load in data from csv now operational

nba_tracking_data_15_16.py

@@ -22,6 +22,8 @@ import py7zr
 import datasets
 import requests
 
+import pandas as pd
+
 
 _CITATION = """\
 @misc{Linou2016,
@@ -39,6 +41,7 @@ By merging all .7z files into one large .json file, access is easier to retrieve
 
 _HOMEPAGE = "https://github.com/linouk23/NBA-Player-Movements/tree/master/"
 _URL = "https://github.com/linouk23/NBA-Player-Movements/raw/master/data/2016.NBA.Raw.SportVU.Game.Logs"
+_PBP_URL = "https://github.com/sumitrodatta/nba-alt-awards/raw/main/Historical/PBP%20Data/2015-16_pbp.csv"
 
 res = requests.get(_URL)
 
@@ -54,6 +57,7 @@ class NbaTracking(datasets.GeneratorBasedBuilder):
     """Tracking data for all games of 2015-2016 season in forms of coordinates for players and ball at each moment."""
     
     _URLS = _URLS
+    _PBP_URL = _PBP_URL
     
     def _info(self):
         features = datasets.Features(
@@ -156,14 +160,21 @@ class NbaTracking(datasets.GeneratorBasedBuilder):
     def _generate_examples(self, filepaths, split):
         # TODO: This method handles input defined in _split_generators to yield (key, example) tuples from the dataset.
         # The `key` is for legacy reasons (tfds) and is not important in itself, but must be unique for each example.
+        pbp_out = dl_manager.download_and_extract(_PBP_URL)
+        pbp = pd.read_csv(pbp_out)
+        
         moment_id = 0
+        
         for game_title, link in filepaths.items():
             with open(link, encoding="utf-8") as fp:
                 game = json.load(fp)
                 game_id = game["gameid"]
                 game_date = game["gamedate"] 
+
                 for event in game["events"]:
                     event_id = event["eventId"]
+
+                    event_row = pbp[(pbp.GAME_ID == int(game_id)) & (pbp.EVENTNUM == int(event_id))]
                     
                     visitor_name = event['visitor']['name']
                     visitor_team_id = event['visitor']['teamid']