmethod
-    def bayes_filter(s_prime: int, o: int, a1: int, b: npt.NDArray[np.float_], pi2: npt.NDArray[Any], l: int,
+    def bayes_filter(s_prime: int, o: int, a1: int, b: npt.NDArray[np.float64], pi2: npt.NDArray[Any], l: int,
                      config: StoppingGameConfig) -> float:
         """
         A Bayesian filter to compute the belief of player 1
@@ -302,8 +302,8 @@ class StoppingGameUtil:
         return float(b_prime_s_prime)
 
     @staticmethod
-    def next_belief(o: int, a1: int, b: npt.NDArray[np.float_], pi2: npt.NDArray[Any],
-                    config: StoppingGameConfig, l: int, a2: int = 0, s: int = 0) -> npt.NDArray[np.float_]:
+    def next_belief(o: int, a1: int, b: npt.NDArray[np.float64], pi2: npt.NDArray[Any],
+                    config: StoppingGameConfig, l: int, a2: int = 0, s: int = 0) -> npt.NDArray[np.float64]:
         """
         Computes the next belief using a Bayesian filter
 
@@ -337,3 +337,52 @@ class StoppingGameUtil:
         :return: a2 is the attacker action
         """
         return int(np.random.choice(np.arange(0, len(pi2[s])), p=pi2[s]))
+
+    @staticmethod
+    def pomdp_solver_file(config: StoppingGameConfig, discount_factor: float, pi2: npt.NDArray[Any]) -> str:
+        """
+        Gets the POMDP environment specification based on the format at http://www.pomdp.org/code/index.html,
+        for the defender's local problem against a static attacker
+
+        :param config: the POMDP config
+        :param discount_factor: the discount factor
+        :param pi2: the attacker strategy
+        :return: the file content as a string
+        """
+        file_str = ""
+        file_str = file_str + f"discount: {discount_factor}\n\n"
+        file_str = file_str + "values: reward\n\n"
+        file_str = file_str + f"states: {len(config.S)}\n\n"
+        file_str = file_str + f"actions: {len(config.A1)}\n\n"
+        file_str = file_str + f"observations: {len(config.O)}\n\n"
+        initial_belief_str = " ".join(list(map(lambda x: str(x), config.b1)))
+        file_str = file_str + f"start: {initial_belief_str}\n\n\n"
+        num_transitions = 0
+        for s in config.S:
+            for a1 in config.A1:
+                probs = []
+                for s_prime in range(len(config.S)):
+                    num_transitions += 1
+                    prob = 0
+                    for a2 in config.A2:
+                        prob += config.T[0][a1][a2][s][s_prime] * pi2[s][a2]
+                    file_str = file_str + f"T: {a1} : {s} : {s_prime} {prob:.80f}\n"
+                    probs.append(prob)
+                assert round(sum(probs), 3) == 1
+        file_str = file_str + "\n\n"
+        for a1 in config.A1:
+            for s_prime in config.S:
+                probs = []
+                for o in range(len(config.O)):
+                    prob = config.Z[0][0][s_prime][o]
+                    file_str = file_str + f"O : {a1} : {s_prime} : {o} {prob:.80f}\n"
+                    probs.append(prob)
+                assert round(sum(probs), 3) == 1
+        file_str = file_str + "\n\n"
+        for s in config.S:
+            for a1 in config.A1:
+                for s_prime in config.S:
+                    for o in config.O:
+                        r = config.R[0][a1][0][s]
+                        file_str = file_str + f"R: {a1} : {s} : {s_prime} : {o} {r:.80f}\n"
+        return file_str

