Upload 27 files

app.py

@@ -1,242 +1,488 @@
-from shiny import App, ui, render, reactive
-import polars as pl
-import numpy as np
-import pandas as pd
-import api_scraper
-scrape = api_scraper.MLB_Scrape()
-from functions import df_update
-from functions import pitch_summary_functions
-update = df_update.df_update()
-from stuff_model import feature_engineering as fe
-from stuff_model import stuff_apply
-import requests
-import joblib
-from matplotlib.gridspec import GridSpec
-import math
-
-colour_palette = ['#FFB000','#648FFF','#785EF0',
-                  '#DC267F','#FE6100','#3D1EB2','#894D80','#16AA02','#B5592B','#A3C1ED']
-
-# df = pl.read_csv("data.csv")
-df = pl.read_parquet("data.parquet")
-print('df')
-
-year_input = 2024
-sport_id = 1
-
-df_schedule = scrape.get_schedule(year_input=[year_input],sport_id=[sport_id])
-df = df.join(df_schedule, on='game_id', how='left')
-
-df = df.with_columns(
-    pl.when((pl.col('batter_team_id') == pl.col('away_id')))
-    .then(pl.lit('Away'))
-    .when((pl.col('batter_team_id') == pl.col('home_id')))
-    .then(pl.lit('Home'))
-    .otherwise(None)
-    .alias('home_away_batter')
-)
-
-df = df.with_columns(
-    pl.when((pl.col('pitcher_team_id') == pl.col('away_id')))
-    .then(pl.lit('Away'))
-    .when((pl.col('pitcher_team_id') == pl.col('home_id')))
-    .then(pl.lit('Home'))
-    .otherwise(None)
-    .alias('home_away_pitcher')
-)
-
-print('schedule')
-
-df_stuff = stuff_apply.stuff_apply(fe.feature_engineering(df))
-print('stuff')
-df_up = update.update(df)
-print('update')
-df_total = df_up.join(df_stuff[['play_id','tj_stuff_plus']], on='play_id', how='left')
-print('total')
-
-stat_titles = {
-
-
-    
-    'pa': 'PA',
-    'bip': 'BBE',
-    'hits': 'H',
-    'k': 'K',
-    'bb': 'BB',
-    'max_launch_speed': 'Max EV',
-    'launch_speed_90': '90th% EV',
-    'launch_speed': 'EV',
-    'pitches': 'Pitches',
-    'tj_stuff_plus_avg': 'tjStuff+',
-    'avg': 'AVG',
-    'obp': 'OBP',
-    'slg': 'SLG',
-    'ops': 'OPS',
-    'k_percent': 'K%',
-    'bb_percent': 'BB%',
-    'k_minus_bb_percent': 'K-BB%',
-    'sweet_spot_percent': 'SwSpot%',
-    'woba_percent': 'wOBA',
-    'xwoba_percent': 'xwOBA',
-    'woba_percent_contact': 'wOBACON',
-    'xwoba_percent_contact': 'xwOBACON',
-    'hard_hit_percent': 'HardHit%',
-    'barrel_percent': 'Barrel%',
-    'zone_contact_percent': 'Z-Contact%',
-    'zone_swing_percent': 'Z-Swing%',
-    'zone_percent': 'Zone%',
-    'chase_percent': 'O-Swing%',
-    'chase_contact': 'O-Contact%',
-    'swing_percent': 'Swing%',
-    'whiff_rate': 'Whiff%',
-    'swstr_rate': 'SwStr%',
-    'ground_ball_percent': 'GB%',
-    'line_drive_percent': 'LD%',
-    'fly_ball_percent': 'FB%',
-    'pop_up_percent': 'PU%',
-    'heart_zone_swing_percent': 'Heart Swing%',
-    'shadow_zone_swing_percent': 'Shadow Swing%',
-    'chase_zone_swing_percent': 'Chase Swing%',
-    'waste_zone_swing_percent': 'Waste Swing%',
-    'heart_zone_whiff_percent': 'Heart Whiff%',
-    'shadow_zone_whiff_percent': 'Shadow Whiff%',
-    'chase_zone_whiff_percent': 'Chase Whiff%',
-    'waste_zone_whiff_percent': 'Waste Whiff%'
-}
-
-agg_titles = {    'batter_id':'Batter ID',
-    'batter_name':'Batter Name',
-    'batter_team':'Batter Team',
-    'batter_hand':'Batter Hand',
-    'pitcher_id':'Pitcher ID',
-    'pitcher_name':'Pitcher Name',
-    'pitcher_team':'Pitcher Team',
-    'pitcher_hand':'Pitcher Hand',
-    'pitch_type':'Pitch Type',
-    'pitch_group':'Pitch Group',
-    'home_away_batter':'Home/Away Batter',
-    'home_away_pitcher':'Home/Away Pitcher'}
-
-stat_selection = [key for key in stat_titles.keys()]
-
-# Pad each value with non-breaking spaces to a length of 20
-stat_titles_padded = {key: value + '\u00A0' * (30 - len(value)) for key, value in (agg_titles | stat_titles).items()}
-
-rounding_dict = {
-    'pa': 0,
-    'bip': 0,
-    'hits': 0,
-    'k': 0,
-    'bb': 0,
-    'max_launch_speed': 1,
-    'launch_speed_90': 1,
-    'launch_speed': 1,
-    'pitches': 0,
-    'tj_stuff_plus_avg': 1,
-    'avg': 3,
-    'obp': 3,
-    'slg': 3,
-    'ops': 3,
-    'k_percent': 3,
-    'bb_percent': 3,
-    'sweet_spot_percent': 3,
-    'woba_percent': 3,
-    'xwoba_percent': 3,
-    'woba_percent_contact': 3,
-    'xwoba_percent_contact': 3,
-    'hard_hit_percent': 3,
-    'barrel_percent': 3,
-    'zone_contact_percent': 3,
-    'zone_swing_percent': 3,
-    'zone_percent': 3,
-    'chase_percent': 3,
-    'chase_contact': 3,
-    'swing_percent': 3,
-    'whiff_rate': 3,
-    'swstr_rate': 3,
-    'ground_ball_percent': 3,
-    'line_drive_percent': 3,
-    'fly_ball_percent': 3,
-    'pop_up_percent': 3,
-    'heart_zone_swing_percent': 3,
-    'shadow_zone_swing_percent': 3,
-    'chase_zone_swing_percent': 3,
-    'waste_zone_swing_percent': 3,
-    'heart_zone_whiff_percent': 3,
-    'shadow_zone_whiff_percent': 3,
-    'chase_zone_whiff_percent': 3,
-    'waste_zone_whiff_percent': 3
-}
-
-
-
-app_ui = ui.page_sidebar(
-    ui.sidebar(
-        
-            
-            ui.input_selectize(
-                "list_input",
-                "Select Aggregation:",
-                choices=agg_titles,
-                multiple=True,
-                selected=['batter_id', 'batter_name']
-            ),
-            ui.input_selectize(
-                "list_stats",
-                "Select Stats:",
-                choices=stat_titles,
-                multiple=True,
-                selected=['pa']
-            ),
-            ui.input_date_range("date_id", "Select Date Range",
-                                   start=df_total['game_date'].min(),
-                                   end=df_total['game_date'].max(),
-                                   min=df_total['game_date'].min(),
-                                    max=df_total['game_date'].max()),
-        
-            ui.input_action_button("generate_table", "Generate Table", class_="btn-primary"),
-            width="400px"
-        
-    ),
-    ui.card(
-        ui.card_header("Leaderboard"),
-        ui.output_data_frame("list_view")
-    )
-)
-
-def server(input, output, session):
-    @output
-    @render.data_frame
-    @reactive.event(input.generate_table, ignore_none=False) 
-    def list_view():
-
-        start_date = str(input.date_id()[0])
-        end_date = str(input.date_id()[1])
-        selection_list = list(input.list_input())
-        stat_list = list(input.list_stats())
-        print(selection_list)
-        df_agg = update.update_summary_select(df=df_total.filter((pl.col('game_date')>=start_date)&(pl.col('game_date')<=end_date)),
-                                    selection=selection_list)
-        df_agg = df_agg.select(selection_list+stat_list)
-        for col in df_agg.columns[len(selection_list):]:
-            if col in rounding_dict:
-                df_agg = df_agg.with_columns(pl.col(col).round(rounding_dict[col]))
-
-        
-        # Only rename columns that exist in stat_titles
-        print(df_agg.columns)
-        rename_dict = {col: stat_titles_padded.get(col, col) for col in df_agg.columns}
-        # print(rename_dict)
-        result_df = df_agg.rename(rename_dict)
-        
-        return render.DataGrid(
-                result_df,
-                width='2000px',
-                height='750px',
-                filters=True,
-                
-                
-                
-            )
-
-app = App(app_ui, server)
+from shiny import App, ui, render, reactive
+import polars as pl
+import numpy as np
+import pandas as pd
+import api_scraper
+scrape = api_scraper.MLB_Scrape()
+from functions import df_update
+from functions import pitch_summary_functions
+update = df_update.df_update()
+from stuff_model import feature_engineering as fe
+from stuff_model import stuff_apply
+import requests
+import joblib
+from matplotlib.gridspec import GridSpec
+import math
+from pytabulator import TableOptions, Tabulator, output_tabulator, render_tabulator, theme
+theme.tabulator_site()
+
+colour_palette = ['#FFB000','#648FFF','#785EF0',
+                  '#DC267F','#FE6100','#3D1EB2','#894D80','#16AA02','#B5592B','#A3C1ED']
+
+# df = pl.read_csv("data.csv")
+# df = pl.read_parquet("data_small.parquet")[:]
+# df = pl.read_parquet("data.parquet")[:]
+# print('df')
+season = 2024
+
+df_mlb = pl.read_parquet("data/data_mlb_2024.parquet")[:]
+df_aaa = pl.read_parquet("data/data_aaa_2024.parquet")[:]
+df_a = pl.read_parquet("data/data_a_2024.parquet")[:]
+
+
+
