Files
Gymnasium/gym/envs/toy_text/blackjack.py

237 lines
8.0 KiB
Python
Raw Normal View History

Seeding update (#2422) * Ditch most of the seeding.py and replace np_random with the numpy default_rng. Let's see if tests pass * Updated a bunch of RNG calls from the RandomState API to Generator API * black; didn't expect that, did ya? * Undo a typo * blaaack * More typo fixes * Fixed setting/getting state in multidiscrete spaces * Fix typo, fix a test to work with the new sampling * Correctly (?) pass the randomly generated seed if np_random is called with None as seed * Convert the Discrete sample to a python int (as opposed to np.int64) * Remove some redundant imports * First version of the compatibility layer for old-style RNG. Mainly to trigger tests. * Removed redundant f-strings * Style fixes, removing unused imports * Try to make tests pass by removing atari from the dockerfile * Try to make tests pass by removing atari from the setup * Try to make tests pass by removing atari from the setup * Try to make tests pass by removing atari from the setup * First attempt at deprecating `env.seed` and supporting `env.reset(seed=seed)` instead. Tests should hopefully pass but throw up a million warnings. * black; didn't expect that, didya? * Rename the reset parameter in VecEnvs back to `seed` * Updated tests to use the new seeding method * Removed a bunch of old `seed` calls. Fixed a bug in AsyncVectorEnv * Stop Discrete envs from doing part of the setup (and using the randomness) in init (as opposed to reset) * Add explicit seed to wrappers reset * Remove an accidental return * Re-add some legacy functions with a warning. * Use deprecation instead of regular warnings for the newly deprecated methods/functions
2021-12-08 22:14:15 +01:00
from typing import Optional
import os
import numpy as np
import pygame
Seeding update (#2422) * Ditch most of the seeding.py and replace np_random with the numpy default_rng. Let's see if tests pass * Updated a bunch of RNG calls from the RandomState API to Generator API * black; didn't expect that, did ya? * Undo a typo * blaaack * More typo fixes * Fixed setting/getting state in multidiscrete spaces * Fix typo, fix a test to work with the new sampling * Correctly (?) pass the randomly generated seed if np_random is called with None as seed * Convert the Discrete sample to a python int (as opposed to np.int64) * Remove some redundant imports * First version of the compatibility layer for old-style RNG. Mainly to trigger tests. * Removed redundant f-strings * Style fixes, removing unused imports * Try to make tests pass by removing atari from the dockerfile * Try to make tests pass by removing atari from the setup * Try to make tests pass by removing atari from the setup * Try to make tests pass by removing atari from the setup * First attempt at deprecating `env.seed` and supporting `env.reset(seed=seed)` instead. Tests should hopefully pass but throw up a million warnings. * black; didn't expect that, didya? * Rename the reset parameter in VecEnvs back to `seed` * Updated tests to use the new seeding method * Removed a bunch of old `seed` calls. Fixed a bug in AsyncVectorEnv * Stop Discrete envs from doing part of the setup (and using the randomness) in init (as opposed to reset) * Add explicit seed to wrappers reset * Remove an accidental return * Re-add some legacy functions with a warning. * Use deprecation instead of regular warnings for the newly deprecated methods/functions
2021-12-08 22:14:15 +01:00
import gym
from gym import spaces
[WIP] add support for seeding environments (#135) * Make environments seedable * Fix monitor bugs - Set monitor_id before setting the infix. This was a bug that would yield incorrect results with multiple monitors. - Remove extra pid from stats recorder filename. This should be purely cosmetic. * Start uploading seeds in episode_batch * Fix _bigint_from_bytes for python3 * Set seed explicitly in random_agent * Pass through seed argument * Also pass through random state to spaces * Pass random state into the observation/action spaces * Make all _seed methods return the list of used seeds * Switch over to np.random where possible * Start hashing seeds, and also seed doom engine * Fixup seeding determinism in many cases * Seed before loading the ROM * Make seeding more Python3 friendly * Make the MuJoCo skipping a bit more forgiving * Remove debugging PDB calls * Make setInt argument into raw bytes * Validate and upload seeds * Skip box2d * Make seeds smaller, and change representation of seeds in upload * Handle long seeds * Fix RandomAgent example to be deterministic * Handle integer types correctly in Python2 and Python3 * Try caching pip * Try adding swap * Add df and free calls * Bump swap * Bump swap size * Try setting overcommit * Try other sysctls * Try fixing overcommit * Try just setting overcommit_memory=1 * Add explanatory comment * Add what's new section to readme * BUG: Mark ElevatorAction-ram-v0 as non-deterministic for now * Document seed * Move nondetermistic check into spec
