def get_season_slots():
slot_data = batch.get_data("stage1/MNCAATourneySlots.csv")
seasons = batch.get_data("stage1/MSeasons.csv")
#print(slot_data)
slot_season = pd.merge(slot_data,seasons,left_on=["Season"],right_on=["Season"])
slot_season = slot_season.drop(["DayZero"],axis=1)
#print(slot_season)
return slot_season
def load_top_level_files():
players = batch.get_data("MPlayers.csv")
print(players)
events15 = batch.get_data("MEvents2015.csv")
#events16 = batch.get_data("MEvents2016.csv")
#events17 = batch.get_data("MEvents2017.csv")
#events18 = batch.get_data("MEvents2018.csv")
#events19 = batch.get_data("MEvents2019.csv")
print(events15)
def show_compact_results():
season_compact_results = batch.get_data(
"stage1/MRegularSeasonCompactResults.csv")
print("<---------------- SEASON_COMPACT_RESULTS -------------->")
print(season_compact_results)
tourney_compact_results = batch.get_data(
"stage1/MNCAATourneyCompactResults.csv")
print("<---------------- TOURNEY_COMPACT_RESULTS -------------->")
print(tourney_compact_results)
def get_all_seed():
#TeamID1 Is Winning Seed
results_df = batch.get_data("stage1/MNCAATourneyCompactResults.csv")
seeds_df = batch.get_data("stage1/MNCAATourneySeeds.csv")
winning_seed = get_winning_seed()
losing_seed = get_losing_seed()
all_seed = pd.concat([winning_seed,losing_seed],sort=True)
#all_seed.to_csv("./data/seed_data-consolidated.csv",index=False)
return all_seed
def get_team_win_loss_data(write=False):
player_data = get_player_wins_losses(write)
team_data = batch.get_data("stage1/MTeamConferences.csv")
#These are teams ids from the player object
team_ids = player_data.TeamID.unique()
#Drop The Teams That Have No Wins or Losses In The Tourney
team_data = team_data[team_data["TeamID"].isin(team_ids)]
visited = []
data = []
for team_id in team_ids:
if team_id in visited:
continue
visited.append(team_id)
team_players = player_data[player_data["TeamID"] == team_id]
wins_stat = team_players["player_total_wins"].iloc[0]
losses_stat = team_players["player_total_losses"].iloc[0]
data.append({"TeamID":team_id,\
"tourney_game_wins":wins_stat,\
"tourney_game_losses":losses_stat\
})
team_win_loss_data = pd.DataFrame(data)
team_win_loss_data = pd.merge(team_data,team_win_loss_data,left_on=["TeamID","Season"],right_on=["TeamID","Season"])
if write:
team_win_loss_data.to_csv("./data/derived/MTeams.csv",index=False)
return team_win_loss_data
def get_player_wins_losses(write=False): #This returns the players df with total number of each player wins in the tourney
myDataWin = get_player_detail_seed_season_win()
winData = []
playerIDs = myDataWin.PlayerID.unique()
for playerid in playerIDs:
total_wins = myDataWin["PlayerID"].value_counts()[playerid]
winData.append({"PlayerID":playerid,\
"player_total_wins":total_wins,\
})
winners_df = (pd.DataFrame(winData))
myDataLoss = get_player_detail_seed_season_loss()
lossData = []
playerIDs = myDataLoss.PlayerID.unique()
for playerid in playerIDs:
total_losses = myDataLoss["PlayerID"].value_counts()[playerid]
lossData.append({"PlayerID":playerid,\
"player_total_losses":total_losses,\
})
losers_df = pd.DataFrame(lossData)
count_df = pd.merge(winners_df,losers_df,left_on=["PlayerID"],right_on=["PlayerID"])
player_data = batch.get_data("stage1/MPlayers.csv")
player_data = pd.merge(player_data,count_df,left_on=["PlayerID"],right_on=["PlayerID"])
if write:
player_data.to_csv("./data/derived/MPlayers.csv",index=False)
return player_data
def get_player_detail_seed_season_loss(): #Players id matched with winning team
player_data = batch.get_data("stage1/MPlayers.csv")
detail_seed_season_data = get_detail_seed_season()
#print(player_data)
detail_seed_player_season_data = pd.merge(player_data,detail_seed_season_data,left_on=["TeamID"],right_on=["LTeamID"])
#print(detail_seed_player_season_data)
return detail_seed_player_season_data
def get_detail_seed_season():
seasons = batch.get_data("stage1/MSeasons.csv")
seed_details = get_seed_detailed_results()
detail_seed_season = pd.merge(seasons,seed_details,left_on=["Season"],right_on=["Season"])
#print(detail_seed_season)
detail_seed_season = detail_seed_season.drop(["DayZero"],axis=1)
return detail_seed_season
def cut_preprocessed_data():
data = batch.get_data("master.csv")
result_ids = []
for index,row in data.iterrows():
team_1 = row["WTeamID"]
team_2 = row["LTeamID"]
season = row["Season"]
ied = str(season) + "_" + str(team_1) + "_" + str(team_2)
result_ids.append(ied)
data["result_id"] = result_ids
print(data)
predict_season = data[data.Season == 2019]
historical_data = data[data.Season < 2019]
print(predict_season)
print(historical_data)
print(historical_data.info())