+def df_final(df:pl.dataframe,year_input:int,sport_id:int):
+
+    df_schedule = scrape.get_schedule(year_input=[year_input],sport_id=[sport_id])
+    df = df.join(df_schedule, on='game_id', how='left')
+
+    df = df.with_columns(
+        pl.when((pl.col('batter_team_id') == pl.col('away_id')))
+        .then(pl.lit('Away'))
+        .when((pl.col('batter_team_id') == pl.col('home_id')))
+        .then(pl.lit('Home'))
+        .otherwise(None)
+        .alias('home_away')
+    )
+
+    df = df.with_columns(
+        pl.when((pl.col('pitcher_team_id') == pl.col('away_id')))
+        .then(pl.lit('Away'))
+        .when((pl.col('pitcher_team_id') == pl.col('home_id')))
+        .then(pl.lit('Home'))
+        .otherwise(None)
+        .alias('home_away_pitcher')
+    )
+
+
+    print('schedule')
+
+    df_stuff = stuff_apply.stuff_apply(fe.feature_engineering(df))
+    print('stuff')
+    df_up = update.update(df)
+    print('update')
+    df_total = df_up.join(df_stuff[['play_id','tj_stuff_plus']], on='play_id', how='left')
+    print('total')
+    return df_total
+
+
+df_mlb_total = df_final(df=df_mlb,year_input=season,sport_id=1)
+df_aaa_total = df_final(df=df_aaa,year_input=season,sport_id=11)
+df_a_total = df_final(df=df_a.drop_nulls(subset=['start_speed']),year_input=season,sport_id=14)
+
+rounding_dict = {
+    'pa': 0,
+    'bip': 0,
+    'hits': 0,
+    'k': 0,
+    'bb': 0,
+    'max_launch_speed': 1,
+    'launch_speed_90': 1,
+    'launch_speed': 1,
+    'pitches': 0,
+    'tj_stuff_plus_avg': 0,
+    'avg': 3,
+    'obp': 3,
+    'slg': 3,
+    'ops': 3,
+    'k_percent': 3,
+    'bb_percent': 3,
+    'k_minus_bb_percent': 3,
+    'sweet_spot_percent': 3,
+    'woba_percent': 3,
+    'xwoba_percent': 3,
+    'woba_percent_contact': 3,
+    'xwoba_percent_contact': 3,
+    'hard_hit_percent': 3,
+    'barrel_percent': 3,
+    'zone_contact_percent': 3,
+    'zone_swing_percent': 3,
+    'zone_percent': 3,
+    'chase_percent': 3,
+    'chase_contact': 3,
+    'swing_percent': 3,
+    'whiff_rate': 3,
+    'swstr_rate': 3,
+    'ground_ball_percent': 3,
+    'line_drive_percent': 3,
+    'fly_ball_percent': 3,
+    'pop_up_percent': 3,
+    'heart_zone_swing_percent': 3,
+    'shadow_zone_swing_percent': 3,
+    'chase_zone_swing_percent': 3,
+    'waste_zone_swing_percent': 3,
+    'heart_zone_whiff_percent': 3,
+    'shadow_zone_whiff_percent': 3,
+    'chase_zone_whiff_percent': 3,
+    'waste_zone_whiff_percent': 3,
+    'start_speed_avg': 1,
+    'vb_avg': 1,
+    'ivb_avg': 1,
+    'hb_avg': 1,
+    'z0_avg': 1,
+    'x0_avg': 1,
+    'vaa_avg': 1,
+    'haa_avg': 1,
+    'spin_rate_avg': 0,
+    'extension_avg': 1   
+}
+
+columns = [
+    { "title": "PA", "field": "pa", "width": 150},
+    { "title": "BBE", "field": "bip", "width": 150 },
+    { "title": "H", "field": "hits", "width": 150 },
+    { "title": "K", "field": "k", "width": 150 },
+    { "title": "BB", "field": "bb", "width": 150 },
+    { "title": "Max EV", "field": "max_launch_speed", "width": 150 },
+    { "title": "90th% EV", "field": "launch_speed_90", "width": 150 },
+    { "title": "EV", "field": "launch_speed", "width": 150 },
+    { "title": "Pitches", "field": "pitches", "width": 150 },
+    { "title": "AVG", "field": "avg", "width": 150 },
+    { "title": "OBP", "field": "obp", "width": 150 },
+    { "title": "SLG", "field": "slg", "width": 150 },
+    { "title": "OPS", "field": "ops", "width": 150 },
+    { "title": "K%", "field": "k_percent", "width": 150,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "BB%", "field": "bb_percent", "width": 150,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "K-BB%", "field": "k_minus_bb_percent", "width": 150,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "SwSpot%", "field": "sweet_spot_percent", "width": 150,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "wOBA", "field": "woba_percent", "width": 150 },
+    { "title": "xwOBA", "field": "xwoba_percent", "width": 150 },
+    { "title": "wOBACON", "field": "woba_percent_contact", "width": 150 },
+    { "title": "xwOBACON", "field": "xwoba_percent_contact", "width": 150 },
+    { "title": "HardHit%", "field": "hard_hit_percent", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "Barrel%", "field": "barrel_percent", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "Z-Contact%", "field": "zone_contact_percent", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "Z-Swing%", "field": "zone_swing_percent", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "Zone%", "field": "zone_percent", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "O-Swing%", "field": "chase_percent", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "O-Contact%", "field": "chase_contact", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "Swing%", "field": "swing_percent", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "Whiff%", "field": "whiff_rate", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "SwStr%", "field": "swstr_rate", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "GB%", "field": "ground_ball_percent", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "LD%", "field": "line_drive_percent", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "FB%", "field": "fly_ball_percent", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "PU%", "field": "pop_up_percent", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "Heart Swing%", "field": "heart_zone_swing_percent", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "Shadow Swing%", "field": "shadow_zone_swing_percent", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "Chase Swing%", "field": "chase_zone_swing_percent", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "Waste Swing%", "field": "waste_zone_swing_percent", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "Heart Whiff%", "field": "heart_zone_whiff_percent", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "Shadow Whiff%", "field": "shadow_zone_whiff_percent", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "Chase Whiff%", "field": "chase_zone_whiff_percent", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "Waste Whiff%", "field": "waste_zone_whiff_percent", "width": 150 ,"formatter": "money", "formatterParams":{"decimal":".","thousand":".","symbol":"%","symbolAfter":"%","negativeSign":True,"precision":1}},
+    { "title": "tjStuff+", "field": "tj_stuff_plus_avg", "width": 150 },
+    { "title": "Velocity", "field": "start_speed_avg", "width": 150 },
+    { "title": "Extension", "field": "extension_avg", "width": 150 },
+    { "title": "VB", "field": "vb_avg", "width": 150 },
+    { "title": "iVB", "field": "ivb_avg", "width": 150 },
+    { "title": "HB", "field": "hb_avg", "width": 150 },
+    { "title": "vRel", "field": "z0_avg", "width": 150 },
+    { "title": "hRel", "field": "x0_avg", "width": 150 },
+    { "title": "VAA", "field": "vaa_avg", "width": 150 },
+    { "title": "HAA", "field": "haa_avg", "width": 150 },
+    { "title": "Spin Rate", "field": "spin_rate_avg", "width": 150 },
+    { "title": "Extension", "field": "extension_avg", "width": 150 },
+
+]
+
+stat_titles = dict(zip([col["field"] for col in columns],[col["title"] for col in columns]))
+
+stat_selection = [key for key in stat_titles.keys()]
+
+agg_titles = {'batter_id':'Batter ID',
+    'batter_name':'Batter Name',
+    'batter_team':'Batter Team',
+    'batter_hand':'Batter Hand',
+    'pitcher_id':'Pitcher ID',
+    'pitcher_name':'Pitcher Name',
+    'pitcher_team':'Pitcher Team',
+    'pitcher_hand':'Pitcher Hand',
+    'pitch_type':'Pitch Type',
+    'pitch_group':'Pitch Group',
+    'home_away_batter':'Home/Away Batter',
+    'home_away_pitcher':'Home/Away Pitcher',
+    'is_swing':'Is Swing?',
+    'is_bip':'Is BIP?',
+    'in_zone_final':'In Zone?',
+    'attack_zone_final':'Attack Zone'}
+
+
+columns_group = [
+    { "title": "Batter ID", "field": "batter_id", "width": 150, "headerFilter":"input","frozen":True,},
+    { "title": "Batter Name", "field": "batter_name", "width": 200,"frozen":True, "headerFilter":"input"  },
+    { "title": "Batter Team", "field": "batter_team", "width": 150,"frozen":True, "headerFilter":"input"  },
+    { "title": "Batter Hand", "field": "batter_hand", "width": 150,"frozen":True, "headerFilter":"input"  },
+    { "title": "Pitcher ID", "field": "pitcher_id", "width": 150,"frozen":True, "headerFilter":"input"  },
+    { "title": "Pitcher Name", "field": "pitcher_name", "width": 200,"frozen":True, "headerFilter":"input"  },
