betbot.prediction package

Submodules

betbot.prediction.BettingOddsPredictor module

class betbot.prediction.BettingOddsPredictor.BettingOddsPredictor(api: betbot.api.ApiConnection.ApiConnection, league: str, season: int)

Bases: betbot.prediction.Predictor.Predictor

Class that determinalistically predicts matches based on Tipico quotes

generate_scores(home_odds: float, draw_odds: float, away_odds: float) → Tuple[int, int]

Generates a scoreline based on the quote data :param home_odds: The odds that the home team wins :param draw_odds: The odds that the match ends in a draw :param away_odds: The odds that the away team wins :return: The generated bet

classmethod name() → str

Returns

The name of the predictor

predict(matches: List[betbot.api.Match.Match]) → List[betbot.api.Bet.Bet]

Performs the prediction :param matches: The matches to predict :return: The predictions as Bet objects

betbot.prediction.DrawPredictor module

class betbot.prediction.DrawPredictor.DrawPredictor(api: betbot.api.ApiConnection.ApiConnection, league: str, season: int)

Bases: betbot.prediction.Predictor.Predictor

Class that always predicts 0:0, 1:1, 2:2 or 3:3

classmethod name() → str

Returns

The name of the predictor

predict(matches: List[betbot.api.Match.Match]) → List[betbot.api.Bet.Bet]

Performs the prediction :param matches: The matches to predict :return: The predictions as Bet objects

betbot.prediction.HomeTeamPredictor module

class betbot.prediction.HomeTeamPredictor.HomeTeamPredictor(api: betbot.api.ApiConnection.ApiConnection, league: str, season: int)

Bases: betbot.prediction.Predictor.Predictor

Class that always predicts 2:1 for the home team

classmethod name() → str

Returns

The name of the predictor

predict(matches: List[betbot.api.Match.Match]) → List[betbot.api.Bet.Bet]

Performs the prediction :param matches: The matches to predict :return: The predictions as Bet objects

betbot.prediction.LeagueTablePredictor module

class betbot.prediction.LeagueTablePredictor.LeagueTablePredictor(api: betbot.api.ApiConnection.ApiConnection, league: str, season: int)

Bases: betbot.prediction.Predictor.Predictor

Class that always 2:1 for the team higher up in the league table

classmethod name() → str

Returns

The name of the predictor

predict(matches: List[betbot.api.Match.Match]) → List[betbot.api.Bet.Bet]

Performs the prediction :param matches: The matches to predict :return: The predictions as Bet objects

betbot.prediction.NameAndOddsPredictor module

class betbot.prediction.NameAndOddsPredictor.NameAndOddsPredictor(api: betbot.api.ApiConnection.ApiConnection, league: str, season: int)

Bases: betbot.prediction.SKLearnPredictor.SKLearnPredictor

scikit-learn powered predictor that uses the name and betting odds

interpret_results(home_result: float, away_result: float) → Tuple[int, int]

Interprets the raw results :param home_result: The home goals result :param away_result: The away goals result :return: The home goals, the away goals

classmethod name() → str

Returns

The name of the predictor

betbot.prediction.NamePredictor module

class betbot.prediction.NamePredictor.NamePredictor(api: betbot.api.ApiConnection.ApiConnection, league: str, season: int)

Bases: betbot.prediction.SKLearnPredictor.SKLearnPredictor

scikit-learn powered predictor that uses only the team names

encode_result(home_score: int, away_score: int) → numpy.ndarray

Encodes a result vector This is done as a normalization step. :param home_score: The home score to encode :param away_score: The away score to encode :return: The encoded result vector

interpret_results(home_result: float, away_result: float) → Tuple[int, int]

Interprets the raw results :param home_result: The home goals result :param away_result: The away goals result :return: The home goals, the away goals

classmethod name() → str

Returns

The name of the predictor

predict_match(match: betbot.api.Match.Match) → Optional[Tuple[int, int]]

Predicts the result of a single match using the trained model :param match: The match to predict :return: The home goals and away goals or None

if no prediction took place

vectorize(match_data: Dict[str, Union[str, float]]) → numpy.ndarray

Defines how a match is vectorized :param match_data: The match data to vectorize :return: The vector for the match

betbot.prediction.Predictor module

class betbot.prediction.Predictor.Predictor(api: betbot.api.ApiConnection.ApiConnection, league: str, season: int)

Bases: object

Class that specifies required methods for predictor objects

__init__(api: betbot.api.ApiConnection.ApiConnection, league: str, season: int)

Initializes the model directory if it does not exist :param api: The bundesliga-tippspiel API connection :param league: The league for which to predict matches :param season: The season for which to predict matches

classmethod name() → str

Returns

The name of the predictor

predict(matches: List[betbot.api.Match.Match]) → List[betbot.api.Bet.Bet]

Performs the prediction :param matches: The matches to predict :return: The predictions as Bet objects

betbot.prediction.RandomPredictor module

class betbot.prediction.RandomPredictor.RandomPredictor(api: betbot.api.ApiConnection.ApiConnection, league: str, season: int)

Bases: betbot.prediction.Predictor.Predictor

Class that always predicts random results

classmethod name() → str

Returns

The name of the predictor

predict(matches: List[betbot.api.Match.Match]) → List[betbot.api.Bet.Bet]

Performs the prediction :param matches: The matches to predict :return: The predictions as Bet objects

betbot.prediction.SKLearnPredictor module

class betbot.prediction.SKLearnPredictor.SKLearnPredictor(api: betbot.api.ApiConnection.ApiConnection, league: str, season: int)

Bases: betbot.prediction.Predictor.Predictor

Abstract class that defines how a scikit-learn-based predictor should operate

__init__(api: betbot.api.ApiConnection.ApiConnection, league: str, season: int)

Initializes the scikit-learn model :param api: The bundesliga-tippspiel API connection :param league: The league for which to predict matches :param season: The season for which to predict matches

encode_result(home_score: int, away_score: int) → numpy.ndarray

Encodes a result vector This is done as a normalization step. :param home_score: The home score to encode :param away_score: The away score to encode :return: The encoded result vector

interpret_results(home_result: float, away_result: float) → Tuple[int, int]

Interprets the raw results :param home_result: The home goals result :param away_result: The away goals result :return: The home goals, the away goals

property odds: Dict[Tuple[str, str], Tuple[float, float, float]]

Retrieves current odds using a mix of football-data.co.uk and oddsportal.com :return: The odds for each match in the selected league

predict(matches: List[betbot.api.Match.Match]) → List[betbot.api.Bet.Bet]

Performs the prediction :param matches: The matches to predict :return: The predictions as Bet objects

predict_match(match: betbot.api.Match.Match) → Optional[Tuple[int, int]]

Predicts the result of a single match using the trained model :param match: The match to predict :return: The home goals and away goals or None

if no prediction took place

classmethod regressor() → sklearn.neural_network._multilayer_perceptron.MLPRegressor

Defines the regressor used during the prediction process :return: The predictor

train()

Trains the prediction model

vectorize(match_data: Dict[str, Union[str, float]]) → numpy.ndarray

Defines how a match is vectorized :param match_data: The match data to vectorize :return: The vector for the match

Module contents