wsba-hockey 0.1.0__tar.gz → 0.1.2__tar.gz

{wsba_hockey-0.1.0/src/wsba_hockey.egg-info → wsba_hockey-0.1.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: wsba_hockey
-Version: 0.1.0
+Version: 0.1.2
 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/
@@ -62,8 +62,15 @@ wsba.nhl_scrape_player_info(wsba.nhl_scrape_roster('20242025'))
 
 ## REPOSITORY 
 ### Past Season Play-by-Play*
-### Team Information*
-### Schedule*
+### Team Information
+```python
+wsba.repo_load_teaminfo()
+wsba.repo_load_rosters(seasons=['20212022','20222023','20232024','20242025])
+```
+### Schedule
+```python
+wsba.repo_load_schedule(seasons=['20212022','20222023','20232024','20242025])
+```
 
 ## ACKNOWLEDGEMENTS AND CREDITS 
 ### Huge thanks to the following:

{wsba_hockey-0.1.0 → wsba_hockey-0.1.2}/README.md

@@ -47,8 +47,15 @@ wsba.nhl_scrape_player_info(wsba.nhl_scrape_roster('20242025'))
 
 ## REPOSITORY 
 ### Past Season Play-by-Play*
-### Team Information*
-### Schedule*
+### Team Information
+```python
+wsba.repo_load_teaminfo()
+wsba.repo_load_rosters(seasons=['20212022','20222023','20232024','20242025])
+```
+### Schedule
+```python
+wsba.repo_load_schedule(seasons=['20212022','20222023','20232024','20242025])
+```
 
 ## ACKNOWLEDGEMENTS AND CREDITS 
 ### Huge thanks to the following:

{wsba_hockey-0.1.0 → wsba_hockey-0.1.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "wsba_hockey"
-version = "0.1.0"
+version = "0.1.2"
 authors = [
   { name="Owen Singh", email="owenbksingh@gmail.com" },
 ]

wsba_hockey-0.1.2/src/wsba_hockey/__init__.py

@@ -0,0 +1 @@
+from wsba_hockey.wsba_main import nhl_scrape_game,nhl_scrape_schedule,nhl_scrape_season,nhl_scrape_seasons_info,nhl_scrape_standings,nhl_scrape_roster,nhl_scrape_player_info,repo_load_rosters,repo_load_schedule,repo_load_teaminfo

{wsba_hockey-0.1.0 → wsba_hockey-0.1.2}/src/wsba_hockey/tools/scraping.py

@@ -1,13 +1,9 @@
 import re
-from bs4 import BeautifulSoup, SoupStrainer
+from bs4 import BeautifulSoup
 import hockey_scraper.utils.shared as shared
-import hockey_scraper.nhl.pbp.html_pbp as html
-import hockey_scraper.nhl.game_scraper as gs
 import numpy as np
 import pandas as pd
 import warnings
-import requests as rs
-from zipfile import ZipFile
 warnings.filterwarnings('ignore')
 
 ### SCRAPING FUNCTIONS ###
@@ -68,6 +64,10 @@ def parse_json(json):
     info = pd.json_normalize(json)
     roster = pd.json_normalize(json['rosterSpots'])
 