2016-05-29 09:07:09 -07:00
from gym.utils import seeding
2016-05-10 17:05:47 +02:00
def cmp(a, b):
return float(a > b) - float(a < b)
2016-05-10 17:05:47 +02:00
# 1 = Ace, 2-10 = Number cards, Jack/Queen/King = 10
deck = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 10, 10, 10]
[WIP] add support for seeding environments (#135) * Make environments seedable * Fix monitor bugs - Set monitor_id before setting the infix. This was a bug that would yield incorrect results with multiple monitors. - Remove extra pid from stats recorder filename. This should be purely cosmetic. * Start uploading seeds in episode_batch * Fix _bigint_from_bytes for python3 * Set seed explicitly in random_agent * Pass through seed argument * Also pass through random state to spaces * Pass random state into the observation/action spaces * Make all _seed methods return the list of used seeds * Switch over to np.random where possible * Start hashing seeds, and also seed doom engine * Fixup seeding determinism in many cases * Seed before loading the ROM * Make seeding more Python3 friendly * Make the MuJoCo skipping a bit more forgiving * Remove debugging PDB calls * Make setInt argument into raw bytes * Validate and upload seeds * Skip box2d * Make seeds smaller, and change representation of seeds in upload * Handle long seeds * Fix RandomAgent example to be deterministic * Handle integer types correctly in Python2 and Python3 * Try caching pip * Try adding swap * Add df and free calls * Bump swap * Bump swap size * Try setting overcommit * Try other sysctls * Try fixing overcommit * Try just setting overcommit_memory=1 * Add explanatory comment * Add what's new section to readme * BUG: Mark ElevatorAction-ram-v0 as non-deterministic for now * Document seed * Move nondetermistic check into spec
2016-05-29 09:07:09 -07:00
def draw_card(np_random):
return int(np_random.choice(deck))
[WIP] add support for seeding environments (#135) * Make environments seedable * Fix monitor bugs - Set monitor_id before setting the infix. This was a bug that would yield incorrect results with multiple monitors. - Remove extra pid from stats recorder filename. This should be purely cosmetic. * Start uploading seeds in episode_batch * Fix _bigint_from_bytes for python3 * Set seed explicitly in random_agent * Pass through seed argument * Also pass through random state to spaces * Pass random state into the observation/action spaces * Make all _seed methods return the list of used seeds * Switch over to np.random where possible * Start hashing seeds, and also seed doom engine * Fixup seeding determinism in many cases * Seed before loading the ROM * Make seeding more Python3 friendly * Make the MuJoCo skipping a bit more forgiving * Remove debugging PDB calls * Make setInt argument into raw bytes * Validate and upload seeds * Skip box2d * Make seeds smaller, and change representation of seeds in upload * Handle long seeds * Fix RandomAgent example to be deterministic * Handle integer types correctly in Python2 and Python3 * Try caching pip * Try adding swap * Add df and free calls * Bump swap * Bump swap size * Try setting overcommit * Try other sysctls * Try fixing overcommit * Try just setting overcommit_memory=1 * Add explanatory comment * Add what's new section to readme * BUG: Mark ElevatorAction-ram-v0 as non-deterministic for now * Document seed * Move nondetermistic check into spec
2016-05-29 09:07:09 -07:00
def draw_hand(np_random):
return [draw_card(np_random), draw_card(np_random)]
def usable_ace(hand): # Does this hand have a usable ace?
return 1 in hand and sum(hand) + 10 <= 21
def sum_hand(hand): # Return current hand total
if usable_ace(hand):
return sum(hand) + 10
return sum(hand)
def is_bust(hand): # Is this hand a bust?
return sum_hand(hand) > 21
def score(hand): # What is the score of this hand (0 if bust)
return 0 if is_bust(hand) else sum_hand(hand)
def is_natural(hand): # Is this hand a natural blackjack?
return sorted(hand) == [1, 10]
class BlackjackEnv(gym.Env):
"""Simple blackjack environment
Blackjack is a card game where the goal is to obtain cards that sum to as
near as possible to 21 without going over. They're playing against a fixed
dealer.
Face cards (Jack, Queen, King) have point value 10.