+    { "title": "Pitcher Team", "field": "pitcher_team", "width": 150,"frozen":True, "headerFilter":"input"  },
+    { "title": "Pitcher Hand", "field": "pitcher_hand", "width": 150,"frozen":True, "headerFilter":"input"  },
+    { "title": "Pitch Type", "field": "pitch_type", "width": 150,"frozen":True, "headerFilter":"input"  },
+    { "title": "Pitch Group", "field": "pitch_group", "width": 150,"frozen":True, "headerFilter":"input"  },
+    { "title": "Home/Away Batter", "field": "home_away_batter", "width": 150,"frozen":True, "headerFilter":"input"  },
+    { "title": "Home/Away Pitcher", "field": "home_away_pitcher", "width": 150,"frozen":True, "headerFilter":"input"  },
+    { "title": "Is Swing?", "field": "is_swing", "width": 150,"frozen":True, "headerFilter":"input"  },
+    { "title": "Is BIP?", "field": "is_bip", "width": 150,"frozen":True, "headerFilter":"input"  },
+    { "title": "In Zone?", "field": "in_zone_final", "width": 150,"frozen":True, "headerFilter":"input"  },
+    { "title": "Attack Zone", "field": "attack_zone_final", "width": 150,"frozen":True, "headerFilter":"input"  } 
+    ]
+
+
+app_ui = ui.page_sidebar(
+    ui.sidebar(
+        ui.input_selectize(
+            "level_input",
+            "Select Level:",
+            choices=['MLB','AAA','A'],
+            multiple=False,
+            selected=['MLB']
+        ),
+        ui.input_selectize(
+            "list_input",
+            "Select Aggregation:",
+            choices=agg_titles,
+            multiple=True,
+            selected=['batter_id', 'batter_name']
+        ),
+        ui.input_selectize(
+            "list_stats",
+            "Select Stats:",
+            choices=stat_titles,
+            multiple=True,
+            selected=['pa']
+        ),
+        ui.input_date_range(
+            "date_id", 
+            "Select Date Range",
+            start=f'{season}-01-01',
+            end=f'{season}-12-01',
+            min=f'{season}-01-01',
+            max=f'{season}-12-01',
+        ),
+        ui.hr(),
+        ui.h4("Filters"),
+        ui.div(
+            {"id": "filter-container"},
+            ui.div(
+                {"class": "filter-row", "id": "filter_row_1"},  # Add id for deletion
+                ui.row(
+                    ui.column(5,  # Adjusted column widths to make room for delete button
+                        ui.input_select(
+                            "filter_column_1",
+                            "Metric",
+                            choices={}
+                        )
+                    ),
+                    ui.column(3,
+                        ui.input_select(
+                            "filter_operator_1",
+                            "Operator",
+                            choices=[">=", "<="]
+                        ),
+                    ),
+                    ui.column(3,
+                        ui.input_numeric(
+                            "filter_value_1",
+                            "Value",
+                            value=0
+                        )
+                    ),
+                    ui.column(1,
+                        ui.markdown("&nbsp;"),
+                 
+
+                        ui.input_action_button(
+                            f"delete_filter_1",
+                            "",
+                            class_="btn-danger btn-sm",
+                            style="padding: 3px 6px;",
+                            icon='✖'
+                        
+                    )
+                    )
+                )
+            )
+        ),
+        ui.input_action_button(
+            "add_filter",
+            "Add Filter",
+            class_="btn-secondary"
+        ),
+        ui.br(),
+        ui.br(),
+        ui.input_action_button(
+            "generate_table",
+            "Generate Table",
+            class_="btn-primary"
+        ),
+        width="400px"
+    ),
+    ui.navset_tab(
+        ui.nav_panel("Leaderboard",
+            ui.card(
+                #ui.card_header("Leaderboard"),
+                output_tabulator("tabulator")
+            )
+        ),
+
+    )
+)
+
+def server(input, output, session):
+    # Store the number of active filters
+    filter_count = reactive.value(1)
+    # Store active filter IDs
+    active_filters = reactive.value([1])
+
+    @reactive.effect
+    @reactive.event(input.list_stats)
+    def _():
+        stat_choices = {k: k for k in input.list_stats()}
+        filtered_stat_choices = {key: stat_titles[key] for key in stat_choices}   
+        ui.update_select("filter_column_1", choices=filtered_stat_choices)
+
+    @reactive.effect
+    @reactive.event(input.add_filter)
+    def _():
+        current_count = filter_count.get()
+        new_count = current_count + 1
+        
+        stat_choices = {k: k for k in input.list_stats()}
+        filtered_stat_choices = {key: stat_titles[key] for key in stat_choices}   
+        
+        ui.insert_ui(
+            selector="#filter-container",
+            where="beforeEnd",
+            ui=ui.div(
+                {"class": "filter-row", "id": f"filter_row_{new_count}"},
+                ui.row(
+                    ui.column(5,
+                        ui.input_select(
+                            f"filter_column_{new_count}",
+                            "Metric",
+                            choices=filtered_stat_choices
+                        ),
+                    ),
+                    ui.column(3,
+                        ui.input_select(
+                            f"filter_operator_{new_count}",
+                            "Operator",
+                            choices=[">=", "<="]
+                        ),
+                    ),
+                    ui.column(3,
+                        ui.input_numeric(
+                            f"filter_value_{new_count}",
+                            "Value",
+                            value=0
+                        )
+                    ),
+                    ui.column(1,
+                        ui.markdown("&nbsp;"),
+                 
+
+                        ui.input_action_button(
+                            f"delete_filter_{new_count}",
+                            "",
+                            class_="btn-danger btn-sm",
+                            style="padding: 3px 6px;",
+                            icon='✖'
+                        
+                    )
+                    )
+                )
+            )
+        )
+        filter_count.set(new_count)
+        current_filters = active_filters.get()
+        current_filters.append(new_count)
+        active_filters.set(current_filters)
+
+    @reactive.effect
+    def _():
+        # Monitor all possible delete buttons
+        for i in range(1, filter_count.get() + 1):
+            try:
+                if getattr(input, f"delete_filter_{i}")() > 0:
+                    # Remove the filter row
+                    ui.remove_ui(f"#filter_row_{i}")
+                    # Update active filters
+                    current_filters = active_filters.get()
+                    if i in current_filters:
+                        current_filters.remove(i)
+                        active_filters.set(current_filters)
+            except:
+                continue
+
+    @output
+    @render_tabulator
+    @reactive.event(input.generate_table, ignore_none=False) 
+    def tabulator():
+        columns_c = columns.copy()
+        selection_list = list(input.list_input())
+        start_date = str(input.date_id()[0])
+        end_date = str(input.date_id()[1])
+        
+
+        if input.level_input() == "MLB":
+            df_agg = update.update_summary_select(df=df_mlb_total.filter((pl.col('game_date')>=start_date)&(pl.col('game_date')<=end_date)),
+                                    selection=selection_list)
+
+        elif input.level_input() == "AAA":
+            df_agg = update.update_summary_select(df=df_aaa_total.filter((pl.col('game_date')>=start_date)&(pl.col('game_date')<=end_date)),
+                                    selection=selection_list)
+        
+        elif input.level_input() == "A":
+            df_agg = update.update_summary_select(df=df_a_total.filter((pl.col('game_date')>=start_date)&(pl.col('game_date')<=end_date)),
+                                    selection=selection_list)
+                
+        
+        df_agg = df_agg.select(selection_list + list(input.list_stats()))#.sort('pa', descending=True)
+        