+    #Return error if game is set in the future
+    if info['gameState'][0] == 'FUT':
+        raise ValueError(f"Game {info['id'][0]} has not occured yet.")
+
     #Game information
     events['game_id'] = info['id'][0]
     events['season'] = info['season'][0]
@@ -127,12 +127,25 @@ def parse_json(json):
     # x, y - Raw coordinates from JSON pbpp
     # x_fixed, y_fixed - Coordinates fixed to the right side of the ice (x is always greater than 0)
     # x_adj, y_adj - Adjusted coordinates configuring away events with negative x vlaues while home events are always positive
-    events['x_fixed'] = abs(events['details.xCoord'])
-    events['y_fixed'] = np.where(events['details.xCoord']<0,-events['details.yCoord'],events['details.yCoord'])
-    events['x_adj'] = np.where(events['event_team_status']=="home",events['x_fixed'],-events['x_fixed'])
-    events['y_adj'] = np.where(events['event_team_status']=="home",events['y_fixed'],-events['y_fixed'])
-    events['event_distance'] = np.sqrt(((89 - events['x_fixed'])**2) + (events['y_fixed']**2))
-    events['event_angle'] = np.degrees(np.arctan2(abs(events['y_fixed']), abs(89 - events['x_fixed'])))
+    
+    #Some games (mostly preseason and all star games) do not include coordinates.  
+    try:
+        events['x_fixed'] = abs(events['details.xCoord'])
+        events['y_fixed'] = np.where(events['details.xCoord']<0,-events['details.yCoord'],events['details.yCoord'])
+        events['x_adj'] = np.where(events['event_team_status']=="home",events['x_fixed'],-events['x_fixed'])
+        events['y_adj'] = np.where(events['event_team_status']=="home",events['y_fixed'],-events['y_fixed'])
+        events['event_distance'] = np.sqrt(((89 - events['x_fixed'])**2) + (events['y_fixed']**2))
+        events['event_angle'] = np.degrees(np.arctan2(abs(events['y_fixed']), abs(89 - events['x_fixed'])))
+    except TypeError:
+        print(f"No coordinates found for game {info['id'][0]}...")
+    
+        events['x_fixed'] = np.nan
+        events['y_fixed'] = np.nan
+        events['x_adj'] = np.nan
+        events['y_adj'] = np.nan
+        events['event_distance'] = np.nan
+        events['event_angle'] = np.nan
+    
     
     events['event_team_abbr'] = events['details.eventOwnerTeamId'].replace(teams)
 
@@ -517,6 +530,23 @@ def fix_names(shifts_df,json):
     
     return shifts_df.replace(replace,regex=True)
 
+def get_col():
+    return [
+        'season','season_type','game_id','game_date',"start_time","venue","venue_location",
+        'away_team_abbr','home_team_abbr','event_num','period','period_type',
+        'seconds_elapsed', "situation_code","strength_state","home_team_defending_side","shift_type",
+        "event_type_code","event_type","description","reason","penalty_duration","penalty_description",
+        "event_team_abbr",'num_on', 'players_on', 'ids_on', 'num_off', 'players_off', 'ids_off',
+        "event_team_status","event_player_1_id","event_player_2_id","event_player_3_id",
+        "event_player_1_name","event_player_2_name","event_player_3_name","event_player_1_pos","event_player_2_pos",
+        "event_player_3_pos","event_goalie_id",
+        "event_goalie_name","shot_type","zone_code","x","y","x_fixed","y_fixed","x_adj","y_adj",
+        "event_skaters","away_skaters","home_skaters",
+        "event_distance","event_angle","away_score","home_score", "away_fenwick", "home_fenwick",
+        "away_on_1","away_on_2","away_on_3","away_on_4","away_on_5","away_on_6","away_goalie",
+        "home_on_1","home_on_2","home_on_3","home_on_4","home_on_5","home_on_6","home_goalie"
+    ]
+
 def combine_data(json,html):
     #Given json pbp and html shifts, total game play-by-play data is provided with additional and corrected details
     df = pd.concat([json,html])
@@ -604,22 +634,6 @@ def combine_data(json,html):
 
