clean_data.py

import sys
import pandas as pd
import numpy as np
#splits the game logs by seasons and removes players who played < 20 games last season. Also removes unrelated stats

columns = ['First Assists','Second Assists','IPP','Rebounds Created','PIM','Total Penalties','Minor','Major','Misconduct','Penalties Drawn','Giveaways','Takeaways','Hits',
'Hits Taken','Shots Blocked','Faceoffs Won','Faceoffs Lost','Faceoffs %','Off. Zone Faceoffs','Neu. Zone Faceoffs','Def. Zone Faceoffs', 'Off. Zone Faceoff %',
'CA','CF%','FA','FF%','SA','SF%','GA','GF%','xGA','xGF%','SCA','SCF%','SCGA','SCGF%','HDGA','HDGF%','On-Ice SV%','PDO','Neu. Zone Starts','Def. Zone Starts','On The Fly Starts','Off. Zone Start %']

data = pd.read_pickle(sys.argv[1])
data.index = data.index.set_levels([data.index.levels[0], pd.to_datetime(data.index.levels[1])])
data = data.drop(columns = columns)
data.replace([np.inf, -np.inf], np.nan, inplace=True)
data = data.fillna(0)
last_season = data[data.index.get_level_values('Date') < "2023-6-20"]
current_season = data[data.index.get_level_values('Date') > "2023-6-20"]

game_counts = last_season.groupby(['Id']).size().to_frame('size')
game_counts = game_counts[game_counts['size'] <= 20]

last_season = last_season.drop(index = game_counts.index, level='Id')

last_season.to_pickle("data/last_season_game_logs.pkl")
current_season.to_pickle("data/current_season_game_logs.pkl")