+        # Apply filters - only for active filters
+        for i in active_filters.get():
+            try:
+                col_name = getattr(input, f"filter_column_{i}")()
+                if col_name:  # Only apply filter if column is selected
+                    operator = getattr(input, f"filter_operator_{i}")()
+                    if col_name in [col["field"] for col in columns_c if col.get("formatter") == "money"]:
+                        value =  getattr(input, f"filter_value_{i}")()/100
+                    else:
+                        value = getattr(input, f"filter_value_{i}")()
+                
+                    if operator == ">=":
+                        df_agg = df_agg.filter(pl.col(col_name) >= value)
+                    elif operator == "<=":
+                        df_agg = df_agg.filter(pl.col(col_name) <= value)
+            except:
+                continue
+        
+        for col in df_agg.columns[len(selection_list):]:
+            if col in rounding_dict:
+                df_agg = df_agg.with_columns(pl.col(col).round(rounding_dict[col]))
+
+        for column in columns_c:
+            if column.get("formatter") == "money" and column.get("field") in df_agg.columns:
+                df_agg = df_agg.with_columns(pl.col(column.get("field"))*100)
+
+        col_group = []
+        for column in columns_group:
+            if column.get("field") in df_agg.columns:
+                col_group.append(column)
+
+        col_group_stats = []
+        for column in columns_c:
+            if column.get("field") in df_agg.columns:
+                col_group_stats.append(column)
+
+        columns_c = col_group + col_group_stats
+        
+        return Tabulator(
+            df_agg.to_pandas(),
+            table_options=TableOptions(
+                height=800,
+                columns=columns_c,
+            )
+        )
+
+app = App(app_ui, server)

data/data_a_2024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11ddeb5b35ced8d2a3c627af2cded75c78fbb333fa4331569b78a7fb29ddce1f
+size 44964738

data/data_aaa_2024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf48b06263a3b0a25b701e98c589f7e11bdf0ec3665562082ea209a61b55468e
+size 117704536

data/data_mlb_2024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0f2c52277a98ecbba343650996f32bbb8a847631d676bce2f07dbd3931f09a8
+size 128295721

functions/df_update.py

@@ -1,506 +1,579 @@
-import polars as pl
-import numpy as np
-import joblib
-
-loaded_model = joblib.load('joblib_model/barrel_model.joblib')
-in_zone_model = joblib.load('joblib_model/in_zone_model_knn_20240410.joblib')
-attack_zone_model = joblib.load('joblib_model/model_attack_zone.joblib')
-xwoba_model = joblib.load('joblib_model/xwoba_model.joblib')
-px_model = joblib.load('joblib_model/linear_reg_model_x.joblib')
-pz_model = joblib.load('joblib_model/linear_reg_model_z.joblib')
-
-
-class df_update:
-    def __init__(self):
-        pass
-
-    def update(self, df_clone: pl.DataFrame):
-
-        df = df_clone.clone()
-        # Assuming px_model is defined and df is your DataFrame
-        hit_codes = ['single',
-            'double','home_run', 'triple']
-
-        ab_codes = ['single', 'strikeout', 'field_out',
-            'grounded_into_double_play', 'fielders_choice', 'force_out',
-            'double', 'field_error', 'home_run', 'triple',
-            'double_play',
-            'fielders_choice_out', 'strikeout_double_play',
-            'other_out','triple_play']
-
-
-        obp_true_codes = ['single', 'walk',
-            'double','home_run', 'triple',
-            'hit_by_pitch', 'intent_walk']
-
-        obp_codes = ['single', 'strikeout', 'walk', 'field_out',
-            'grounded_into_double_play', 'fielders_choice', 'force_out',
-            'double', 'sac_fly', 'field_error', 'home_run', 'triple',
-            'hit_by_pitch', 'double_play', 'intent_walk',
-            'fielders_choice_out', 'strikeout_double_play',
-            'sac_fly_double_play',
-            'other_out','triple_play']
-
-
-        contact_codes = ['In play, no out',
-                'Foul', 'In play, out(s)',
-            'In play, run(s)',
-            'Foul Bunt']
-
-        bip_codes = ['In play, no out', 'In play, run(s)','In play, out(s)']
-
-
-        conditions_barrel = [
-            df['launch_speed'].is_null(),
-            (df['launch_speed'] * 1.5 - df['launch_angle'] >= 117) & 
-            (df['launch_speed'] + df['launch_angle'] >= 124) & 
-            (df['launch_speed'] >= 98) & 
-            (df['launch_angle'] >= 4) & (df['launch_angle'] <= 50)
-        ]
-        choices_barrel = [False, True]
-
-        conditions_tb = [
-            (df['event_type'] == 'single'),
-            (df['event_type'] == 'double'),
-            (df['event_type'] == 'triple'),
-            (df['event_type'] == 'home_run')
-        ]
-        choices_tb = [1, 2, 3, 4]
-
-
-        conditions_woba = [
-            df['event_type'].is_in(['strikeout', 'field_out', 'sac_fly', 'force_out', 'grounded_into_double_play', 'fielders_choice', 'field_error', 'sac_bunt', 'double_play', 'fielders_choice_out', 'strikeout_double_play', 'sac_fly_double_play', 'other_out']),
-            df['event_type'] == 'walk',
-            df['event_type'] == 'hit_by_pitch',
-            df['event_type'] == 'single',
-            df['event_type'] == 'double',
-            df['event_type'] == 'triple',
-            df['event_type'] == 'home_run'
-        ]
-        choices_woba = [0, 0.689, 0.720, 0.881, 1.254, 1.589, 2.048]
-
-        woba_codes = ['strikeout', 'field_out', 'single', 'walk', 'hit_by_pitch', 'double', 'sac_fly', 'force_out', 'home_run', 'grounded_into_double_play', 'fielders_choice', 'field_error', 'triple', 'sac_bunt', 'double_play', 'fielders_choice_out', 'strikeout_double_play', 'sac_fly_double_play', 'other_out']
-
-        pitch_cat = {'FA': 'Fastball',
-                    'FF': 'Fastball',
-                    'FT': 'Fastball',
-                    'FC': 'Fastball',
-                    'FS': 'Off-Speed',
-                    'FO': 'Off-Speed',
-                    'SI': 'Fastball',
-                    'ST': 'Breaking',
-                    'SL': 'Breaking',
-                    'CU': 'Breaking',
-                    'KC': 'Breaking',
-                    'SC': 'Off-Speed',
-                    'GY': 'Off-Speed',
-                    'SV': 'Breaking',
-                    'CS': 'Breaking',
-                    'CH': 'Off-Speed',
-                    'KN': 'Off-Speed',
-                    'EP': 'Breaking',
-                    'UN': None,
-                    'IN': None,
-                    'PO': None,
-                    'AB': None,
-                    'AS': None,
-                    'NP': None}
-
-
-        df = df.with_columns([
-            pl.when(df['type_ab'].is_not_null()).then(1).otherwise(0).alias('pa'),
-            pl.when(df['is_pitch']).then(1).otherwise(0).alias('pitches'),
-            pl.when(df['sz_top'] == 0).then(None).otherwise(df['sz_top']).alias('sz_top'),
-            pl.when(df['sz_bot'] == 0).then(None).otherwise(df['sz_bot']).alias('sz_bot'),
-            pl.when(df['zone'] > 0).then(df['zone'] < 10).otherwise(None).alias('in_zone'),
-            pl.Series(px_model.predict(df[['x']].fill_null(0).to_numpy())[:, 0]).alias('px_predict'),
-            pl.Series(pz_model.predict(df[['y']].fill_null(0).to_numpy())[:, 0] + 3.2).alias('pz_predict'),
-            pl.Series(in_zone_model.predict(df[['px','pz','sz_top','sz_bot']].fill_null(0).to_numpy())[:]).alias('in_zone_predict'),
-            pl.Series(attack_zone_model.predict(df[['px','pz','sz_top','sz_bot']].fill_null(0).to_numpy())[:]).alias('attack_zone_predict'),
-            pl.when(df['event_type'].is_in(hit_codes)).then(True).otherwise(False).alias('hits'),
-            pl.when(df['event_type'].is_in(ab_codes)).then(True).otherwise(False).alias('ab'),
-            pl.when(df['event_type'].is_in(obp_true_codes)).then(True).otherwise(False).alias('on_base'),
-            pl.when(df['event_type'].is_in(obp_codes)).then(True).otherwise(False).alias('obp'),
-            pl.when(df['play_description'].is_in(bip_codes)).then(True).otherwise(False).alias('bip'),
-            pl.when(conditions_barrel[0]).then(choices_barrel[0]).when(conditions_barrel[1]).then(choices_barrel[1]).otherwise(None).alias('barrel'),
-            pl.when(df['launch_angle'].is_null()).then(False).when((df['launch_angle'] >= 8) & (df['launch_angle'] <= 32)).then(True).otherwise(None).alias('sweet_spot'),
