wsba-hockey 1.2.0__tar.gz → 1.2.2__tar.gz

{wsba_hockey-1.2.0/src/wsba_hockey.egg-info → wsba_hockey-1.2.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: wsba_hockey
-Version: 1.2.0
+Version: 1.2.2
 Summary: WeakSide Breakout's complete Python package of access to hockey data, primairly including the scraping of National Hockey League schedule, play-by-play, and shifts information.
 Author-email: Owen Singh <owenbksingh@gmail.com>
 Project-URL: Homepage, https://github.com/owensingh38/wsba_hockey/

{wsba_hockey-1.2.0 → wsba_hockey-1.2.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "wsba_hockey"
-version = "1.2.0"
+version = "1.2.2"
 authors = [
   { name="Owen Singh", email="owenbksingh@gmail.com" },
 ]

{wsba_hockey-1.2.0 → wsba_hockey-1.2.2}/src/wsba_hockey/__init__.py

@@ -7,6 +7,7 @@ from wsba_hockey.wsba_main import (
     nhl_scrape_roster,
     nhl_scrape_draft_rankings,
     nhl_scrape_prospects,
+    nhl_scrape_team_info,
     nhl_scrape_game_info,
     nhl_calculate_stats,
     nhl_apply_xG,

wsba_hockey-1.2.2/src/wsba_hockey/tools/game_pred.py

@@ -0,0 +1,33 @@
+import joblib
+import os
+import pandas as pd
+import numpy as np
+import xgboost as xgb
+import scipy.sparse as sp
+import wsba_hockey.wsba_main as wsba
+import wsba_hockey.tools.scraping as scraping
+import matplotlib.pyplot as plt
+from sklearn.calibration import calibration_curve
+from sklearn.metrics import roc_curve, auc
+
+### GAME PREDICTION MODEL FUNCTIONS ###
+# Provided in this file are functions vital to the game prediction model in the WSBA Hockey Python package. #
+
+## GLOBAL VARIABLES ##
+dir = os.path.dirname(os.path.realpath(__file__))
+roster_path = os.path.join(dir,'rosters\\nhl_rosters.csv')
+schedule_path = os.path.join(dir,'schedule/schedule.csv')
+
+def prep_game_data(pbp):
+    #Prepare schedule data for model development given full-season pbp
+
+    #Calculate necessary team stats (by game) for the prediction model
+    #The model will evaluate based on three different qualities for valid EV, PP, and SH strength 
+    dfs = []
+    for strength in [['5v5'],['5v4'],['4v5']]:
+        team_games = wsba.nhl_calculate_stats(pbp,'team',[2,3],strength,True)
+        team_games['Year'] = team_games['Season'].str[0:4].astype(int)
+        dfs.append(team_games)
+
+    #Place the games in order and create sums for 
+    df = pd.concat(dfs).sort_values(by=['Year','Game'])

{wsba_hockey-1.2.0 → wsba_hockey-1.2.2}/src/wsba_hockey/wsba_main.py

@@ -100,14 +100,12 @@ KNOWN_PROBS = {
     2009020885:'Missing shifts data for game between Sharks and Blue Jackets.',
     2010020124:'Game between Capitals and Hurricanes is sporadically missing player on-ice data',
     2012020018:'HTML events contain mislabeled events.',
-    2013020971:'On March 10th, 2014, Stars forward Rich Peverley suffered from a cardiac episode midgame and as a result, the remainder of the game was postponed.  \nThe game resumed on April 9th, and the only goal scorer in the game, Blue Jackets forward Nathan Horton, did not appear in the resumed game due to injury.  Interestingly, Horton would never play in the NHL again.',
     2018021133:'Game between Lightning and Capitals has incorrectly labeled event teams (i.e. WSH TAKEAWAY - #71 CIRELLI (Cirelli is a Tampa Bay skater in this game)).',
-    2019020876:'Due to the frightening collapse of Blues defensemen Jay Bouwmeester, a game on February 2nd, 2020 between the Ducks and Blues was postponed.  \nWhen the game resumed, Ducks defensemen Hampus Lindholm, who assisted on a goal in the inital game, did not play in the resumed match.'
 }
 
 SHOT_TYPES = ['wrist','deflected','tip-in','slap','backhand','snap','wrap-around','poke','bat','cradle','between-legs']
 
-NEW = 2024
+NEW = 2025
 
 EVENTS = ['faceoff','hit','giveaway','takeaway','blocked-shot','missed-shot','shot-on-goal','goal','penalty']
 
@@ -461,14 +459,14 @@ def nhl_scrape_seasons_info(seasons:list[int] = []):
     else:
         return df.sort_values(by=['id'])
 
-def nhl_scrape_standings(arg:str | int = "now", season_type:int = 2):
+def nhl_scrape_standings(arg:int | list[int] | Literal['now'] = 'now', season_type:int = 2):
     """
     Returns standings or playoff bracket
     Args:
-        arg (str or int, optional):
-            Date formatted as 'YYYY-MM-DD' to scrape standings, NHL season such as "20242025", or 'now' for current standings. Default is 'now'.
+        arg (int or list[int] or str, optional):
+            Date formatted as 'YYYY-MM-DD' to scrape standings, NHL season such as "20242025", list of NHL seasons, or 'now' for current standings. Default is 'now'.
         season_type (int, optional):
-            Part of season to scrape.  If 3 (playoffs) then scrape the playoff bracket for the season implied by arg. When arg = 'now' this is ignored. Default is 2.
+            Part of season to scrape.  If 3 (playoffs) then scrape the playoff bracket for the season implied by arg. When arg = 'now' this is defaulted to the most recent playoff year.  Any dates passed through are parsed as seasons. Default is 2.
 