gym_csle_stopping_game-0.6.5.dist-info/METADATA

@@ -0,0 +1,46 @@
+Metadata-Version: 2.1
+Name: gym-csle-stopping-game
+Version: 0.6.5
+Summary: OpenAI gym reinforcement learning environment of a Dynkin (Optimal stopping) game in CSLE
+Author: Kim Hammar
+Author-email: hammar.kim@gmail.com
+License: Creative Commons Attribution-ShareAlike 4.0 International
+Keywords: Reinforcement-Learning Cyber-Security Markov-Games Markov-Decision-Processes
+Platform: unix
+Platform: linux
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3 :: Only
+Classifier: Programming Language :: Python :: 3.8
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Intended Audience :: Science/Research
+Requires-Python: >=3.8
+Requires-Dist: gymnasium>=0.27.1
+Requires-Dist: csle-base>=0.6.5
+Requires-Dist: csle-common>=0.6.5
+Requires-Dist: csle-attacker>=0.6.5
+Requires-Dist: csle-defender>=0.6.5
+Requires-Dist: csle-collector>=0.6.5
+Provides-Extra: testing
+Requires-Dist: pytest>=6.0; extra == "testing"
+Requires-Dist: pytest-cov>=2.0; extra == "testing"
+Requires-Dist: pytest-mock>=3.6.0; extra == "testing"
+Requires-Dist: grpcio>=1.57.0; extra == "testing"
+Requires-Dist: grpcio-tools>=1.57.0; extra == "testing"
+Requires-Dist: pytest-grpc>=0.8.0; extra == "testing"
+Requires-Dist: mypy>=1.4.1; extra == "testing"
+Requires-Dist: mypy-extensions>=1.0.0; extra == "testing"
+Requires-Dist: mypy-protobuf>=3.5.0; extra == "testing"
+Requires-Dist: types-PyYAML>=6.0.12.11; extra == "testing"
+Requires-Dist: types-paramiko>=3.2.0.0; extra == "testing"
+Requires-Dist: types-protobuf>=4.23.0.3; extra == "testing"
+Requires-Dist: types-requests>=2.31.0.1; extra == "testing"
+Requires-Dist: types-urllib3>=1.26.25.13; extra == "testing"
+Requires-Dist: flake8>=6.1.0; extra == "testing"
+Requires-Dist: flake8-rst-docstrings>=0.3.0; extra == "testing"
+Requires-Dist: tox>=3.24; extra == "testing"
+Requires-Dist: sphinx>=5.3.0; extra == "testing"
+Requires-Dist: sphinxcontrib-napoleon>=0.7; extra == "testing"
+Requires-Dist: sphinx-rtd-theme>=1.1.1; extra == "testing"
+Requires-Dist: twine>=4.0.2; extra == "testing"
+Requires-Dist: build>=0.10.0; extra == "testing"
+

{gym_csle_stopping_game-0.6.4.dist-info → gym_csle_stopping_game-0.6.5.dist-info}/RECORD

@@ -1,19 +1,19 @@
 gym_csle_stopping_game/__init__.py,sha256=ooy6TjxvBi1sZMEX3_mVlvfskqI5GqwITWzI882tfk0,657
-gym_csle_stopping_game/__version__.py,sha256=9V8HNBWlWd9kKrEEJE04ypjOSYSmFVS4xh0H3Zq8EGE,22
+gym_csle_stopping_game/__version__.py,sha256=g0OeNAF6f8ojnlyb43abksBvbLUrnL-AlwclSoULgd0,22
 gym_csle_stopping_game/constants/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 gym_csle_stopping_game/constants/constants.py,sha256=eIoD9eXifZ73kP-lSlvG-IXCpe4n6D-_aDygx0zOr5U,1030
 gym_csle_stopping_game/dao/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 gym_csle_stopping_game/dao/stopping_game_attacker_mdp_config.py,sha256=kmtrVa2CCVbi5xfd6uPWqMvhGmP8ccrtn1e_VmVvH7k,3494
-gym_csle_stopping_game/dao/stopping_game_config.py,sha256=0_CtrFod1NkZ5eqlwIq68IKzv87FIP5s4crDnc4eFZI,6640
+gym_csle_stopping_game/dao/stopping_game_config.py,sha256=r77noSBWbP1AeBFiBVmiDeEn58SkBcHGklJStzAEEkE,6641
 gym_csle_stopping_game/dao/stopping_game_defender_pomdp_config.py,sha256=3FfNi2-R6n1LqjA644EVq-v7wtp6sqyEkEdBN90-2n0,3753
-gym_csle_stopping_game/dao/stopping_game_state.py,sha256=v9jEG7-reOQTkSyTZH64jCZ8IyixXFEIk4HIcmhBHiw,2828
+gym_csle_stopping_game/dao/stopping_game_state.py,sha256=bwlvjzPkrgD61yp9EIZRZJWexEN-4SmY2OU__iIp3c0,2829
 gym_csle_stopping_game/envs/__init__.py,sha256=SQHaqXI0_2HYsC8i9swXEHDFcXKEYpb8GRP9l_S0Sw8,74
 gym_csle_stopping_game/envs/stopping_game_env.py,sha256=J9h73alytskNgHxa7LHM83R1PUNIALDoy0PCItA4P9k,16403
-gym_csle_stopping_game/envs/stopping_game_mdp_attacker_env.py,sha256=bgoddSXUT-TkDN5T8jJljsul4CRYiPRW7VMmtM7QrwU,10877
+gym_csle_stopping_game/envs/stopping_game_mdp_attacker_env.py,sha256=UoSDY2U8x3AmOqiquwnAzolsya6bM2vVsAkW7i8ljWk,10878
 gym_csle_stopping_game/envs/stopping_game_pomdp_defender_env.py,sha256=cyC2OuJA41aqE84KA9-oJRWsKLIqzekh8A8zyf6Qo5I,8766
 gym_csle_stopping_game/util/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-gym_csle_stopping_game/util/stopping_game_util.py,sha256=s3a5jkUlbh_2K6ALz5riE1w9gfdosf4N3yvrNMBxHUg,11363
-gym_csle_stopping_game-0.6.4.dist-info/METADATA,sha256=Q9AikBlL-ZWon-v7TOy_bdZUYxd-xt4W53CgqQp9NLw,2162
-gym_csle_stopping_game-0.6.4.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
-gym_csle_stopping_game-0.6.4.dist-info/top_level.txt,sha256=3DBHkAEI00nq0aXZlJUkXJrLiwkcJCfaFoYcaOzEZUU,23
-gym_csle_stopping_game-0.6.4.dist-info/RECORD,,
+gym_csle_stopping_game/util/stopping_game_util.py,sha256=IlgdrQCVjkt4ta2T18Hws3b9vxJFlcywsPgWWmN06H4,13629
+gym_csle_stopping_game-0.6.5.dist-info/METADATA,sha256=ZAhlxpiwp6QftyAYkNXYVrD6bpYmLhkt9tD3_UFXIvE,2112
+gym_csle_stopping_game-0.6.5.dist-info/WHEEL,sha256=eOLhNAGa2EW3wWl_TU484h7q1UNgy0JXjjoqKoxAAQc,92
+gym_csle_stopping_game-0.6.5.dist-info/top_level.txt,sha256=3DBHkAEI00nq0aXZlJUkXJrLiwkcJCfaFoYcaOzEZUU,23
+gym_csle_stopping_game-0.6.5.dist-info/RECORD,,

