main.py

import os
import sys
from datetime import datetime

import pandas as pd
import rpy2.robjects.packages as rpackages
from joblib import Parallel, delayed
from tqdm import tqdm

from src.match import Match
from src.myconstants import *
from src.record_manager import RecordManager
from src.soccercpd import SoccerCPD

# Uncomment this code if it raises an error
# os.environ['R_HOME'] = "/usr/lib/R"  # or whereever your R is installed"


pd.set_option("display.width", 250)
pd.set_option("display.max_rows", 100)
pd.set_option("display.max_columns", 20)


class HiddenPrints:
    def __enter__(self):
        self._original_stdout = sys.stdout
        sys.stdout = open(os.devnull, "w")

    def __exit__(self, exc_type, exc_val, exc_tb):
        sys.stdout.close()
        sys.stdout = self._original_stdout


def analyze_activity_inner(i, outliers, apply_cpd, formcpd_type):
    tic = datetime.now()
    activity_id = activity_records.at[i, LABEL_ACTIVITY_ID]
    date = activity_records.at[i, LABEL_DATE]
    team_name = activity_records.at[i, LABEL_TEAM_NAME]
    print()
    print("=" * 78)
    print(f"[{i}] activity_id: {activity_id}, date: {date}, team_name: {team_name}")

    activity_args = rm.load_activity_data(activity_id)
    activity_outliers = None
    if outliers is not None:
        activity_outliers = outliers[outliers[LABEL_ACTIVITY_ID] == activity_id][LABEL_PLAYER_ID].tolist()
    match = Match(*activity_args, outliers=activity_outliers)

    if match.player_periods[LABEL_PLAYER_IDS].iloc[1:].apply(lambda x: len(x)).max() >= 10:
        # Filter in-play data from the measured data using the start, end, and substitution records
        match.construct_inplay_ugp()

        # Rotate the pitch for one of the sessions so that the team always attacks from left to right
        match.rotate_pitch()

        # Apply SoccerCPD on the preprocessed match data
        cpd = SoccerCPD(match, apply_cpd=apply_cpd, formcpd_type=formcpd_type)
        cpd.run()
        if not cpd.fgp_df.empty:
            cpd.visualize()
            cpd.save_stats()
            toc = datetime.now()
            print("The total process takes {:.3f} sec.".format((toc - tic).total_seconds()))
            return i, 1
        else:
            return i, 0

    else:
        # If at least one player has not been measured during the match,
        # skip the process for the match since the data is incomplete
        print("Not enough players to estimate a formation.")
        return i, 0


def analyze_activity(i, outliers, apply_cpd=True, formcpd_type="gseg_avg", verbose=True):
    if verbose:
        return analyze_activity_inner(i, outliers, apply_cpd, formcpd_type)
    else:
        with HiddenPrints():
            return analyze_activity_inner(i, outliers, apply_cpd, formcpd_type)


if __name__ == "__main__":
    # Install and import the R package 'gSeg' to be used in SoccerCPD
    utils = rpackages.importr("utils")
    utils.chooseCRANmirror(ind=1)
    if not rpackages.isinstalled("gSeg"):
        utils.install_packages("gSeg")
    rpackages.importr("gSeg")

    # Select the records of target matches (having data files but not analyzed yet)
    rm = RecordManager()
    activity_ids = [int(os.path.splitext(f)[0]) for f in os.listdir(DIR_UGP_DATA) if f.endswith(".ugp")]
    activity_records = rm.activity_records[
        (rm.activity_records[LABEL_DATA_SAVED] == 1) & (rm.activity_records[LABEL_STATS_SAVED] == 0)
    ]
    print()
    print("Activity Records:")
    print(activity_records)

    outliers = pd.read_csv("data/outliers.csv", header=0) if os.path.exists("data/outliers.csv") else None

    # Perform SoccerCPD per match using parallel processing
    # results = Parallel(n_jobs=50)(
    #     delayed(analyze_activity)(i, outliers, formcpd_type='rank', verbose=False)
    #     for i in tqdm(activity_records.index)
    # )
    # print(results)
    # for i, flag in results:
    #     activity_records.at[i, LABEL_STATS_SAVED] = flag
    # print(activity_records)

    # Perform SoccerCPD per match using for loop
    for i in activity_records.index:
        analyze_activity(i, outliers, apply_cpd=False, formcpd_type="gseg_union", verbose=True)

        # Set 'stats_saved' value for the match to 1 to avoid redundant executions
        rm.activity_records.at[i, LABEL_STATS_SAVED] = 1
        rm.save_records(VARNAME_ACTIVITY_RECORDS)