"""wrappers from https://github.com/openai/baselines/blob/master/baselines/common/atari_wrappers.py"""
import os
os.environ.setdefault('PATH', '')
import gym
import cv2
import numpy as np
from collections import deque
cv2.ocl.setUseOpenCL(False)
from pytorchrl import EMBED
from pytorchrl.envs.common import FrameStack
from pytorchrl.envs.atari.utils import imdownscale
[docs]class StickyActionEnv(gym.Wrapper):
def __init__(self, env, p=0.25):
super(StickyActionEnv, self).__init__(env)
self.p = p
self.last_action = 0
[docs] def step(self, action):
if np.random.uniform() < self.p:
action = self.last_action
self.last_action = action
return self.env.step(action)
[docs] def reset(self):
self.last_action = 0
return self.env.reset()
[docs]class TimeLimit(gym.Wrapper):
def __init__(self, env, max_episode_steps=None):
super(TimeLimit, self).__init__(env)
self._max_episode_steps = max_episode_steps
self._elapsed_steps = 0
[docs] def step(self, ac):
observation, reward, done, info = self.env.step(ac)
self._elapsed_steps += 1
if self._elapsed_steps >= self._max_episode_steps:
done = True
info['TimeLimit.truncated'] = True
return observation, reward, done, info
[docs] def reset(self, **kwargs):
self._elapsed_steps = 0
return self.env.reset(**kwargs)
[docs]class NoopResetEnv(gym.Wrapper):
def __init__(self, env, noop_max=30):
"""Sample initial states by taking random number of no-ops on reset.
No-op is assumed to be action 0.
"""
gym.Wrapper.__init__(self, env)
self.noop_max = noop_max
self.override_num_noops = None
self.noop_action = 0
assert env.unwrapped.get_action_meanings()[0] == 'NOOP'
[docs] def reset(self, **kwargs):
""" Do no-op action for a number of steps in [1, noop_max]."""
self.env.reset(**kwargs)
if self.override_num_noops is not None:
noops = self.override_num_noops
else:
noops = self.unwrapped.np_random.randint(1, self.noop_max + 1) # pylint: disable=E1101
assert noops > 0
obs = None
for _ in range(noops):
obs, _, done, _ = self.env.step(self.noop_action)
if done:
obs = self.env.reset(**kwargs)
return obs
[docs] def step(self, ac):
return self.env.step(ac)
[docs]class FireResetEnv(gym.Wrapper):
def __init__(self, env):
"""Take action on reset for environments that are fixed until firing."""
gym.Wrapper.__init__(self, env)
assert env.unwrapped.get_action_meanings()[1] == 'FIRE'
assert len(env.unwrapped.get_action_meanings()) >= 3
[docs] def reset(self, **kwargs):
self.env.reset(**kwargs)
obs, _, done, _ = self.env.step(1)
if done:
self.env.reset(**kwargs)
obs, _, done, _ = self.env.step(2)
if done:
self.env.reset(**kwargs)
return obs
[docs] def step(self, ac):
return self.env.step(ac)
[docs]class EpisodicLifeEnv(gym.Wrapper):
def __init__(self, env):
"""Make end-of-life == end-of-episode, but only reset on true game over.
Done by DeepMind for the DQN and co. since it helps value estimation.
"""
gym.Wrapper.__init__(self, env)
self.lives = 0
self.was_real_done = True
self.total_reward = 0.0
[docs] def step(self, action):
obs, reward, done, info = self.env.step(action)
self.total_reward += reward
info['EpisodicReward'] = self.total_reward
self.was_real_done = done
# check current lives, make loss of life terminal,
# then update lives to handle bonus lives
lives = self.env.unwrapped.ale.lives()
if lives < self.lives and lives > 0:
# for Qbert sometimes we stay in lives == 0 condition for a few frames
# so it's important to keep lives > 0, so that we only reset once
# the environment advertises done.
done = True
info['Lives'] = lives
self.lives = lives
return obs, reward, done, info
[docs] def reset(self, **kwargs):
"""Reset only when lives are exhausted.
This way all states are still reachable even though lives are episodic,
and the learner need not know about any of this behind-the-scenes.
"""
if self.was_real_done:
self.total_reward = 0.0
obs = self.env.reset(**kwargs)
else:
# no-op step to advance from terminal/lost life state
obs, _, _, _ = self.env.step(0)
self.lives = self.env.unwrapped.ale.lives()
return obs
[docs]class ClipRewardEnv(gym.Wrapper):
def __init__(self, env):
gym.Wrapper.__init__(self, env)
self.total_reward = 0.0
[docs] def step(self, action):
obs, reward, done, info = self.env.step(action)
self.total_reward += reward
info['UnclippedReward'] = self.total_reward
reward = np.sign(reward)
return obs, reward, done, info
[docs] def reset(self, **kwargs):
self.total_reward = 0.0
return self.env.reset(**kwargs)
[docs]class ScaleRewardEnv(gym.Wrapper):
def __init__(self, env, scaling=0.001):
gym.Wrapper.__init__(self, env)
self.scaling = scaling
self.total_reward = 0.0
[docs] def step(self, action):
obs, reward, done, info = self.env.step(action)
self.total_reward += reward
info['UnscaledReward'] = self.total_reward
reward = self.scaling * reward
return obs, reward, done, info
[docs] def reset(self, **kwargs):
self.total_reward = 0.0
return self.env.reset(**kwargs)
[docs]class MaxAndSkipEnv(gym.Wrapper):
def __init__(self, env, skip=4):
"""Return only every `skip`-th frame"""
gym.Wrapper.__init__(self, env)
# most recent raw observations (for max pooling across time steps)
self._obs_buffer = np.zeros((2,) + env.observation_space.shape, dtype=np.uint8)
self._skip = skip
[docs] def step(self, action):
"""Repeat action, sum reward, and max over last observations."""