     df['strength_state'] = df['event_skaters'].astype(str) + "v" + df['event_skaters_against'].astype(str)
     df['situation_code'] = np.where(df['situation_code'].isna(),df['away_goalie_in'].astype(str) + df['away_skaters'].astype(str) + df['home_skaters'].astype(str) + df['home_goalie_in'].astype(str),df['situation_code'])
-    
-    col = [
-        'season','season_type','game_id','game_date',"start_time","venue","venue_location",
-        'away_team_abbr','home_team_abbr','event_num','period','period_type',
-        'seconds_elapsed', "situation_code","strength_state","home_team_defending_side","shift_type",
-        "event_type_code","event_type","description","reason","penalty_duration","penalty_description",
-        "event_team_abbr",'num_on', 'players_on', 'ids_on', 'num_off', 'players_off', 'ids_off',
-        "event_team_status","event_player_1_id","event_player_2_id","event_player_3_id",
-        "event_player_1_name","event_player_2_name","event_player_3_name","event_player_1_pos","event_player_2_pos",
-        "event_player_3_pos","event_goalie_id",
-        "event_goalie_name","shot_type","zone_code","x","y","x_fixed","y_fixed","x_adj","y_adj",
-        "event_skaters","away_skaters","home_skaters",
-        "event_distance","event_angle","away_score","home_score", "away_fenwick", "home_fenwick",
-        "away_on_1","away_on_2","away_on_3","away_on_4","away_on_5","away_on_6","away_goalie",
-        "home_on_1","home_on_2","home_on_3","home_on_4","home_on_5","home_on_6","home_goalie"
-    ]
 
     #Return: complete play-by-play with all important data for each event in a provided game
-    return df[col].replace(r'^\s*$', np.nan, regex=True)
+    return df[get_col()].replace(r'^\s*$', np.nan, regex=True)