predict_season.to_csv("data/test.csv")
historical_data.to_csv("data/train.csv")
return
def show_seasons():
seasons = batch.get_data("stage1/MSeasons.csv")
print("<--------------- SEASONS --------------->")
print(seasons)
print("<---------------- SEASON SEED DETAILS -------------->")
print(consolidation.get_detail_seed_season())
return
def development():
team_data = batch.get_data("stage1/MTeams.csv")
print(team_data)
seed_data = consolidation.get_seed_detailed_results()
print(seed_data)
combo_win = pd.merge(team_data,
seed_data,
left_on=["TeamID"],
right_on=["WTeamID"])
combo_loss = pd.merge(team_data,
seed_data,
left_on=["TeamID"],
right_on=["LTeamID"])
print(combo_win)
print(combo_loss)
wins = []
for index, row in combo_win.iterrows():
wins.append(1)
combo_win["result"] = wins
losses = []
for index, row in combo_loss.iterrows():
losses.append(0)
combo_loss["result"] = losses
combo = pd.concat([combo_win, combo_loss])
print(combo.info())
combo.to_csv("./data/master.csv", index=0)
return
def get_seed_compact_results():
all_seed = get_all_seed()
tourney_compact_data = batch.get_data("stage1/MNCAATourneyCompactResults.csv")
all_seed = all_seed.rename(mapper={"TeamID1":"WTeamID","TeamID2":"LTeamID"},axis=1)
#tourney_compact_data = tourney_compact_data.drop(["Season"],axis=1)
seed_compact = pd.merge(tourney_compact_data,all_seed,left_on=["Season","WTeamID","LTeamID"],right_on=["Season","WTeamID","LTeamID"])
seed_compact = seed_compact.rename(mapper={"Seed1":"WSeed","Seed2":"LSeed"},axis=1)
#print(seed_compact)
return seed_compact
def derive_team_data(write=False):
team_data = batch.get_data("derived/MTeams.csv")
slot_data = batch.get_data("stage1/MNCAATourneySlots.csv")
seed_data = get_seed_detailed_results()
print(slot_data)
print(seed_data)
print(team_data)
team_seeds = []
teams = team_data["TeamID"].unique()
for nTeam in teams:
team = seed_data[seed_data["WTeamID"] == nTeam]
team_seed = team.WSeed.iloc[0]
team_seeds.append({"TeamID":nTeam,\
"Seed":team_seed,})
team_seeds = pd.DataFrame(team_seeds)
team_data = pd.merge(team_data,team_seeds,left_on="TeamID",right_on="TeamID")
team_data.sort_values("Season",axis=0,ascending=True)
print(team_data)
return
def show_conference_tourney_games():
conference_tourney_games = batch.get_data(
"stage1/MConferenceTourneyGames.csv")
print("<-------------- CONFERENCE GAMES ---------------->")
print(conference_tourney_games)
def show_team_data():
print("<----------- TEAM DATA ---------->")
team_data = batch.get_data("stage1/MTeams.csv")
print(team_data)
def load_stage1_files():
cities = batch.get_data("stage1/Cities.csv")
print(cities)
conferences = batch.get_data("stage1/Conferences.csv")
print(conferences)
conference_tourney_games = batch.get_data(
"stage1/MConferenceTourneyGames.csv")
print(conference_tourney_games)
seasons = batch.get_data("stage1/MSeasons.csv")
print(seasons)
tourney_compact_results = batch.get_data(
"stage1/MNCAATourneyCompactResults.csv")
print(tourney_compact_results)
tourney_detailed_results = batch.get_data(
"stage1/MNCAATourneyDetailedResults.csv")
print(tourney_detailed_results)
season_compact_results = batch.get_data(
"stage1/MRegularSeasonCompactResults.csv")
print(season_compact_results)
season_detailed_results = batch.get_data(
"stage1/MRegularSeasonDetailedResults.csv")
print(season_detailed_results)
ordinals = batch.get_data("stage1/MMasseyOrdinals.csv")
print(ordinals)
seeds = batch.get_data("stage1/MNCAATourneySeeds.csv")
print(seeds)
seed_round_slots = batch.get_data("stage1/MNCAATourneySeedRoundSlots.csv")
print(seed_round_slots)
def show_slot_data():
slot_data = batch.get_data("stage1/MNCAATourneySlots.csv")
print("<--------------- SLOTS --------------->")
print(slot_data)
print("<--------------- SEASON SLOTS --------------->")
print(consolidation.get_season_slots())
#!/usr/bin/env python3
import pandas as pd
import refine, display
import batch_handler as batch
import sklearn
import pandas as pd
import numpy as np
results_df = batch.get_data("stage1/MNCAATourneyCompactResults.csv")
seeds_df = batch.get_data("stage1/MNCAATourneySeeds.csv")
"""
losing seed is seed 2
"""
def get_winning_seed():
winning_seed = pd.merge(results_df,seeds_df,left_on=["Season","WTeamID"],right_on=["Season","TeamID"])
winning_seed = winning_seed.drop(["TeamID"],axis=1)
winning_seed = winning_seed.rename(mapper={"Seed": "Seed1", "WTeamID": "TeamID1"},axis=1)
winning_seed["Result"] = 1
winning_seed = pd.merge(winning_seed,seeds_df,left_on=["Season","LTeamID"],right_on=["Season","TeamID"])
winning_seed = winning_seed.drop(["TeamID"],axis=1)
winning_seed = winning_seed.rename(mapper={"Seed": "Seed2", "LTeamID": "TeamID2"},axis=1)
