Package kuimaze :: Module gym_wrapper :: Class MDPMaze

Class MDPMaze

gym.Env --+    
          |    
    MazeEnv --+
              |
             MDPMaze

maze for solving MDP problems

Instance Methods

__init__(self, map_image=None, grad=(0,0), probs=None, node_rewards=None)
Class wrapping Maze into gym enviroment.

source code

_get_reward(self, curr, last)
returns reward and indication of goal state

source code

get_actions(self, state)

source code

is_goal_state(self, state)

source code

is_terminal_state(self, state)

source code

get_next_states_and_probs(self, state, action)

source code

get_state_reward(self, curr)

source code

Inherited from MazeEnv: close, get_all_states, render, reset, save_eps, save_path, seed, step, visualise

Inherited from MazeEnv (private): _get_action_space, _get_depth, _get_observation

Class Variables

Inherited from MazeEnv: MAP, metadata

Inherited from MazeEnv (private): _path, _visited

Method Details

Class wrapping Maze into gym enviroment.

Parameters:

(inherited documentation)

returns reward and indication of goal state

Parameters:

Returns:

float, boolean