-            pl.when(df['launch_speed'].is_null()).then(False).when(df['launch_speed'] >= 94.5).then(True).otherwise(None).alias('hard_hit'),
-            pl.when(conditions_tb[0]).then(choices_tb[0]).when(conditions_tb[1]).then(choices_tb[1]).when(conditions_tb[2]).then(choices_tb[2]).when(conditions_tb[3]).then(choices_tb[3]).otherwise(None).alias('tb'),
-            pl.when(conditions_woba[0]).then(choices_woba[0]).when(conditions_woba[1]).then(choices_woba[1]).when(conditions_woba[2]).then(choices_woba[2]).when(conditions_woba[3]).then(choices_woba[3]).when(conditions_woba[4]).then(choices_woba[4]).when(conditions_woba[5]).then(choices_woba[5]).when(conditions_woba[6]).then(choices_woba[6]).otherwise(None).alias('woba'),
-            pl.when((df['play_code'] == 'S') | (df['play_code'] == 'W') | (df['play_code'] == 'T')).then(1).otherwise(0).alias('whiffs'),
-            pl.when((df['play_code'] == 'S') | (df['play_code'] == 'W') | (df['play_code'] == 'T') | (df['play_code'] == 'C')).then(1).otherwise(0).alias('csw'),
-            pl.when(pl.col('is_swing').cast(pl.Boolean)).then(1).otherwise(0).alias('swings'),
-            pl.col('event_type').is_in(['strikeout','strikeout_double_play']).alias('k'),
-            pl.col('event_type').is_in(['walk', 'intent_walk']).alias('bb'),
-            pl.lit(None).alias('attack_zone'),
-            pl.lit(None).alias('woba_pred'),
-            pl.lit(None).alias('woba_pred_contact')
-
-        ])
-
-        df = df.with_columns([
-            pl.when(df['event_type'].is_in(woba_codes)).then(1).otherwise(None).alias('woba_codes'),
-            pl.when(df['event_type'].is_in(woba_codes)).then(1).otherwise(None).alias('xwoba_codes'),
-            pl.when((pl.col('tb') >= 0)).then(df['woba']).otherwise(None).alias('woba_contact'),
-            pl.when(pl.col('px').is_null()).then(pl.col('px_predict')).otherwise(pl.col('px')).alias('px'),
-            pl.when(pl.col('pz').is_null()).then(pl.col('pz_predict')).otherwise(pl.col('pz')).alias('pz'),
-            pl.when(pl.col('in_zone').is_null()).then(pl.col('in_zone_predict')).otherwise(pl.col('in_zone')).alias('in_zone'),
-            pl.when(df['launch_speed'].is_null()).then(None).otherwise(df['barrel']).alias('barrel'),
-            pl.lit('average').alias('average'),
-                pl.when(pl.col('in_zone') == False).then(True).otherwise(False).alias('out_zone'),
-            pl.when((pl.col('in_zone') == True) & (pl.col('swings') == 1)).then(True).otherwise(False).alias('zone_swing'),
-            pl.when((pl.col('in_zone') == True) & (pl.col('swings') == 1) & (pl.col('whiffs') == 0)).then(True).otherwise(False).alias('zone_contact'),
-            pl.when((pl.col('in_zone') == False) & (pl.col('swings') == 1)).then(True).otherwise(False).alias('ozone_swing'),
-            pl.when((pl.col('in_zone') == False) & (pl.col('swings') == 1) & (pl.col('whiffs') == 0)).then(True).otherwise(False).alias('ozone_contact'),
-            pl.when(pl.col('event_type').str.contains('strikeout')).then(True).otherwise(False).alias('k'),
-            pl.when(pl.col('event_type').is_in(['walk', 'intent_walk'])).then(True).otherwise(False).alias('bb'),
-            pl.when(pl.col('attack_zone').is_null()).then(pl.col('attack_zone_predict')).otherwise(pl.col('attack_zone')).alias('attack_zone'),
-            
-
-        ])
-
-        df = df.with_columns([
-            (df['k'].cast(pl.Float32) - df['bb'].cast(pl.Float32)).alias('k_minus_bb'),
-            (df['bb'].cast(pl.Float32) - df['k'].cast(pl.Float32)).alias('bb_minus_k'),
-            (df['launch_speed'] > 0).alias('bip_div'),
-            (df['attack_zone'] == 0).alias('heart'),
-            (df['attack_zone'] == 1).alias('shadow'),
-            (df['attack_zone'] == 2).alias('chase'),
-            (df['attack_zone'] == 3).alias('waste'),
-            ((df['attack_zone'] == 0) & (df['swings'] == 1)).alias('heart_swing'),
-            ((df['attack_zone'] == 1) & (df['swings'] == 1)).alias('shadow_swing'),
-            ((df['attack_zone'] == 2) & (df['swings'] == 1)).alias('chase_swing'),
-            ((df['attack_zone'] == 3) & (df['swings'] == 1)).alias('waste_swing'),
-            ((df['attack_zone'] == 0) & (df['whiffs'] == 1)).alias('heart_whiff'),
-            ((df['attack_zone'] == 1) & (df['whiffs'] == 1)).alias('shadow_whiff'),
-            ((df['attack_zone'] == 2) & (df['whiffs'] == 1)).alias('chase_whiff'),
-            ((df['attack_zone'] == 3) & (df['whiffs'] == 1)).alias('waste_whiff')
-        ])
-
-
-        [0, 0.689, 0.720, 0.881, 1.254, 1.589, 2.048]
-
-        df = df.with_columns([
-                pl.Series(
-                    [sum(x) for x in xwoba_model.predict_proba(df[['launch_angle', 'launch_speed']].fill_null(0).to_numpy()[:]) * ([0, 0.881, 1.254, 1.589, 2.048])]
-                ).alias('woba_pred_predict')
-            ])
-
-        df = df.with_columns([
-            pl.when(pl.col('event_type').is_in(['walk'])).then(0.689)
-            .when(pl.col('event_type').is_in(['hit_by_pitch'])).then(0.720)
-            .when(pl.col('event_type').is_in(['strikeout', 'strikeout_double_play'])).then(0)
-            .otherwise(pl.col('woba_pred_predict')).alias('woba_pred_predict')
-        ])
-
-        df = df.with_columns([
-            pl.when(pl.col('woba_codes').is_null()).then(None).otherwise(pl.col('woba_pred_predict')).alias('woba_pred'),
-            pl.when(pl.col('bip')!=1).then(None).otherwise(pl.col('woba_pred_predict')).alias('woba_pred_contact'),
-        ])
-
-        df = df.with_columns([
-            pl.when(pl.col('trajectory').is_in(['bunt_popup'])).then(pl.lit('popup'))
-            .when(pl.col('trajectory').is_in(['bunt_grounder'])).then(pl.lit('ground_ball'))
-            .when(pl.col('trajectory').is_in(['bunt_line_drive'])).then(pl.lit('line_drive'))
-            .when(pl.col('trajectory').is_in([''])).then(pl.lit(None))
-            .otherwise(pl.col('trajectory')).alias('trajectory')
-        ])
-
-
-        # Create one-hot encoded columns for the trajectory column
-        dummy_df = df.select(pl.col('trajectory')).to_dummies()
-
-        # Rename the one-hot encoded columns
-        dummy_df = dummy_df.rename({
-            'trajectory_fly_ball': 'trajectory_fly_ball',
-            'trajectory_ground_ball': 'trajectory_ground_ball',
-            'trajectory_line_drive': 'trajectory_line_drive',
-            'trajectory_popup': 'trajectory_popup'
-        })
-
-        # Ensure the columns are present in the DataFrame
-        for col in ['trajectory_fly_ball', 'trajectory_ground_ball', 'trajectory_line_drive', 'trajectory_popup']:
-            if col not in dummy_df.columns:
-                dummy_df = dummy_df.with_columns(pl.lit(0).alias(col))
-
-        # Join the one-hot encoded columns back to the original DataFrame
-        df = df.hstack(dummy_df)
-
-        # Check if 'trajectory_null' column exists and drop it
-        if 'trajectory_null' in df.columns:
-            df = df.drop('trajectory_null')
-
-
-        pitch_cat = {'FA': None,
-                            'FF': 'Fastball',
-                            'FT': 'Fastball',
-                            'FC': 'Fastball',
-                            'FS': 'Off-Speed',
-                            'FO': 'Off-Speed',
-                            'SI': 'Fastball',
-                            'ST': 'Breaking',
-                            'SL': 'Breaking',
-                            'CU': 'Breaking',
-                            'KC': 'Breaking',
-                            'SC': 'Off-Speed',
-                            'GY': 'Off-Speed',
-                            'SV': 'Breaking',
-                            'CS': 'Breaking',
-                            'CH': 'Off-Speed',
-                            'KN': 'Off-Speed',
-                            'EP': 'Breaking',
-                            'UN': None,
-                            'IN': None,
-                            'PO': None,
-                            'AB': None,
-                            'AS': None,
-                            'NP': None}
-        df = df.with_columns(
-            df["pitch_type"].map_elements(lambda x: pitch_cat.get(x, x)).alias("pitch_group")
-        )
-
-        return df
-
-    # Assuming df is your Polars DataFrame
-    def update_summary(self, df: pl.DataFrame, pitcher: bool = True) -> pl.DataFrame:
-        """
-        Update summary statistics for pitchers or batters.