     Returns:
         pd.DataFrame: 
@@ -477,31 +475,59 @@ def nhl_scrape_standings(arg:str | int = "now", season_type:int = 2):
 
     if season_type == 3:
         if arg == "now":
-            arg = NEW
+            arg = [NEW]
+        elif type(arg) == int:
+            #Find year from season
+            arg = [str(arg)[4:8]]
+        elif type(arg) == list:
+            #Find year from seasons
+            arg = [str(s)[4:8] for s in arg]
+        else:
+            #Find year from season from date
+            arg = [int(arg[0:4])+1 if (9 < int(arg[5:7]) < 13) else int(arg[0:4])]
 
-        print(f"Scraping playoff bracket for date: {arg}")
-        api = f"https://api-web.nhle.com/v1/playoff-bracket/{arg}"
+        print(f"Scraping playoff bracket for season{'s' if len(arg)>1 else ''}: {arg}")
+        
+        dfs = []
+        for season in arg:
+            api = f"https://api-web.nhle.com/v1/playoff-bracket/{season}"
 
-        data = rs.get(api).json()['series']
+            data = rs.get(api).json()['series']
+            dfs.append(pd.json_normalize(data))
 
-        return pd.json_normalize(data)
+        #Return: playoff bracket
+        return pd.concat(dfs)
 
     else:
         if arg == "now":
             print("Scraping standings as of now...")
+            arg = [arg]
         elif arg in SEASONS:
             print(f'Scraping standings for season: {arg}')
+            arg = [arg]
+        elif type(arg) == list:
+            print(f'Scraping standings for seasons: {arg}')
         else:
             print(f"Scraping standings for date: {arg}")
+            arg = [arg]
+
+        dfs = []
+        for search in arg:
+            #If the end is an int then its a season otherwise it is either 'now' or a date as a string
+            if type(search) == int:
+                season_data = rs.get('https://api.nhle.com/stats/rest/en/season').json()['data']
+                season_data = [s for s in season_data if s['id'] == search][0]
+                end = season_data['regularSeasonEndDate'][0:10]
+            else:
+                end = search
+                
+            api = f"https://api-web.nhle.com/v1/standings/{end}"
 
-        season_data = rs.get('https://api.nhle.com/stats/rest/en/season').json()['data']
-        season_data = [s for s in season_data if s['id'] == arg][0]
-        end = season_data['regularSeasonEndDate'][0:10]
-
-        api = f"https://api-web.nhle.com/v1/standings/{end}"
-        data = rs.get(api).json()['standings']
+            data = rs.get(api).json()['standings']
+            dfs.append(pd.json_normalize(data))
 
-        return pd.json_normalize(data)
+        #Return: standings data
+        return pd.concat(dfs)
 
 def nhl_scrape_roster(season: int):
     """
@@ -629,7 +655,7 @@ def nhl_scrape_player_data(player_ids:list[int]):
     else:
         return pd.DataFrame()
 
-def nhl_scrape_draft_rankings(arg:str = 'now', category:int = 0):
+def nhl_scrape_draft_rankings(arg:str | Literal['now'] = 'now', category:int = 0):
     """
     Returns draft rankings
     Args:
@@ -1165,7 +1191,7 @@ def nhl_calculate_stats(pbp:pd.DataFrame, type:Literal['skater','goalie','team']
                                             'birthCountry':'Nationality'})
         
         #WSBA
-        complete['WSBA'] = complete['Goalie']+complete['Team']+complete['Season'].astype(str)
+        complete['WSBA'] = complete['ID'].astype(str).str.replace('.0','')+complete['Team']+complete['Season'].astype(str)
 
         #Add player age
         complete['Birthday'] = pd.to_datetime(complete['Birthday'])
@@ -1306,7 +1332,7 @@ def nhl_calculate_stats(pbp:pd.DataFrame, type:Literal['skater','goalie','team']
         #Remove goalies that occasionally appear in a set
         complete = complete.loc[complete['Position']!='G']
         #Add WSBA ID
-        complete['WSBA'] = complete['Player']+complete['Season'].astype(str)+complete['Team']
+        complete['WSBA'] = complete['ID'].astype(str).str.replace('.0','')+complete['Season'].astype(str)+complete['Team']
 
         #Add per 60 stats
         for stat in PER_SIXTY:

{wsba_hockey-1.2.0 → wsba_hockey-1.2.2/src/wsba_hockey.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: wsba_hockey
-Version: 1.2.0
+Version: 1.2.2
 Summary: WeakSide Breakout's complete Python package of access to hockey data, primairly including the scraping of National Hockey League schedule, play-by-play, and shifts information.
 Author-email: Owen Singh <owenbksingh@gmail.com>
 Project-URL: Homepage, https://github.com/owensingh38/wsba_hockey/

{wsba_hockey-1.2.0 → wsba_hockey-1.2.2}/src/wsba_hockey.egg-info/SOURCES.txt

@@ -9,6 +9,7 @@ src/wsba_hockey.egg-info/dependency_links.txt
 src/wsba_hockey.egg-info/top_level.txt
 src/wsba_hockey/tools/__init__.py
 src/wsba_hockey/tools/agg.py
+src/wsba_hockey/tools/game_pred.py
 src/wsba_hockey/tools/plotting.py
 src/wsba_hockey/tools/scraping.py
 src/wsba_hockey/tools/xg_model.py