The source code and dockerfile for the GSW2024 AI Lab.
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
This repo is archived. You can view files and clone it, but cannot push or open issues/pull-requests.
 
 
 
 
 
 

133 lines
4.7 KiB

import gymnasium as gym
import numpy as np
import random
from moviepy.editor import ImageSequenceClip
from utils import MiniGridShieldHandler, common_parser
from stable_baselines3.common.callbacks import BaseCallback, CheckpointCallback
from stable_baselines3.common.logger import Image
class MiniGridSbShieldingWrapper(gym.core.Wrapper):
def __init__(self,
env,
shield_handler : MiniGridShieldHandler,
create_shield_at_reset = False,
):
super().__init__(env)
self.shield_handler = shield_handler
self.create_shield_at_reset = create_shield_at_reset
shield = self.shield_handler.create_shield(env=self.env)
self.shield = shield
def create_action_mask(self):
try:
return self.shield[self.env.get_symbolic_state()]
except:
return [0.0] * 3 + [0.0] * 4
def reset(self, *, seed=None, options=None):
obs, infos = self.env.reset(seed=seed, options=options)
if self.create_shield_at_reset:
shield = self.shield_handler.create_shield(env=self.env)
self.shield = shield
return obs, infos
def step(self, action):
obs, rew, done, truncated, info = self.env.step(action)
info["no_shield_action"] = not self.shield.__contains__(self.env.get_symbolic_state())
return obs, rew, done, truncated, info
def parse_sb3_arguments():
parser = common_parser()
args = parser.parse_args()
return args
class ImageRecorderCallback(BaseCallback):
def __init__(self, eval_env, render_freq, n_eval_episodes, evaluation_method, log_dir, deterministic=True, verbose=0):
super().__init__(verbose)
self._eval_env = eval_env
self._render_freq = render_freq
self._n_eval_episodes = n_eval_episodes
self._deterministic = deterministic
self._evaluation_method = evaluation_method
self._log_dir = log_dir
def _on_training_start(self):
image = self.training_env.render(mode="rgb_array")
self.logger.record("trajectory/image", Image(image, "HWC"), exclude=("stdout", "log", "json", "csv"))
def _on_step(self) -> bool:
#if self.n_calls % self._render_freq == 0:
# self.record_video()
return True
def _on_training_end(self) -> None:
self.record_video()
def record_video(self) -> bool:
screens = []
def grab_screens(_locals, _globals) -> None:
"""
Renders the environment in its current state, recording the screen in the captured `screens` list
:param _locals: A dictionary containing all local variables of the callback's scope
:param _globals: A dictionary containing all global variables of the callback's scope
"""
screen = self._eval_env.render()
screens.append(screen)
self._evaluation_method(
self.model,
self._eval_env,
callback=grab_screens,
n_eval_episodes=self._n_eval_episodes,
deterministic=self._deterministic,
)
clip = ImageSequenceClip(list(screens), fps=3)
clip.write_gif(f"{self._log_dir}/{self.n_calls}.gif", fps=3)
return True
class InfoCallback(BaseCallback):
"""
Custom callback for plotting additional values in tensorboard.
"""
def __init__(self, verbose=0):
super().__init__(verbose)
self.sum_goal = 0
self.sum_lava = 0
self.sum_collisions = 0
self.sum_opened_door = 0
self.sum_picked_up = 0
self.no_shield_action = 0
def _on_step(self) -> bool:
infos = self.locals["infos"][0]
if infos["reached_goal"]:
self.sum_goal += 1
if infos["ran_into_lava"]:
self.sum_lava += 1
self.logger.record("info/sum_reached_goal", self.sum_goal)
self.logger.record("info/sum_ran_into_lava", self.sum_lava)
if "collision" in infos:
if infos["collision"]:
self.sum_collisions += 1
self.logger.record("info/sum_collision", self.sum_collisions)
if "opened_door" in infos:
if infos["opened_door"]:
self.sum_opened_door += 1
self.logger.record("info/sum_opened_door", self.sum_opened_door)
if "picked_up" in infos:
if infos["picked_up"]:
self.sum_picked_up += 1
self.logger.record("info/sum_picked_up", self.sum_picked_up)
if "no_shield_action" in infos:
if infos["no_shield_action"]:
self.no_shield_action += 1
self.logger.record("info/no_shield_action", self.no_shield_action)
return True