roomai.texas package

Submodules

roomai.texas.TexasHoldemAction module

class roomai.texas.TexasHoldemAction.TexasHoldemAction(key)

Bases: roomai.common.common.AbstractAction

The TexasHoldemAction. The action consists of two parts, namely option and price.

The option is [“Fold”,”Check”,”Call”,”Raise”,”AllIn”], and the price is the chips used by this action.

The TexasHoldemAction has a key “%s_%d”%(option, price) as its identification. Examples of usages:

>> import roomai.TexasHoldem

>> a = roomai.TexasHoldem.TexasHoldemAction.lookup(“Fold_0”)

>> ## We strongly recommend you to get a TexasHoldemAction using the lookup function

>> a.option

“Fold”

>> a.price

0

AllIn = 'Allin'

Call = 'Call'

Check = 'Check'

Fold = 'Fold'

Raise = 'Raise'

key

The key of this action

classmethod lookup(key)

lookup an action with this specified key

Parameters:key – The specified key Returns:The action

option

The option of this action

price

The price of this action

roomai.texas.TexasHoldemEnv module

class roomai.texas.TexasHoldemEnv.TexasHoldemEnv

Bases: roomai.common.common.AbstractEnv

The TexasHoldem game environment

classmethod available_actions(public_state, person_state)

Generate all valid actions given the public state and the person state

Parameters:

• public_state –
• person_state –

Returns:

all valid actions

classmethod compete(env, players)

Use the game environment to hold a compete for the players

Parameters:

• env – The game environment
• players – The players

Returns:

scores for the players

classmethod compute_rank_pattern_cards(pattern_cards)

forward(action)

The TexasHoldem game environments steps with the action taken by the current player

Parameters:action – The action taken by the current player Returns:infos, public_state, person_states, private_state

init(params={})

Initialize the TexasHoldem game environment with the initialization params.

The initialization is a dict with some options

1. allcards: the order of all poker cards appearing
2. record_history: whether to record all history states. if you need call the backward function, please set it to True. default False
3. num_players: how many players are in the game, default 3
4. dealer_id: the player id of the dealer, default random
5. chips: the initialization chips, default [1000,1000,...]
6. big_blind_bet: the number of chips for the big blind bet, default 10

An example of the initialization param is {“num_players”:2,”record_history”:True}

Parameters:params – the initialization params Returns:infos, public_state, person_states, private_state

classmethod is_action_valid(action, public_state, person_state)

Args:

action: public_state: person_state:

Returns:

roomai.texas.TexasHoldemInfo module

class roomai.texas.TexasHoldemInfo.TexasHoldemPersonState

Bases: roomai.common.common.AbstractPersonState

hand_cards

The hand cards of the corresponding player. It contains two poker cards. For example, hand_cards=[roomai.coomon.PokerCard.lookup(“A_Spade”),roomai.coomon.PokerCard.lookup(“A_Heart”)]

class roomai.texas.TexasHoldemInfo.TexasHoldemPrivateState

Bases: roomai.common.common.AbstractPrivateState

The private state of TexasHoldem

keep_cards

the keep cards

class roomai.texas.TexasHoldemInfo.TexasHoldemPublicState

Bases: roomai.common.common.AbstractPublicState

The public state of TexasHoldem

bets

bets is an array which contains the bets from all players. For example, bets=[50,25,25]

big_blind_bet

The big blind bet

chips

chips is an array of the chips of all players. For example, chips=[50,50,50]

dealer_id

The player id of the dealer. The next player after the dealer is the small blind. The next player after the small blind is the big blind.

is_allin

is_allin is an array of which player has take the allin action. For example, is_allin = [true,true,false] denotes the player0 and player1 have taken the allin action

is_fold

is_fold is an array of which player has take the fold action. For example, is_fold = [true,true,false] denotes the player0 and player1 have taken the fold action

is_needed_to_action

is_needed_to_action is an array of which player has take the needed_to_action action. For example, is_needed_to_action = [true,true,false] denotes the player0 and player1 have taken the needed_to_action action

max_bet_sofar

The max bet used by one player so far

num_allin

The number of players who has taken the allin action

num_fold

The number of players who has taken the fold action

num_needed_to_action

The number of players who has taken the needed_to_action action

num_players

The number of players in this game

public_cards

The public cards of this game. For example, public_cards = [roomai.common.PokerCards.lookup(“A_Spade”), roomai.common.PokerCards.lookup(“A_Heart”)]

raise_account

The raise account. If a player want to raise, the price must be max_bet_sofar + raise_account * N. The raise account will increases as the game goes forward

stage

stage in [1,2,3,4]

roomai.texas.TexasHoldemUtil module

class roomai.texas.TexasHoldemUtil.StageSpace

firstStage = 1

fourthStage = 4

secondStage = 2

thirdStage = 3

roomai.texas.play module

class roomai.texas.play.Player

Bases: roomai.common.common.AbstractPlayer

receive_info(info)

reset()

take_action()

roomai.texas.play.show_info(info)

roomai.texas.play.show_public(public_state)

Module contents