Aces can either count as 11 or 1, and it's called 'usable' at 11.
This game is placed with an infinite deck (or with replacement).
The game starts with dealer having one face up and one face down card, while
player having two face up cards. (Virtually for all Blackjack games today).
The player can request additional cards (hit=1) until they decide to stop
(stick=0) or exceed 21 (bust).
After the player sticks, the dealer reveals their facedown card, and draws
until their sum is 17 or greater. If the dealer goes bust the player wins.
If neither player nor dealer busts, the outcome (win, lose, draw) is
decided by whose sum is closer to 21. The reward for winning is +1,
drawing is 0, and losing is -1.
The observation of a 3-tuple of: the players current sum,
the dealer's one showing card (1-10 where 1 is ace),
and whether or not the player holds a usable ace (0 or 1).
This environment corresponds to the version of the blackjack problem
described in Example 5.1 in Reinforcement Learning: An Introduction
by Sutton and Barto.
http://incompleteideas.net/book/the-book-2nd.html
"""
2021-07-29 02:26:34 +02:00
metadata = {"render.modes": ["human", "rgb_array"]}
def __init__(self, natural=False, sab=False):
self.action_space = spaces.Discrete(2)
2021-07-29 15:39:42 -04:00
self.observation_space = spaces.Tuple(
(spaces.Discrete(32), spaces.Discrete(11), spaces.Discrete(2))
)
[WIP] add support for seeding environments (#135) * Make environments seedable * Fix monitor bugs - Set monitor_id before setting the infix. This was a bug that would yield incorrect results with multiple monitors. - Remove extra pid from stats recorder filename. This should be purely cosmetic. * Start uploading seeds in episode_batch * Fix _bigint_from_bytes for python3 * Set seed explicitly in random_agent * Pass through seed argument * Also pass through random state to spaces * Pass random state into the observation/action spaces * Make all _seed methods return the list of used seeds * Switch over to np.random where possible * Start hashing seeds, and also seed doom engine * Fixup seeding determinism in many cases * Seed before loading the ROM * Make seeding more Python3 friendly * Make the MuJoCo skipping a bit more forgiving * Remove debugging PDB calls * Make setInt argument into raw bytes * Validate and upload seeds * Skip box2d * Make seeds smaller, and change representation of seeds in upload * Handle long seeds * Fix RandomAgent example to be deterministic * Handle integer types correctly in Python2 and Python3 * Try caching pip * Try adding swap * Add df and free calls * Bump swap * Bump swap size * Try setting overcommit * Try other sysctls * Try fixing overcommit * Try just setting overcommit_memory=1 * Add explanatory comment * Add what's new section to readme * BUG: Mark ElevatorAction-ram-v0 as non-deterministic for now * Document seed * Move nondetermistic check into spec
2016-05-29 09:07:09 -07:00
# Flag to payout 1.5 on a "natural" blackjack win, like casino rules
# Ref: http://www.bicyclecards.com/how-to-play/blackjack/
self.natural = natural
# Flag for full agreement with the (Sutton and Barto, 2018) definition. Overrides self.natural
self.sab = sab
def step(self, action):
2016-06-16 00:20:22 -07:00
assert self.action_space.contains(action)
if action: # hit: add a card to players hand and return
[WIP] add support for seeding environments (#135) * Make environments seedable * Fix monitor bugs - Set monitor_id before setting the infix. This was a bug that would yield incorrect results with multiple monitors. - Remove extra pid from stats recorder filename. This should be purely cosmetic. * Start uploading seeds in episode_batch * Fix _bigint_from_bytes for python3 * Set seed explicitly in random_agent * Pass through seed argument * Also pass through random state to spaces * Pass random state into the observation/action spaces * Make all _seed methods return the list of used seeds * Switch over to np.random where possible * Start hashing seeds, and also seed doom engine * Fixup seeding determinism in many cases * Seed before loading the ROM * Make seeding more Python3 friendly * Make the MuJoCo skipping a bit more forgiving * Remove debugging PDB calls * Make setInt argument into raw bytes * Validate and upload seeds * Skip box2d * Make seeds smaller, and change representation of seeds in upload * Handle long seeds * Fix RandomAgent example to be deterministic * Handle integer types correctly in Python2 and Python3 * Try caching pip * Try adding swap * Add df and free calls * Bump swap * Bump swap size * Try setting overcommit * Try other sysctls * Try fixing overcommit * Try just setting overcommit_memory=1 * Add explanatory comment * Add what's new section to readme * BUG: Mark ElevatorAction-ram-v0 as non-deterministic for now * Document seed * Move nondetermistic check into spec