wsba_hockey-0.1.2/src/wsba_hockey/tools/xg_model.py

@@ -0,0 +1,146 @@
+import pandas as pd
+import numpy as np
+import xgboost as xgb
+import scipy.sparse as sp
+import joblib
+
+### XG_MODEL FUNCTIONS ###
+# Provided in this file are functions vital to the goal prediction model in the WSBA Hockey Python package. #
+
+def prep_xG_data(pbp):
+    #Prep data for xG training and calculation
+
+    events = ['faceoff','hit','giveaway','takeaway','blocked-shot','missed-shot','shot-on-goal','goal']
+    shot_types = ['wrist','deflected','tip-in','slap','backhand','snap','wrap-around','poke','bat','cradle','between-legs']
+    fenwick_events = ['missed-shot','shot-on-goal','goal']
+    strengths = ['3v3',
+                '3v4',
+                '3v5',
+                '4v3',
+                '4v4',
+                '4v5',
+                '4v6',
+                '5v3',
+                '5v4',
+                '5v5',
+                '5v6',
+                '6v4',
+                '6v5']
+    
+    #Filter unwanted date:
+    #Shots must occur in specified events and strength states, occur before the shootout, and have valid coordinates
+    data = pbp.loc[(pbp['event_type'].isin(events))&
+                   (pbp['strength_state'].isin(strengths))&
+                   (pbp['period'] < 5)&
+                   (pbp['x_fixed'].notna())&
+                   (pbp['y_fixed'].notna())&
+                   ~((pbp['x_fixed']==0)&(pbp['y_fixed']==0)&(pbp['x_fixed'].isin(fenwick_events))&(pbp['event_distance']!=90))]
+    #Create last event columns
+    data = data.sort_values(by=['season','game_id','period','seconds_elapsed','event_num'])
+
+    data["seconds_since_last"] = data['seconds_elapsed']-data['seconds_elapsed'].shift(1)
+    data["event_team_last"] = data['event_team_abbr'].shift(1)
+    data["event_type_last"] = data['event_type'].shift(1)
+    data["x_fixed_last"] = data['x_fixed'].shift(1)
+    data["y_fixed_last"] = data['y_fixed'].shift(1)
+    data["zone_code_last"] = data['zone_code'].shift(1)
+    data['shot_type'] = data['shot_type'].fillna('wrist')
+    
+
+    data.sort_values(['season','game_id','period','seconds_elapsed','event_num'],inplace=True)
+    data['score_state'] = np.where(data['away_team_abbr']==data['event_team_abbr'],data['away_score']-data['home_score'],data['home_score']-data['away_score'])
+    data['fenwick_state'] = np.where(data['away_team_abbr']==data['event_team_abbr'],data['away_fenwick']-data['home_fenwick'],data['home_fenwick']-data['away_fenwick'])
+    data['distance_from_last'] = np.sqrt((data['x_fixed'] - data['x_fixed_last'])**2 + (data['y_fixed'] - data['y_fixed_last'])**2)
+    data['rush_mod'] = np.where((data['event_type'].isin(fenwick_events))&(data['zone_code_last'].isin(['N','D']))&(data['x_fixed']>25)&(data['seconds_since_last']<5),5-data['seconds_since_last'],0)
+    data['rebound_mod'] = np.where((data['event_type'].isin(fenwick_events))&(data['event_type_last'].isin(fenwick_events))&(data['seconds_since_last']<3),3-data['seconds_since_last'],0)
+
+    #Create boolean variables
+    data["is_goal"]=(data['event_type']=='goal').astype(int)
+    data["is_home"]=(data['home_team_abbr']==data['event_team_abbr']).astype(int)
+
+
+    for shot in shot_types:
+        data[shot] = (data['shot_type']==shot).astype(int)
+    for strength in strengths:
+        data[f'state_{strength}'] = (data['strength_state']==strength).astype(int)
+    for event in events[0:len(events)-1]:
+        data[f'prior_{event}_same'] = ((data['event_type_last']==event)&(data['event_team_last']==data['event_team_abbr'])).astype(int)
+        data[f'prior_{event}_opp'] = ((data['event_type_last']==event)&(data['event_team_last']!=data['event_team_abbr'])).astype(int)
+    
+    #Return: pbp data prepared to train and calculate the xG model
+    return data
+
+def wsba_xG(pbp, train = False, overwrite = False, model_path = "tools/xg_model/wsba_xg.joblib", train_runs = 20, test_runs = 20):
+    #Train and calculate the WSBA Expected Goals model
+    
+    target = "is_goal"
+    continous = ['event_distance',
+                'event_angle',
+                'seconds_elapsed',
+                'period',
+                'x_fixed',
+                'y_fixed',
+                'x_fixed_last',
+                'y_fixed_last',
+                'distance_from_last',
+                'seconds_since_last',
+                'score_state',
+                'fenwick_state',
+                'rush_mod',
+                'rebound_mod']
+    boolean = ['is_home',
+            'state_3v3',
+            'state_3v4',
+            'state_3v5',
+            'state_4v3',
+            'state_4v4',
+            'state_4v5',
+            'state_4v6',
+            'state_5v3',
+            'state_5v4',
+            'state_5v5',
+            'state_5v6',
+            'state_6v4',
+            'state_6v5',
+            'wrist',
+            'deflected',
+            'tip-in',
+            'slap',
+            'backhand',
+            'snap',
+            'wrap-around',
+            'poke',
+            'bat',
+            'cradle',
+            'between-legs',
+            'prior_shot-on-goal_same',
+            'prior_missed-shot_same',
+            'prior_blocked-shot_same',
+            'prior_giveaway_same',
+            'prior_takeaway_same',
+            'prior_hit_same',
+            'prior_shot-on-goal_opp',
+            'prior_missed-shot_opp',
+            'prior_blocked-shot_opp',
+            'prior_giveaway_opp',
+            'prior_takeaway_opp',
+            'prior_hit_opp',
+            'prior_faceoff']
+    
+    #Prep Data
+    data = prep_xG_data(pbp)
+
+    #Convert to sparse
+    data_sparse = sp.csr_matrix(data[[target]+continous+boolean])
+
+    #Target and Predictors
+    is_goal_vect = data_sparse[:, 0].A
+    predictors = data_sparse[:, 1:]
+
+    #XGB DataModel
+    xgb_matrix = xgb.DMatrix(data=predictors,label=is_goal_vect)
+
+    if train == True:
+        run_num = 
+    else:
+        print("No data to add yet...")

