| import random | |
| import pandas as pd | |
| import os | |
| class Model: | |
| """ | |
| Class containing the info of a model. | |
| :param name: Name of the model | |
| :param elo: Elo rating of the model | |
| :param games_played: Number of games played by the model (useful if we implement sigma uncertainty) | |
| """ | |
| def __init__(self, name, elo): | |
| self.name = name | |
| self.elo = elo | |
| self.games_played = 0 | |
| class Matchmaking: | |
| """ | |
| Class managing the matchmaking between the models. | |
| :param models: List of models | |
| :param queue: Temporary list of models used for the matching process | |
| :param k: Dev coefficient | |
| :param max_diff: Maximum difference considered between two models' elo | |
| :param matches: Dictionary containing the match history (to later upload as CSV) | |
| """ | |
| def __init__(self): | |
| self.models = [] | |
| self.queue = [] | |
| self.start_elo = 1200 | |
| self.k = 20 | |
| self.max_diff = 500 | |
| self.matches = pd.DataFrame() | |
| def read_history(self): | |
| """ Read the match history from the CSV files, concat the Dataframes and sort them by datetime. """ | |
| path = "match_history" | |
| files = os.listdir(path) | |
| for file in files: | |
| self.matches = pd.concat([self.matches, pd.read_csv(os.path.join(path, file))], ignore_index=True) | |
| self.matches["datetime"] = pd.to_datetime(self.matches["datetime"], format="%Y-%m-%d %H:%M:%S.%f", errors="coerce") | |
| self.matches = self.matches.dropna() | |
| self.matches = self.matches.sort_values("datetime") | |
| self.matches.reset_index(drop=True, inplace=True) | |
| model_names = self.matches["model1"].unique() | |
| self.models = [Model(name, self.start_elo) for name in model_names] | |
| def compute_elo(self): | |
| """ Compute the elo for each model after each match. """ | |
| for i, row in self.matches.iterrows(): | |
| model1 = self.get_model(row["model1"]) | |
| model2 = self.get_model(row["model2"]) | |
| result = row["result"] | |
| delta = model1.elo - model2.elo | |
| win_probability = 1 / (1 + 10 ** (-delta / 500)) | |
| model1.elo += self.k * (result - win_probability) | |
| model2.elo -= self.k * (result - win_probability) | |
| model1.games_played += 1 | |
| model2.games_played += 1 | |
| def save_elo_data(self): | |
| """ Save the match history as a CSV file to the hub. """ | |
| df = pd.DataFrame(columns=['name', 'elo']) | |
| for model in self.models: | |
| df = pd.concat([df, pd.DataFrame([[model.name, model.elo]], columns=['name', 'elo'])]) | |
| df.to_csv('elo.csv', index=False) | |
| def get_model(self, name): | |
| """ Return the Model with the given name. """ | |
| for model in self.models: | |
| if model.name == name: | |
| return model | |
| return None | |