2016-05-29 09:07:09 -07:00
self.player.append(draw_card(self.np_random))
if is_bust(self.player):
done = True
2021-07-29 02:26:34 +02:00
reward = -1.0
else:
done = False
2021-07-29 02:26:34 +02:00
reward = 0.0
else: # stick: play out the dealers hand, and score
done = True
while sum_hand(self.dealer) < 17:
[WIP] add support for seeding environments (#135) * Make environments seedable * Fix monitor bugs - Set monitor_id before setting the infix. This was a bug that would yield incorrect results with multiple monitors. - Remove extra pid from stats recorder filename. This should be purely cosmetic. * Start uploading seeds in episode_batch * Fix _bigint_from_bytes for python3 * Set seed explicitly in random_agent * Pass through seed argument * Also pass through random state to spaces * Pass random state into the observation/action spaces * Make all _seed methods return the list of used seeds * Switch over to np.random where possible * Start hashing seeds, and also seed doom engine * Fixup seeding determinism in many cases * Seed before loading the ROM * Make seeding more Python3 friendly * Make the MuJoCo skipping a bit more forgiving * Remove debugging PDB calls * Make setInt argument into raw bytes * Validate and upload seeds * Skip box2d * Make seeds smaller, and change representation of seeds in upload * Handle long seeds * Fix RandomAgent example to be deterministic * Handle integer types correctly in Python2 and Python3 * Try caching pip * Try adding swap * Add df and free calls * Bump swap * Bump swap size * Try setting overcommit * Try other sysctls * Try fixing overcommit * Try just setting overcommit_memory=1 * Add explanatory comment * Add what's new section to readme * BUG: Mark ElevatorAction-ram-v0 as non-deterministic for now * Document seed * Move nondetermistic check into spec
2016-05-29 09:07:09 -07:00
self.dealer.append(draw_card(self.np_random))
reward = cmp(score(self.player), score(self.dealer))
if self.sab and is_natural(self.player) and not is_natural(self.dealer):
# Player automatically wins. Rules consistent with S&B
reward = 1.0
elif (
not self.sab
and self.natural
and is_natural(self.player)
and reward == 1.0
):
# Natural gives extra points, but doesn't autowin. Legacy implementation
reward = 1.5
return self._get_obs(), reward, done, {}
def _get_obs(self):
return (sum_hand(self.player), self.dealer[0], usable_ace(self.player))
def reset(self, *, seed: Optional[int] = None, options: Optional[dict] = None):
Seeding update (#2422) * Ditch most of the seeding.py and replace np_random with the numpy default_rng. Let's see if tests pass * Updated a bunch of RNG calls from the RandomState API to Generator API * black; didn't expect that, did ya? * Undo a typo * blaaack * More typo fixes * Fixed setting/getting state in multidiscrete spaces * Fix typo, fix a test to work with the new sampling * Correctly (?) pass the randomly generated seed if np_random is called with None as seed * Convert the Discrete sample to a python int (as opposed to np.int64) * Remove some redundant imports * First version of the compatibility layer for old-style RNG. Mainly to trigger tests. * Removed redundant f-strings * Style fixes, removing unused imports * Try to make tests pass by removing atari from the dockerfile * Try to make tests pass by removing atari from the setup * Try to make tests pass by removing atari from the setup * Try to make tests pass by removing atari from the setup * First attempt at deprecating `env.seed` and supporting `env.reset(seed=seed)` instead. Tests should hopefully pass but throw up a million warnings. * black; didn't expect that, didya? * Rename the reset parameter in VecEnvs back to `seed` * Updated tests to use the new seeding method * Removed a bunch of old `seed` calls. Fixed a bug in AsyncVectorEnv * Stop Discrete envs from doing part of the setup (and using the randomness) in init (as opposed to reset) * Add explicit seed to wrappers reset * Remove an accidental return * Re-add some legacy functions with a warning. * Use deprecation instead of regular warnings for the newly deprecated methods/functions