-
-        Parameters:
-        df (pl.DataFrame): The input Polars DataFrame containing player statistics.
-        pitcher (bool): A flag indicating whether to calculate statistics for pitchers (True) or batters (False).
-
-        Returns:
-        pl.DataFrame: A Polars DataFrame with aggregated and calculated summary statistics.
-        """
-
-        # Determine the position based on the pitcher flag
-        if pitcher:
-            position = 'pitcher'
-        else:
-            position = 'batter'
-
-        # Group by position_id and position_name, then aggregate various statistics
-        df_summ = df.group_by([f'{position}_id', f'{position}_name']).agg([
-            pl.col('pa').sum().alias('pa'),
-            pl.col('ab').sum().alias('ab'),
-            pl.col('obp').sum().alias('obp_pa'),
-            pl.col('hits').sum().alias('hits'),
-            pl.col('on_base').sum().alias('on_base'),
-            pl.col('k').sum().alias('k'),
-            pl.col('bb').sum().alias('bb'),
-            pl.col('bb_minus_k').sum().alias('bb_minus_k'),
-            pl.col('csw').sum().alias('csw'),
-            pl.col('bip').sum().alias('bip'),
-            pl.col('bip_div').sum().alias('bip_div'),
-            pl.col('tb').sum().alias('tb'),
-            pl.col('woba').sum().alias('woba'),
-            pl.col('woba_contact').sum().alias('woba_contact'),
-            pl.col('woba_pred').sum().alias('xwoba'),
-            pl.col('woba_pred_contact').sum().alias('xwoba_contact'),
-            pl.col('woba_codes').sum().alias('woba_codes'),
-            pl.col('xwoba_codes').sum().alias('xwoba_codes'),
-            pl.col('hard_hit').sum().alias('hard_hit'),
-            pl.col('barrel').sum().alias('barrel'),
-            pl.col('sweet_spot').sum().alias('sweet_spot'),
-            pl.col('launch_speed').max().alias('max_launch_speed'),
-            pl.col('launch_speed').quantile(0.90).alias('launch_speed_90'),
-            pl.col('launch_speed').mean().alias('launch_speed'),
-            pl.col('launch_angle').mean().alias('launch_angle'),
-            pl.col('is_pitch').sum().alias('pitches'),
-            pl.col('swings').sum().alias('swings'),
-            pl.col('in_zone').sum().alias('in_zone'),
-            pl.col('out_zone').sum().alias('out_zone'),
-            pl.col('whiffs').sum().alias('whiffs'),
-            pl.col('zone_swing').sum().alias('zone_swing'),
-            pl.col('zone_contact').sum().alias('zone_contact'),
-            pl.col('ozone_swing').sum().alias('ozone_swing'),
-            pl.col('ozone_contact').sum().alias('ozone_contact'),
-            pl.col('trajectory_ground_ball').sum().alias('ground_ball'),
-            pl.col('trajectory_line_drive').sum().alias('line_drive'),
-            pl.col('trajectory_fly_ball').sum().alias('fly_ball'),
-            pl.col('trajectory_popup').sum().alias('pop_up'),
-            pl.col('attack_zone').count().alias('attack_zone'),
-            pl.col('heart').sum().alias('heart'),
-            pl.col('shadow').sum().alias('shadow'),
-            pl.col('chase').sum().alias('chase'),
-            pl.col('waste').sum().alias('waste'),
-            pl.col('heart_swing').sum().alias('heart_swing'),
-            pl.col('shadow_swing').sum().alias('shadow_swing'),
-            pl.col('chase_swing').sum().alias('chase_swing'),
-            pl.col('waste_swing').sum().alias('waste_swing'),
-            pl.col('heart_whiff').sum().alias('heart_whiff'),
-            pl.col('shadow_whiff').sum().alias('shadow_whiff'),
-            pl.col('chase_whiff').sum().alias('chase_whiff'),
-            pl.col('waste_whiff').sum().alias('waste_whiff')
-        ])
-
-        # Add calculated columns to the summary DataFrame
-        df_summ = df_summ.with_columns([
-            (pl.col('hits') / pl.col('ab')).alias('avg'),
-            (pl.col('on_base') / pl.col('obp_pa')).alias('obp'),
-            (pl.col('tb') / pl.col('ab')).alias('slg'),
-            (pl.col('on_base') / pl.col('obp_pa') + pl.col('tb') / pl.col('ab')).alias('ops'),
-            (pl.col('k') / pl.col('pa')).alias('k_percent'),
-            (pl.col('bb') / pl.col('pa')).alias('bb_percent'),
-            (pl.col('bb_minus_k') / pl.col('pa')).alias('bb_minus_k_percent'),
-            (pl.col('bb') / pl.col('k')).alias('bb_over_k_percent'),
-            (pl.col('csw') / pl.col('pitches')).alias('csw_percent'),
-            (pl.col('sweet_spot') / pl.col('bip_div')).alias('sweet_spot_percent'),
-            (pl.col('woba') / pl.col('woba_codes')).alias('woba_percent'),
-            (pl.col('woba_contact') / pl.col('bip')).alias('woba_percent_contact'),
-            (pl.col('hard_hit') / pl.col('bip_div')).alias('hard_hit_percent'),
-            (pl.col('barrel') / pl.col('bip_div')).alias('barrel_percent'),
-            (pl.col('zone_contact') / pl.col('zone_swing')).alias('zone_contact_percent'),
-            (pl.col('zone_swing') / pl.col('in_zone')).alias('zone_swing_percent'),
-            (pl.col('in_zone') / pl.col('pitches')).alias('zone_percent'),
-            (pl.col('ozone_swing') / (pl.col('pitches') - pl.col('in_zone'))).alias('chase_percent'),
-            (pl.col('ozone_contact') / pl.col('ozone_swing')).alias('chase_contact'),
-            (pl.col('swings') / pl.col('pitches')).alias('swing_percent'),
-            (pl.col('whiffs') / pl.col('swings')).alias('whiff_rate'),
-            (pl.col('whiffs') / pl.col('pitches')).alias('swstr_rate'),
-            (pl.col('ground_ball') / pl.col('bip')).alias('ground_ball_percent'),
-            (pl.col('line_drive') / pl.col('bip')).alias('line_drive_percent'),
-            (pl.col('fly_ball') / pl.col('bip')).alias('fly_ball_percent'),
-            (pl.col('pop_up') / pl.col('bip')).alias('pop_up_percent'),
-            (pl.col('heart') / pl.col('attack_zone')).alias('heart_zone_percent'),
-            (pl.col('shadow') / pl.col('attack_zone')).alias('shadow_zone_percent'),
-            (pl.col('chase') / pl.col('attack_zone')).alias('chase_zone_percent'),
-            (pl.col('waste') / pl.col('attack_zone')).alias('waste_zone_percent'),
-            (pl.col('heart_swing') / pl.col('heart')).alias('heart_zone_swing_percent'),
-            (pl.col('shadow_swing') / pl.col('shadow')).alias('shadow_zone_swing_percent'),
-            (pl.col('chase_swing') / pl.col('chase')).alias('chase_zone_swing_percent'),
-            (pl.col('waste_swing') / pl.col('waste')).alias('waste_zone_swing_percent'),
-            (pl.col('heart_whiff') / pl.col('heart_swing')).alias('heart_zone_whiff_percent'),
-            (pl.col('shadow_whiff') / pl.col('shadow_swing')).alias('shadow_zone_whiff_percent'),
-            (pl.col('chase_whiff') / pl.col('chase_swing')).alias('chase_zone_whiff_percent'),
-            (pl.col('waste_whiff') / pl.col('waste_swing')).alias('waste_zone_whiff_percent'),
-            (pl.col('xwoba') / pl.col('xwoba_codes')).alias('xwoba_percent'),
-            (pl.col('xwoba_contact') / pl.col('bip')).alias('xwoba_percent_contact')
-        ])
-
-        return df_summ
-    
-
-
-
-
-    
-    # Assuming df is your Polars DataFrame
-    def update_summary_select(self, df: pl.DataFrame, selection: list) -> pl.DataFrame:
-        """
-        Update summary statistics for pitchers or batters.
-
-        Parameters:
-        df (pl.DataFrame): The input Polars DataFrame containing player statistics.
-        pitcher (bool): A flag indicating whether to calculate statistics for pitchers (True) or batters (False).
-
-        Returns:
-        pl.DataFrame: A Polars DataFrame with aggregated and calculated summary statistics.
