jaclearn.rl.envs.maze package

Submodules

jaclearn.rl.envs.maze.maze module

class jaclearn.rl.envs.maze.maze.MazeEnv(map_size=14, visible_size=None, obs_ratio=0.3, enable_path_checking=True, random_action_mapping=None, enable_noaction=False, dense_reward=False, reward_move=None, reward_noaction=0, reward_final=10, reward_error=-2, state_mode='DEFAULT')

Bases: jaclearn.rl.env.SimpleRLEnvBase

Create a maze environment.

action_delta

the tuple (dy, dx) when you perform action i

Type:Action deltas

action_mapping

If random action mapping is enabled, return the internal mapping

canvas()

Return the raw canvas (full)

canvas_size

Canvas size

current_point()

Current point (r, c)

distance_mat

Distance matrix

distance_prev

Distance-prev matrix

final_point()

Finish point (r, c)

inv_distance_mat

inv_distance_prev

map_size

Map size

obstacles()

origin_canvas()

Return the original canvas (at time 0, full)

quick_distance_mat()

Distance matrix: this is done during the first run of SPFA, so if you ensure that all valid points are in the same connected component, you can use it

quick_distance_prev()

Distance-prev matrix: see also quick_distance_mat

restart(obstacles=None, start_point=None, final_point=None)

rewards

(Move, Noaction, Arrive final point, Move Err)

Type:A tuple of 4 value, representing the rewards for each action

shortest_path()

One of the shortest paths from start to finish, list of point (r, c)

start_point()

Start point (r, c)

visible_size

Visible size

class jaclearn.rl.envs.maze.maze.CustomLavaWorldEnv(map_size=15, mode=None, **kwargs)

Bases: jaclearn.rl.envs.maze.maze.MazeEnv

A maze similar to Lava World in OpenAI Gym

lv_finals

lv_obstacles

lv_starts

restart(obstacles=None, start_point=None, final_point=None)

jaclearn.rl.envs.maze.taxi module

class jaclearn.rl.envs.maze.taxi.CustomTaxiEnv(maze_env=None, use_coord=False, *args, **kwargs)

Bases: jaclearn.rl.env.SimpleRLEnvBase

maze_env

phase

restart(start_point=None, final_point1=None, final_point2=None)

class jaclearn.rl.envs.maze.taxi.CustomLavaWorldTaxiEnv(maze_env=None, use_coord=False, *args, **kwargs)

Bases: jaclearn.rl.envs.maze.taxi.CustomTaxiEnv