{gym_csle_stopping_game-0.6.4.dist-info → gym_csle_stopping_game-0.6.5.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: bdist_wheel (0.43.0)
+Generator: bdist_wheel (0.44.0)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

gym_csle_stopping_game-0.6.4.dist-info/METADATA

@@ -1,46 +0,0 @@
-Metadata-Version: 2.1
-Name: gym-csle-stopping-game
-Version: 0.6.4
-Summary: OpenAI gym reinforcement learning environment of a Dynkin (Optimal stopping) game in CSLE
-Author: Kim Hammar
-Author-email: hammar.kim@gmail.com
-License: Creative Commons Attribution-ShareAlike 4.0 International
-Keywords: Reinforcement-Learning Cyber-Security Markov-Games Markov-Decision-Processes
-Platform: unix
-Platform: linux
-Classifier: Programming Language :: Python :: 3
-Classifier: Programming Language :: Python :: 3 :: Only
-Classifier: Programming Language :: Python :: 3.8
-Classifier: Programming Language :: Python :: 3.9
-Classifier: Intended Audience :: Science/Research
-Requires-Python: >=3.8
-Requires-Dist: gymnasium >=0.27.1
-Requires-Dist: csle-base >=0.6.4
-Requires-Dist: csle-common >=0.6.4
-Requires-Dist: csle-attacker >=0.6.4
-Requires-Dist: csle-defender >=0.6.4
-Requires-Dist: csle-collector >=0.6.4
-Provides-Extra: testing
-Requires-Dist: pytest >=6.0 ; extra == 'testing'
-Requires-Dist: pytest-cov >=2.0 ; extra == 'testing'
-Requires-Dist: pytest-mock >=3.6.0 ; extra == 'testing'
-Requires-Dist: grpcio >=1.57.0 ; extra == 'testing'
-Requires-Dist: grpcio-tools >=1.57.0 ; extra == 'testing'
-Requires-Dist: pytest-grpc >=0.8.0 ; extra == 'testing'
-Requires-Dist: mypy >=1.4.1 ; extra == 'testing'
-Requires-Dist: mypy-extensions >=1.0.0 ; extra == 'testing'
-Requires-Dist: mypy-protobuf >=3.5.0 ; extra == 'testing'
-Requires-Dist: types-PyYAML >=6.0.12.11 ; extra == 'testing'
-Requires-Dist: types-paramiko >=3.2.0.0 ; extra == 'testing'
-Requires-Dist: types-protobuf >=4.23.0.3 ; extra == 'testing'
-Requires-Dist: types-requests >=2.31.0.1 ; extra == 'testing'
-Requires-Dist: types-urllib3 >=1.26.25.13 ; extra == 'testing'
-Requires-Dist: flake8 >=6.1.0 ; extra == 'testing'
-Requires-Dist: flake8-rst-docstrings >=0.3.0 ; extra == 'testing'
-Requires-Dist: tox >=3.24 ; extra == 'testing'
-Requires-Dist: sphinx >=5.3.0 ; extra == 'testing'
-Requires-Dist: sphinxcontrib-napoleon >=0.7 ; extra == 'testing'
-Requires-Dist: sphinx-rtd-theme >=1.1.1 ; extra == 'testing'
-Requires-Dist: twine >=4.0.2 ; extra == 'testing'
-Requires-Dist: build >=0.10.0 ; extra == 'testing'
-