-        """
-
-        # Group by position_id and position_name, then aggregate various statistics
-        df_summ = df.group_by(selection).agg([
-            pl.col('pa').sum().alias('pa'),
-            pl.col('ab').sum().alias('ab'),
-            pl.col('obp').sum().alias('obp_pa'),
-            pl.col('hits').sum().alias('hits'),
-            pl.col('on_base').sum().alias('on_base'),
-            pl.col('k').sum().alias('k'),
-            pl.col('bb').sum().alias('bb'),
-            pl.col('bb_minus_k').sum().alias('bb_minus_k'),
-            pl.col('k_minus_bb').sum().alias('k_minus_bb'),
-            pl.col('csw').sum().alias('csw'),
-            pl.col('bip').sum().alias('bip'),
-            pl.col('bip_div').sum().alias('bip_div'),
-            pl.col('tb').sum().alias('tb'),
-            pl.col('woba').sum().alias('woba'),
-            pl.col('woba_contact').sum().alias('woba_contact'),
-            pl.col('woba_pred').sum().alias('xwoba'),
-            pl.col('woba_pred_contact').sum().alias('xwoba_contact'),
-            pl.col('woba_codes').sum().alias('woba_codes'),
-            pl.col('xwoba_codes').sum().alias('xwoba_codes'),
-            pl.col('hard_hit').sum().alias('hard_hit'),
-            pl.col('barrel').sum().alias('barrel'),
-            pl.col('sweet_spot').sum().alias('sweet_spot'),
-            pl.col('launch_speed').max().alias('max_launch_speed'),
-            pl.col('launch_speed').quantile(0.90).alias('launch_speed_90'),
-            pl.col('launch_speed').mean().alias('launch_speed'),
-            pl.col('launch_angle').mean().alias('launch_angle'),
-            pl.col('is_pitch').sum().alias('pitches'),
-            pl.col('swings').sum().alias('swings'),
-            pl.col('in_zone').sum().alias('in_zone'),
-            pl.col('out_zone').sum().alias('out_zone'),
-            pl.col('whiffs').sum().alias('whiffs'),
-            pl.col('zone_swing').sum().alias('zone_swing'),
-            pl.col('zone_contact').sum().alias('zone_contact'),
-            pl.col('ozone_swing').sum().alias('ozone_swing'),
-            pl.col('ozone_contact').sum().alias('ozone_contact'),
-            pl.col('trajectory_ground_ball').sum().alias('ground_ball'),
-            pl.col('trajectory_line_drive').sum().alias('line_drive'),
-            pl.col('trajectory_fly_ball').sum().alias('fly_ball'),
-            pl.col('trajectory_popup').sum().alias('pop_up'),
-            pl.col('attack_zone').count().alias('attack_zone'),
-            pl.col('heart').sum().alias('heart'),
-            pl.col('shadow').sum().alias('shadow'),
-            pl.col('chase').sum().alias('chase'),
-            pl.col('waste').sum().alias('waste'),
-            pl.col('heart_swing').sum().alias('heart_swing'),
-            pl.col('shadow_swing').sum().alias('shadow_swing'),
-            pl.col('chase_swing').sum().alias('chase_swing'),
-            pl.col('waste_swing').sum().alias('waste_swing'),
-            pl.col('heart_whiff').sum().alias('heart_whiff'),
-            pl.col('shadow_whiff').sum().alias('shadow_whiff'),
-            pl.col('chase_whiff').sum().alias('chase_whiff'),
-            pl.col('waste_whiff').sum().alias('waste_whiff'),
-            pl.col('tj_stuff_plus').sum().alias('tj_stuff_plus')
-        ])
-
-        # Add calculated columns to the summary DataFrame
-        df_summ = df_summ.with_columns([
-            (pl.col('hits') / pl.col('ab')).alias('avg'),
-            (pl.col('on_base') / pl.col('obp_pa')).alias('obp'),
-            (pl.col('tb') / pl.col('ab')).alias('slg'),
-            (pl.col('on_base') / pl.col('obp_pa') + pl.col('tb') / pl.col('ab')).alias('ops'),
-            (pl.col('k') / pl.col('pa')).alias('k_percent'),
-            (pl.col('bb') / pl.col('pa')).alias('bb_percent'),
-            (pl.col('bb_minus_k') / pl.col('pa')).alias('bb_minus_k_percent'),
-            (pl.col('k_minus_bb') / pl.col('pa')).alias('k_minus_bb_percent'),
-            (pl.col('bb') / pl.col('k')).alias('bb_over_k_percent'),
-            (pl.col('csw') / pl.col('pitches')).alias('csw_percent'),
-            (pl.col('sweet_spot') / pl.col('bip_div')).alias('sweet_spot_percent'),
-            (pl.col('woba') / pl.col('woba_codes')).alias('woba_percent'),
-            (pl.col('woba_contact') / pl.col('bip')).alias('woba_percent_contact'),
-            (pl.col('hard_hit') / pl.col('bip_div')).alias('hard_hit_percent'),
-            (pl.col('barrel') / pl.col('bip_div')).alias('barrel_percent'),
-            (pl.col('zone_contact') / pl.col('zone_swing')).alias('zone_contact_percent'),
-            (pl.col('zone_swing') / pl.col('in_zone')).alias('zone_swing_percent'),
-            (pl.col('in_zone') / pl.col('pitches')).alias('zone_percent'),
-            (pl.col('ozone_swing') / (pl.col('pitches') - pl.col('in_zone'))).alias('chase_percent'),
-            (pl.col('ozone_contact') / pl.col('ozone_swing')).alias('chase_contact'),
-            (pl.col('swings') / pl.col('pitches')).alias('swing_percent'),
-            (pl.col('whiffs') / pl.col('swings')).alias('whiff_rate'),
-            (pl.col('whiffs') / pl.col('pitches')).alias('swstr_rate'),
-            (pl.col('ground_ball') / pl.col('bip')).alias('ground_ball_percent'),
-            (pl.col('line_drive') / pl.col('bip')).alias('line_drive_percent'),
-            (pl.col('fly_ball') / pl.col('bip')).alias('fly_ball_percent'),
-            (pl.col('pop_up') / pl.col('bip')).alias('pop_up_percent'),
-            (pl.col('heart') / pl.col('attack_zone')).alias('heart_zone_percent'),
-            (pl.col('shadow') / pl.col('attack_zone')).alias('shadow_zone_percent'),
-            (pl.col('chase') / pl.col('attack_zone')).alias('chase_zone_percent'),
-            (pl.col('waste') / pl.col('attack_zone')).alias('waste_zone_percent'),
-            (pl.col('heart_swing') / pl.col('heart')).alias('heart_zone_swing_percent'),
-            (pl.col('shadow_swing') / pl.col('shadow')).alias('shadow_zone_swing_percent'),
-            (pl.col('chase_swing') / pl.col('chase')).alias('chase_zone_swing_percent'),
-            (pl.col('waste_swing') / pl.col('waste')).alias('waste_zone_swing_percent'),
-            (pl.col('heart_whiff') / pl.col('heart_swing')).alias('heart_zone_whiff_percent'),
-            (pl.col('shadow_whiff') / pl.col('shadow_swing')).alias('shadow_zone_whiff_percent'),
-            (pl.col('chase_whiff') / pl.col('chase_swing')).alias('chase_zone_whiff_percent'),
-            (pl.col('waste_whiff') / pl.col('waste_swing')).alias('waste_zone_whiff_percent'),
-            (pl.col('xwoba') / pl.col('xwoba_codes')).alias('xwoba_percent'),
-            (pl.col('xwoba_contact') / pl.col('bip')).alias('xwoba_percent_contact'),
-            (pl.col('tj_stuff_plus') / pl.col('pitches')).alias('tj_stuff_plus_avg'),
-            
-        ])
-
+import polars as pl
+import numpy as np
+import joblib
+
+loaded_model = joblib.load('joblib_model/barrel_model.joblib')
+in_zone_model = joblib.load('joblib_model/in_zone_model_knn_20240410.joblib')
+attack_zone_model = joblib.load('joblib_model/model_attack_zone.joblib')
+xwoba_model = joblib.load('joblib_model/xwoba_model.joblib')
+px_model = joblib.load('joblib_model/linear_reg_model_x.joblib')
+pz_model = joblib.load('joblib_model/linear_reg_model_z.joblib')
+
+
+class df_update:
+    def __init__(self):
+        pass
+
+    def update(self, df_clone: pl.DataFrame):
+
+        df = df_clone.clone()
+        # Assuming px_model is defined and df is your DataFrame
+        hit_codes = ['single',
+            'double','home_run', 'triple']
+
+        ab_codes = ['single', 'strikeout', 'field_out',
+            'grounded_into_double_play', 'fielders_choice', 'force_out',
+            'double', 'field_error', 'home_run', 'triple',
+            'double_play',