total_reward = 0.0
done = None
for i in range(self._skip):
obs, reward, done, info = self.env.step(action)
if i == self._skip - 2:
self._obs_buffer[0] = obs
if i == self._skip - 1:
self._obs_buffer[1] = obs
total_reward += reward
if done:
break
# Note that the observation on the done=True frame
# doesn't matter
max_frame = self._obs_buffer.max(axis=0)
return max_frame, total_reward, done, info
[docs] def reset(self, **kwargs):
return self.env.reset(**kwargs)
[docs]class MontezumaVisitedRoomEnv(gym.Wrapper):
def __init__(self, env, room_address):
gym.Wrapper.__init__(self, env)
self.room_address = room_address
self.visited_rooms = set() # Only stores unique numbers.
[docs] def step(self, action):
state, reward, done, info = self.env.step(action)
ram = self.unwrapped.ale.getRAM()
assert len(ram) == 128
self.visited_rooms.add(ram[self.room_address])
info['VisitedRooms'] = len(self.visited_rooms)
return state, reward, done, info
[docs] def reset(self):
self.visited_rooms.clear()
return self.env.reset()
[docs]class MontezumaEmbeddingsEnv(gym.Wrapper):
def __init__(self, env, embeddings_shape=(11, 8), embeddings_num_values=8, use_domain_knowledge=False,
domain_knowledge_embedding="default", double_state=False):
gym.Wrapper.__init__(self, env)
# pos 0: The current frame of the episode.
# pos 3: The current screen. Room?
# pod 57: room level?
# pos 19, 20, 21: The score, represented in Binary Coded Decimal. This is, every nibble represents a decimal digit
# pos 42: joe_x, from 0 to 153
# pos 43: joe_y, from 0 to 122 0 135 to 253
# pos 52: agent orientation, 76 or 128
# pos 65: inventory
# mallet +1
# key +2
# key +4
# key +8
# key +16
# sword +32
# sword +64
# torch +128
self.joe_x = 42
self.joe_y = 43
self.room_level = 57
self.room_address = 3
self.joe_inventory = 65
self.last_state = None
self.embeddings_shape = embeddings_shape
self.embeddings_num_values = embeddings_num_values
self.use_domain_knowledge = use_domain_knowledge
self.domain_knowledge_embedding = domain_knowledge_embedding
self.double_state = double_state
if double_state:
self._embed_buff = deque(maxlen=2)
[docs] def step(self, action):
# Create embedding
if self.use_domain_knowledge:
ram = self.unwrapped.ale.getRAM()
assert len(ram) == 128
if self.domain_knowledge_embedding == "default":
embed_state = np.array(
[
np.clip(ram[self.joe_x], 0, 153), # range 0 - 153
np.clip(ram[self.joe_y], 135, 253), # range 135 - 253
ram[self.room_address] + 24 * ram[self.room_level],
ram[self.joe_inventory],
]
)
elif self.domain_knowledge_embedding == "room_inventory":
embed_state = np.array(
[
ram[self.room_address] + 24 * ram[self.room_level],
ram[self.joe_inventory],
]
)
elif self.domain_knowledge_embedding == "room":
embed_state = np.array(
[
ram[self.room_address] + 24 * ram[self.room_level],
]
)
state, reward, done, info = self.env.step(action)
else:
state, reward, done, info = self.env.step(action)
embed_state = imdownscale(
state=self.last_state[:, :, -1],
target_shape=self.embeddings_shape,
max_pix_value=self.embeddings_num_values)
# Concat last 2 embeddings if specified
if self.double_state:
if len(self._embed_buff) < 2:
self._embed_buff.append(embed_state)
self._embed_buff.append(embed_state)
if (embed_state != self._embed_buff[-1]).any():
self._embed_buff.append(embed_state)
embed_state = np.concatenate(self._embed_buff)
info.update({EMBED: embed_state})
self.last_state = state
return state, reward, done, info
[docs] def reset(self):
self.last_state = self.env.reset()
return self.last_state
[docs]class PitfallEmbeddingsEnv(gym.Wrapper):
def __init__(self, env, embeddings_shape=(11, 8), embeddings_num_values=8, use_domain_knowledge=False,
double_state=False):
gym.Wrapper.__init__(self, env)
# pos 1: room ID
self.room_level = 1
self.last_state = None
self.embeddings_shape = embeddings_shape
self.embeddings_num_values = embeddings_num_values
self.use_domain_knowledge = use_domain_knowledge
self.double_state = double_state
if double_state:
self._embed_buff = deque(maxlen=2)
[docs] def step(self, action):
# Create embedding
if self.use_domain_knowledge:
ram = self.unwrapped.ale.getRAM()
assert len(ram) == 128
embed_state = np.array([np.clip(ram[self.room_level], 0, 255)]) # range 0 - 255