{wsba_hockey-0.1.0 → wsba_hockey-0.1.2}/src/wsba_hockey/wsba_main.py

@@ -2,7 +2,7 @@ import requests as rs
 import pandas as pd
 import numpy as np
 from datetime import datetime, timedelta
-from tools.scraping import *
+from wsba_hockey.tools.scraping import *
 
 ### WSBA HOCKEY ###
 ## Provided below are all integral functions in the WSBA Hockey Python package. ##
@@ -26,30 +26,43 @@ def nhl_scrape_game(game_ids,split_shifts = False,remove = ['period-start','peri
         away_log = "https://www.nhl.com/scores/htmlreports/"+season+"/TV"+str(game_id)[-6:]+".HTM"
 
         #Retrieve raw data
-        json = rs.get(api).json()
-        home_shift = rs.get(home_log).content
-        away_shift = rs.get(away_log).content
-
-        if int(game_id[:4]) < 2010:
-            print()
-            raise Exception('Games before 2010-2011 are not available yet.')
-        else:
-            #Parse Json
-            pbp = parse_json(json) 
-        
-        #Create shifts
-        shifts = fix_names(combine_shifts(home_shift,away_shift,json,game_id),json)
-
-        #Combine and append data to list
-        data = combine_data(pbp,shifts)
+        try: 
+            json = rs.get(api).json()
+            home_shift = rs.get(home_log).content
+            away_shift = rs.get(away_log).content
+
+            if int(game_id[:4]) < 2010:
+                print()
+                raise Exception('Games before 2010-2011 are not available yet.')
+            else:
+                #Parse Json
+                pbp = parse_json(json) 
+            
+            #Create shifts
+            #If no shifts data exists only export play-by-play
+            try: 
+                shifts = fix_names(combine_shifts(home_shift,away_shift,json,game_id),json)
+                data = combine_data(pbp,shifts)
+            
+            except:
+                print(f"Cannot find or create shifts for game {game_id}...")
+                data = combine_data(pbp,pd.DataFrame(columns=get_col()))
 
-        pbps.append(data)
+            #Combine and append data to list
+            pbps.append(data)
+        except:
+            print(f"Unable to scrape game {game_id}.  Ensure the ID is properly inputted and formatted.")
+            pbps.append(pd.DataFrame())
 
     #Add all pbps together
     df = pd.concat(pbps)
 
     #Split pbp and shift events if necessary
     #Return: complete play-by-play with data removed or split as necessary
+    try: df['event_type']
+    except KeyError:
+        raise KeyError("No data is available to return.")
+
     if split_shifts == True:
         if len(remove) == 0:
             remove = ['change']
@@ -113,7 +126,7 @@ def nhl_scrape_schedule(season,start = "09-01", end = "08-01"):
     #Return: specificed schedule data (excluding preseason games)
     return df.loc[df['season_type']>1]
 
-def nhl_scrape_season(season,split_shifts = False, remove = ['period-start','period-end','challenge','stoppage','change'], start = "09-01", end = "08-01", local=False, local_path = "schedule/schedule.csv"):
+def nhl_scrape_season(season,split_shifts = False, remove = ['period-start','period-end','game-end','challenge','stoppage'], start = "09-01", end = "08-01", local=False, local_path = "schedule/schedule.csv"):
     #Given season, scrape all play-by-play occuring within the season
     # param 'season' - NHL season to scrape
     # param 'split_shifts' - boolean which splits pbp and shift events if true
@@ -146,29 +159,26 @@ def nhl_scrape_season(season,split_shifts = False, remove = ['period-start','per
                 df.append(data)
 
         except: 
-            #Errors should be rare; testing of eight full-season scraped produced just one missing game due to erro
-            #Games which have not happened yet also print as errors
-            print("An error occurred...")
-            errors.append(pd.DataFrame({"id":game_id}))
-    
-    pbp = pd.concat(df)
+            #Errors should be rare; testing of eight full-season scraped produced just one missing regular season game due to error
+            continue
+
+    #Missing data handled as a KeyError
+    try: pbp = pd.concat(df)
+    except: 
+        raise KeyError("No data is available to return.")
+        
     if split_shifts == True:
-        shifts = pd.concat(df_s)
+        try: shifts = pd.concat(df_s)
+        except: raise KeyError("No data is available to return.")
     else:
         ""
-    try: 
-        errors = pd.concat(errors)
-    except:
-        errors = pd.DataFrame()
 
