ast_toolbox.mcts.ASTSim module¶
-
class
ast_toolbox.mcts.ASTSim.AcionSequence(sequence, index=0)[source]¶ Bases:
objectSturcture storing the actions sequences.
Parameters: - sequence (list) – The list of actions.
- index (int, optional) – The initial action index in the sequence.
-
ast_toolbox.mcts.ASTSim.action_seq_policy(action_seq, s)[source]¶ The policy wrapper for the action sequence.
Parameters: - action_seq (
ast_toolbox.mcts.ASTSim.AcionSequence) – The action sequence. - s (
ast_toolbox.mcts.AdaptiveStressTesting.ASTState) – The AST state.
Returns: action (ast_toolbox.mcts.AdaptiveStressTesting.ASTAction) – The AST action.
- action_seq (
-
ast_toolbox.mcts.ASTSim.play_sequence(ast, actions, verbose=False, sleeptime=0.0)[source]¶ Rollout the action sequence.
Parameters: - ast (
ast_toolbox.mcts.AdaptiveStressTesting.AdaptiveStressTest) – The AST object. - actions (list) – The action sequence.
- verbose (bool, optional) – Whether to log the rollout information.
- sleeptime (float, optional) – The pause time between each step.
Returns: - rewards (list[float]) – The rewards.
- actions2 (list) – The action sequence of the path. Should be the same as the input actions.
- ast (