Note
Click here to download the full example code
Saving and loading a bettor
This example shows save_bettor and load_bettor.
# Author: Georgios Douzas <gdouzas@icloud.com>
# Licence: MIT
import tempfile
from pathlib import Path
import matplotlib.pyplot as plt
import pandas as pd
from sportsbet.dataloaders import DataLoader
from sportsbet.evaluation import OddsComparisonBettor, load_bettor, save_bettor
from sportsbet.sources import SampleSoccerOdds, SampleSoccerStats
A fitted bettor comes back fitted
The model you backtested is the model that should place the bets. Saving it makes them the same object. Otherwise you have two models that were fitted separately and are only hopefully the same.
dataloader = DataLoader(
param_grid={'league': ['England']},
stats=SampleSoccerStats(),
odds=SampleSoccerOdds(),
)
X_train, Y_train, O_train = dataloader.extract_train_data(odds_type='market_average')
bettor = OddsComparisonBettor(alpha=0.03, betting_markets=['home_win', 'draw', 'away_win'])
_ = bettor.fit(X_train, Y_train, O_train)
path = str(Path(tempfile.mkdtemp()) / 'bettor.pkl')
save_bettor(bettor, path)
It is ready to bet without being fitted again.
reloaded = load_bettor(path)
reloaded.betting_markets_.tolist()
Out:
['home_win', 'draw', 'away_win']
reloaded.bet(X_train, O_train)
Out:
array([[False, False, True],
[ True, False, False],
[False, False, False],
...,
[False, False, False],
[ True, False, False],
[False, False, True]], shape=(380, 3))
A picture of it
markets = reloaded.betting_markets_.tolist()
counts = pd.DataFrame(reloaded.bet(X_train, O_train), columns=markets).sum()
fig, ax = plt.subplots()
ax.bar(counts.index, counts.to_numpy())
ax.set_title('Value bets found, by market')
ax.set_ylabel('bets')
Out:
Text(46.722222222222214, 0.5, 'bets')
Total running time of the script: ( 0 minutes 0.731 seconds)
Download Python source code: plot_save_load_bettor.py