2021-12-08 22:14:15 +01:00
super().reset(seed=seed)
[WIP] add support for seeding environments (#135) * Make environments seedable * Fix monitor bugs - Set monitor_id before setting the infix. This was a bug that would yield incorrect results with multiple monitors. - Remove extra pid from stats recorder filename. This should be purely cosmetic. * Start uploading seeds in episode_batch * Fix _bigint_from_bytes for python3 * Set seed explicitly in random_agent * Pass through seed argument * Also pass through random state to spaces * Pass random state into the observation/action spaces * Make all _seed methods return the list of used seeds * Switch over to np.random where possible * Start hashing seeds, and also seed doom engine * Fixup seeding determinism in many cases * Seed before loading the ROM * Make seeding more Python3 friendly * Make the MuJoCo skipping a bit more forgiving * Remove debugging PDB calls * Make setInt argument into raw bytes * Validate and upload seeds * Skip box2d * Make seeds smaller, and change representation of seeds in upload * Handle long seeds * Fix RandomAgent example to be deterministic * Handle integer types correctly in Python2 and Python3 * Try caching pip * Try adding swap * Add df and free calls * Bump swap * Bump swap size * Try setting overcommit * Try other sysctls * Try fixing overcommit * Try just setting overcommit_memory=1 * Add explanatory comment * Add what's new section to readme * BUG: Mark ElevatorAction-ram-v0 as non-deterministic for now * Document seed * Move nondetermistic check into spec
2016-05-29 09:07:09 -07:00
self.dealer = draw_hand(self.np_random)
self.player = draw_hand(self.np_random)
return self._get_obs()
def render(self, mode="human"):
player_sum, dealer_card_value, usable_ace = self._get_obs()
screen_width, screen_height = 600, 500
card_img_height = screen_height // 3
card_img_width = int(card_img_height * 142 / 197)
spacing = screen_height // 20
bg_color = (7, 99, 36)
white = (255, 255, 255)
if not hasattr(self, "screen"):
if mode == "human":
pygame.init()
self.screen = pygame.display.set_mode((screen_width, screen_height))
else:
pygame.font.init()
self.screen = pygame.Surface((screen_width, screen_height))
self.screen.fill(bg_color)
def get_image(path):
cwd = os.path.dirname(__file__)
image = pygame.image.load(os.path.join(cwd, path))
return image
def get_font(path, size):
cwd = os.path.dirname(__file__)
font = pygame.font.Font(os.path.join(cwd, path), size)
return font
small_font = get_font(
os.path.join("font", "Minecraft.ttf"), screen_height // 15
)
dealer_text = small_font.render(
"Dealer: " + str(dealer_card_value), True, white
)
dealer_text_rect = self.screen.blit(dealer_text, (spacing, spacing))
suits = ["C", "D", "H", "S"]
dealer_card_suit = self.np_random.choice(suits)
if dealer_card_value == 1:
dealer_card_value_str = "A"
elif dealer_card_value == 10:
dealer_card_value_str = self.np_random.choice(["J", "Q", "K"])
else:
dealer_card_value_str = str(dealer_card_value)
def scale_card_img(card_img):
return pygame.transform.scale(card_img, (card_img_width, card_img_height))
dealer_card_img = scale_card_img(
get_image(
os.path.join("img", dealer_card_suit + dealer_card_value_str + ".png")
)
)
dealer_card_rect = self.screen.blit(
dealer_card_img,
(
screen_width // 2 - card_img_width - spacing // 2,
dealer_text_rect.bottom + spacing,
),
)
hidden_card_img = scale_card_img(get_image(os.path.join("img", "Card.png")))
self.screen.blit(
hidden_card_img,
(
screen_width // 2 + spacing // 2,
dealer_text_rect.bottom + spacing,
),
)
player_text = small_font.render("Player", True, white)
player_text_rect = self.screen.blit(
player_text, (spacing, dealer_card_rect.bottom + 1.5 * spacing)
)
large_font = get_font(os.path.join("font", "Minecraft.ttf"), screen_height // 6)
player_sum_text = large_font.render(str(player_sum), True, white)
player_sum_text_rect = self.screen.blit(
player_sum_text,
(
screen_width // 2 - player_sum_text.get_width() // 2,
player_text_rect.bottom + spacing,
),
)
if usable_ace:
usable_ace_text = small_font.render("usable ace", True, white)
self.screen.blit(
usable_ace_text,
(
screen_width // 2 - usable_ace_text.get_width() // 2,
player_sum_text_rect.bottom + spacing // 2,
),
)
if mode == "human":
pygame.display.update()
else:
return np.transpose(
np.array(pygame.surfarray.pixels3d(self.screen)), axes=(1, 0, 2)
)