     #Return: Complete pbp and shifts data for specified season as well as dataframe of game_ids which failed to return data
     if split_shifts == True:
         return {"pbp":pbp,
-            'shifts':shifts,
-            "errors":errors}
+            'shifts':shifts}
     else:
-        return {"pbp":pbp,
-            "errors":errors}
+        return pbp
 
 def nhl_scrape_seasons_info(seasons = []):
     #Returns info related to NHL seasons (by default, all seasons are included)
@@ -192,7 +202,7 @@ def nhl_scrape_seasons_info(seasons = []):
 
 def nhl_scrape_standings(arg = "now"):
     #Returns standings
-    # parma 'arg' - by default, this is "now" returning active NHL standings.  May also be a specific date formatted as YYYY-MM-DD
+    # param 'arg' - by default, this is "now" returning active NHL standings.  May also be a specific date formatted as YYYY-MM-DD
     
     if arg == "now":
         print("Scraping standings as of now...")
@@ -279,4 +289,29 @@ def nhl_scrape_player_info(roster):
     players['Season'] = players['Season'].astype(str)
     players['Player'] = players['Player'].replace(r'^\s*$', np.nan, regex=True)
 
-    return players.loc[players['Player'].notna()].sort_values(by=['Player','Season','Team'])
+    return players.loc[players['Player'].notna()].sort_values(by=['Player','Season','Team'])
+
+def repo_load_rosters(seasons = []):
+    #Returns roster data from repository
+    # param 'seasons' - list of seasons to include
+
+    data = pd.read_csv("rosters/nhl_rosters.csv")
+    if len(seasons)>0:
+        data = data.loc[data['season'].isin(seasons)]
+
+    return data
+
+def repo_load_schedule(seasons = []):
+    #Returns schedule data from repository
+    # param 'seasons' - list of seasons to include
+
+    data = pd.read_csv("schedule/schedule.csv")
+    if len(seasons)>0:
+        data = data.loc[data['season'].isin(seasons)]
+
+    return data
+
+def repo_load_teaminfo():
+    #Returns team data from repository
+
+    return pd.read_csv("teaminfo/nhl_teaminfo.csv")

{wsba_hockey-0.1.0 → wsba_hockey-0.1.2/src/wsba_hockey.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: wsba_hockey
-Version: 0.1.0
+Version: 0.1.2
 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/
@@ -62,8 +62,15 @@ wsba.nhl_scrape_player_info(wsba.nhl_scrape_roster('20242025'))
 
 ## REPOSITORY 
 ### Past Season Play-by-Play*
-### Team Information*
-### Schedule*
+### Team Information
+```python
+wsba.repo_load_teaminfo()
+wsba.repo_load_rosters(seasons=['20212022','20222023','20232024','20242025])
+```
+### Schedule
+```python
+wsba.repo_load_schedule(seasons=['20212022','20222023','20232024','20242025])
+```
 
 ## ACKNOWLEDGEMENTS AND CREDITS 
 ### Huge thanks to the following:

{wsba_hockey-0.1.0 → wsba_hockey-0.1.2}/src/wsba_hockey.egg-info/SOURCES.txt

@@ -1,9 +1,11 @@
 LICENSE
 README.md
 pyproject.toml
+src/wsba_hockey/__init__.py
 src/wsba_hockey/wsba_main.py
 src/wsba_hockey.egg-info/PKG-INFO
 src/wsba_hockey.egg-info/SOURCES.txt
 src/wsba_hockey.egg-info/dependency_links.txt
 src/wsba_hockey.egg-info/top_level.txt
-src/wsba_hockey/tools/scraping.py
+src/wsba_hockey/tools/scraping.py
+src/wsba_hockey/tools/xg_model.py