FrozenLake 6x6ΒΆ

[1]:

import gymnasium as gym

[2]:

from gymcts.gymcts_agent import GymctsAgent
from gymcts.gymcts_deepcopy_wrapper import DeepCopyMCTSGymEnvWrapper

[3]:

from gymcts.logger import log

[4]:

log.setLevel(20)

[5]:

if __name__ == '__main__':
    log.debug("Starting example")

    # 0. create the environment
    custom_map = [
        "SFFFFF",
        "FFFFFF",
        "FFFHFF",
        "FFFFFH",
        "FHFFFF",
        "FFFFFG"
    ]
    env = gym.make(
        'FrozenLake-v1',
        desc=custom_map,
        map_name=None,
        is_slippery=True,
        render_mode="ansi"
    )
    env.reset()

    # 1. wrap the environment with the naive wrapper or a custom gymcts wrapper
    env = DeepCopyMCTSGymEnvWrapper(env)

    # 2. create the agent
    agent = GymctsAgent(env=env, clear_mcts_tree_after_step=False)

    # 3. solve the environment
    actions = agent.solve(num_simulations_per_step=200)

    # 4. render the environment solution in the terminal
    print(env.render())
    for a in actions:
        obs, rew, term, trun, info = env.step(a)
        print(env.render())

[16:47:17] INFO     selected action 1 after 200 simulations.

           INFO     current action list: [1]

[16:47:18] INFO     selected action 1 after 200 simulations.

           INFO     current action list: [1, 1]

[16:47:19] INFO     selected action 0 after 200 simulations.

           INFO     current action list: [1, 1, 0]

           INFO     selected action 0 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0]

[16:47:20] INFO     selected action 1 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0, 1]

[16:47:21] INFO     selected action 2 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0, 1, 2]

           INFO     selected action 0 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0, 1, 2, 0]

[16:47:22] INFO     selected action 2 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0, 1, 2, 0, 2]

[16:47:23] INFO     selected action 1 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0, 1, 2, 0, 2, 1]

[16:47:24] INFO     selected action 3 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0, 1, 2, 0, 2, 1, 3]

           INFO     selected action 1 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0, 1, 2, 0, 2, 1, 3, 1]

[16:47:25] INFO     selected action 0 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0, 1, 2, 0, 2, 1, 3, 1, 0]

[16:47:26] INFO     selected action 3 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0, 1, 2, 0, 2, 1, 3, 1, 0, 3]

           INFO     selected action 2 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0, 1, 2, 0, 2, 1, 3, 1, 0, 3, 2]

[16:47:27] INFO     selected action 1 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0, 1, 2, 0, 2, 1, 3, 1, 0, 3, 2, 1]

[16:47:28] INFO     selected action 2 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0, 1, 2, 0, 2, 1, 3, 1, 0, 3, 2, 1, 2]

           INFO     selected action 1 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0, 1, 2, 0, 2, 1, 3, 1, 0, 3, 2, 1, 2, 1]

[16:47:29] INFO     selected action 0 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0, 1, 2, 0, 2, 1, 3, 1, 0, 3, 2, 1, 2, 1, 0]

[16:47:30] INFO     selected action 3 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0, 1, 2, 0, 2, 1, 3, 1, 0, 3, 2, 1, 2, 1, 0, 3]

           INFO     selected action 2 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0, 1, 2, 0, 2, 1, 3, 1, 0, 3, 2, 1, 2, 1, 0, 3, 2]

           INFO     selected action 1 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0, 1, 2, 0, 2, 1, 3, 1, 0, 3, 2, 1, 2, 1, 0, 3, 2, 1]

           INFO     selected action 0 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0, 1, 2, 0, 2, 1, 3, 1, 0, 3, 2, 1, 2, 1, 0, 3, 2, 1, 0]

           INFO     selected action 2 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0, 1, 2, 0, 2, 1, 3, 1, 0, 3, 2, 1, 2, 1, 0, 3, 2, 1, 0, 2]

[16:47:31] INFO     selected action 1 after 200 simulations.

           INFO     current action list: [1, 1, 0, 0, 1, 2, 0, 2, 1, 3, 1, 0, 3, 2, 1, 2, 1, 0, 3, 2, 1, 0, 2, 1]

           INFO     Final action list: [1, 1, 0, 0, 1, 2, 0, 2, 1, 3, 1, 0, 3, 2, 1, 2, 1, 0, 3, 2, 1, 0, 2, 1]


SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Down)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Down)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Left)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Left)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Down)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Right)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Left)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Right)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Down)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Up)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Down)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Left)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Up)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Right)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Down)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Right)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Down)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Left)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Up)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Right)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Down)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Left)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Right)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG

  (Down)
SFFFFF
FFFFFF
FFFHFF
FFFFFH
FHFFFF
FFFFFG