state, reward, done, info = self.env.step(action)
else:
state, reward, done, info = self.env.step(action)
embed_state = imdownscale(
state=self.last_state[:, :, -1],
target_shape=self.embeddings_shape,
max_pix_value=self.embeddings_num_values)
# Concat last 2 embeddings if specified
if self.double_state:
if len(self._embed_buff) < 2:
self._embed_buff.append(embed_state)
self._embed_buff.append(embed_state)
if (embed_state != self._embed_buff[-1]).any():
self._embed_buff.append(embed_state)
embed_state = np.concatenate(self._embed_buff)
info.update({EMBED: embed_state})
self.last_state = state
return state, reward, done, info
[docs] def reset(self):
self.last_state = self.env.reset()
return self.last_state
[docs]class WarpFrame(gym.ObservationWrapper):
def __init__(self, env, width=84, height=84, grayscale=True, dict_space_key=None):
"""
Warp frames to 84x84 as done in the Nature paper and later work.
If the environment uses dictionary observations, `dict_space_key` can be specified which indicates which
observation should be warped.
"""
super().__init__(env)
self._width = width
self._height = height
self._grayscale = grayscale
self._key = dict_space_key
if self._grayscale:
num_colors = 1
else:
num_colors = 3
new_space = gym.spaces.Box(
low=0,
high=255,
shape=(self._height, self._width, num_colors),
dtype=np.uint8,
)
if self._key is None:
original_space = self.observation_space
self.observation_space = new_space
else:
original_space = self.observation_space.spaces[self._key]
self.observation_space.spaces[self._key] = new_space
assert original_space.dtype == np.uint8 and len(original_space.shape) == 3
[docs] def observation(self, obs):
if self._key is None:
frame = obs
else:
frame = obs[self._key]
if self._grayscale:
frame = cv2.cvtColor(frame, cv2.COLOR_RGB2GRAY)
frame = cv2.resize(
frame, (self._width, self._height), interpolation=cv2.INTER_AREA
)
if self._grayscale:
frame = np.expand_dims(frame, -1)
if self._key is None:
obs = frame
else:
obs = obs.copy()
obs[self._key] = frame
return obs
[docs]class ScaledFloatFrame(gym.ObservationWrapper):
def __init__(self, env):
gym.ObservationWrapper.__init__(self, env)
self.observation_space = gym.spaces.Box(low=0, high=1, shape=env.observation_space.shape, dtype=np.float32)
[docs] def observation(self, observation):
# careful! This undoes the memory optimization, use
# with smaller replay buffers only.
return np.array(observation).astype(np.float32) / 255.0
[docs]class LazyFrames(object):
def __init__(self, frames):
"""
This object ensures that common frames between the observations are only stored once.
It exists purely to optimize memory usage which can be huge for DQN's 1M frames replay
buffers.
This object should only be converted to numpy array before being passed to the model.
You'd not believe how complex the previous solution was.
"""
self._frames = frames
self._out = None
def _force(self):
if self._out is None:
self._out = np.concatenate(self._frames, axis=-1)
self._frames = None
return self._out
def __array__(self, dtype=None):
out = self._force()
if dtype is not None:
out = out.astype(dtype)
return out
def __len__(self):
return len(self._force())
def __getitem__(self, i):
return self._force()[i]
[docs] def count(self):
frames = self._force()
return frames.shape[frames.ndim - 1]
[docs] def frame(self, i):
return self._force()[..., i]
[docs]def wrap_deepmind(env, episode_life=True, clip_rewards=True, frame_stack=1, scale=False):
"""Configure environment for DeepMind-style Atari"""
if episode_life:
env = EpisodicLifeEnv(env)
if 'FIRE' in env.unwrapped.get_action_meanings():
env = FireResetEnv(env)
env = WarpFrame(env)
if scale:
env = ScaledFloatFrame(env)
if clip_rewards:
env = ClipRewardEnv(env)
env = FrameStack(env, frame_stack)
return env
[docs]def make_atari(env_id, max_episode_steps=None, sticky_actions=False):
env = gym.make(env_id)
env._max_episode_steps = max_episode_steps * 4
assert 'NoFrameskip' in env.spec.id
env = NoopResetEnv(env, noop_max=30)
env = MaxAndSkipEnv(env, skip=4)
if sticky_actions:
env = StickyActionEnv(env)
if max_episode_steps is not None:
env = TimeLimit(env, max_episode_steps=max_episode_steps)
return env