+            'fielders_choice_out', 'strikeout_double_play',
+            'other_out','triple_play']
+
+
+        obp_true_codes = ['single', 'walk',
+            'double','home_run', 'triple',
+            'hit_by_pitch', 'intent_walk']
+
+        obp_codes = ['single', 'strikeout', 'walk', 'field_out',
+            'grounded_into_double_play', 'fielders_choice', 'force_out',
+            'double', 'sac_fly', 'field_error', 'home_run', 'triple',
+            'hit_by_pitch', 'double_play', 'intent_walk',
+            'fielders_choice_out', 'strikeout_double_play',
+            'sac_fly_double_play',
+            'other_out','triple_play']
+
+
+        contact_codes = ['In play, no out',
+                'Foul', 'In play, out(s)',
+            'In play, run(s)',
+            'Foul Bunt']
+
+        bip_codes = ['In play, no out', 'In play, run(s)','In play, out(s)']
+
+
+        conditions_barrel = [
+            df['launch_speed'].is_null(),
+            (df['launch_speed'] * 1.5 - df['launch_angle'] >= 117) & 
+            (df['launch_speed'] + df['launch_angle'] >= 124) & 
+            (df['launch_speed'] >= 98) & 
+            (df['launch_angle'] >= 4) & (df['launch_angle'] <= 50)
+        ]
+        choices_barrel = [False, True]
+
+        conditions_tb = [
+            (df['event_type'] == 'single'),
+            (df['event_type'] == 'double'),
+            (df['event_type'] == 'triple'),
+            (df['event_type'] == 'home_run')
+        ]
+        choices_tb = [1, 2, 3, 4]
+
+
+        conditions_woba = [
+            df['event_type'].is_in(['strikeout', 'field_out', 'sac_fly', 'force_out', 'grounded_into_double_play', 'fielders_choice', 'field_error', 'sac_bunt', 'double_play', 'fielders_choice_out', 'strikeout_double_play', 'sac_fly_double_play', 'other_out']),
+            df['event_type'] == 'walk',
+            df['event_type'] == 'hit_by_pitch',
+            df['event_type'] == 'single',
+            df['event_type'] == 'double',
+            df['event_type'] == 'triple',
+            df['event_type'] == 'home_run'
+        ]
+        choices_woba = [0, 0.689, 0.720, 0.881, 1.254, 1.589, 2.048]
+
+        woba_codes = ['strikeout', 'field_out', 'single', 'walk', 'hit_by_pitch', 'double', 'sac_fly', 'force_out', 'home_run', 'grounded_into_double_play', 'fielders_choice', 'field_error', 'triple', 'sac_bunt', 'double_play', 'fielders_choice_out', 'strikeout_double_play', 'sac_fly_double_play', 'other_out']
+
+        pitch_cat = {'FA': 'Fastball',
+                    'FF': 'Fastball',
+                    'FT': 'Fastball',
+                    'FC': 'Fastball',
+                    'FS': 'Off-Speed',
+                    'FO': 'Off-Speed',
+                    'SI': 'Fastball',
+                    'ST': 'Breaking',
+                    'SL': 'Breaking',
+                    'CU': 'Breaking',
+                    'KC': 'Breaking',
+                    'SC': 'Off-Speed',
+                    'GY': 'Off-Speed',
+                    'SV': 'Breaking',
+                    'CS': 'Breaking',
+                    'CH': 'Off-Speed',
+                    'KN': 'Off-Speed',
+                    'EP': 'Breaking',
+                    'UN': None,
+                    'IN': None,
+                    'PO': None,
+                    'AB': None,
+                    'AS': None,
+                    'NP': None}
+
+
+        df = df.with_columns([
+            pl.when(df['type_ab'].is_not_null()).then(1).otherwise(0).alias('pa'),
+            pl.when(df['is_pitch']).then(1).otherwise(0).alias('pitches'),
+            pl.when(df['sz_top'] == 0).then(None).otherwise(df['sz_top']).alias('sz_top'),
+            pl.when(df['sz_bot'] == 0).then(None).otherwise(df['sz_bot']).alias('sz_bot'),
+            pl.when(df['zone'] > 0).then(df['zone'] < 10).otherwise(None).alias('in_zone'),
+            pl.Series(px_model.predict(df[['x']].fill_null(0).to_numpy())[:, 0]).alias('px_predict'),
+            pl.Series(pz_model.predict(df[['y']].fill_null(0).to_numpy())[:, 0] + 3.2).alias('pz_predict'),
+            pl.Series(in_zone_model.predict(df[['px','pz','sz_top','sz_bot']].fill_null(0).to_numpy())[:]).alias('in_zone_predict'),
+            pl.Series(attack_zone_model.predict(df[['px','pz','sz_top','sz_bot']].fill_null(0).to_numpy())[:]).alias('attack_zone_predict'),
+            pl.when(df['event_type'].is_in(hit_codes)).then(True).otherwise(False).alias('hits'),
+            pl.when(df['event_type'].is_in(ab_codes)).then(True).otherwise(False).alias('ab'),
+            pl.when(df['event_type'].is_in(obp_true_codes)).then(True).otherwise(False).alias('on_base'),
+            pl.when(df['event_type'].is_in(obp_codes)).then(True).otherwise(False).alias('obp'),
+            pl.when(df['play_description'].is_in(bip_codes)).then(True).otherwise(False).alias('bip'),
+            pl.when(conditions_barrel[0]).then(choices_barrel[0]).when(conditions_barrel[1]).then(choices_barrel[1]).otherwise(None).alias('barrel'),
+            pl.when(df['launch_angle'].is_null()).then(False).when((df['launch_angle'] >= 8) & (df['launch_angle'] <= 32)).then(True).otherwise(None).alias('sweet_spot'),
+            pl.when(df['launch_speed'].is_null()).then(False).when(df['launch_speed'] >= 94.5).then(True).otherwise(None).alias('hard_hit'),
+            pl.when(conditions_tb[0]).then(choices_tb[0]).when(conditions_tb[1]).then(choices_tb[1]).when(conditions_tb[2]).then(choices_tb[2]).when(conditions_tb[3]).then(choices_tb[3]).otherwise(None).alias('tb'),
+            pl.when(conditions_woba[0]).then(choices_woba[0]).when(conditions_woba[1]).then(choices_woba[1]).when(conditions_woba[2]).then(choices_woba[2]).when(conditions_woba[3]).then(choices_woba[3]).when(conditions_woba[4]).then(choices_woba[4]).when(conditions_woba[5]).then(choices_woba[5]).when(conditions_woba[6]).then(choices_woba[6]).otherwise(None).alias('woba'),
+            pl.when((df['play_code'] == 'S') | (df['play_code'] == 'W') | (df['play_code'] == 'T')).then(1).otherwise(0).alias('whiffs'),
+            pl.when((df['play_code'] == 'S') | (df['play_code'] == 'W') | (df['play_code'] == 'T') | (df['play_code'] == 'C')).then(1).otherwise(0).alias('csw'),
+            pl.when(pl.col('is_swing').cast(pl.Boolean)).then(1).otherwise(0).alias('swings'),
+            pl.col('event_type').is_in(['strikeout','strikeout_double_play']).alias('k'),
+            pl.col('event_type').is_in(['walk', 'intent_walk']).alias('bb'),
+            pl.lit(None).alias('attack_zone'),
+            pl.lit(None).alias('woba_pred'),
+            pl.lit(None).alias('woba_pred_contact')
+
+        ])
+
+
+        df = df.with_columns([
+            pl.when(df['event_type'].is_in(woba_codes)).then(1).otherwise(None).alias('woba_codes'),
+            pl.when(df['event_type'].is_in(woba_codes)).then(1).otherwise(None).alias('xwoba_codes'),
+            pl.when((pl.col('tb') >= 0)).then(df['woba']).otherwise(None).alias('woba_contact'),
+            pl.when(pl.col('px').is_null()).then(pl.col('px_predict')).otherwise(pl.col('px')).alias('px'),
+            pl.when(pl.col('pz').is_null()).then(pl.col('pz_predict')).otherwise(pl.col('pz')).alias('pz'),
+            pl.when(pl.col('in_zone').is_null()).then(pl.col('in_zone_predict')).otherwise(pl.col('in_zone')).alias('in_zone_final'),
+
+        ])
+
+        df = df.with_columns([
+            pl.when(df['launch_speed'].is_null()).then(None).otherwise(df['barrel']).alias('barrel'),
+            pl.lit('average').alias('average'),
+            pl.when(pl.col('in_zone_final') == False).then(True).otherwise(False).alias('out_zone'),
+            pl.when((pl.col('in_zone_final') == True) & (pl.col('swings') == 1)).then(True).otherwise(False).alias('zone_swing'),
+            pl.when((pl.col('in_zone_final') == True) & (pl.col('swings') == 1) & (pl.col('whiffs') == 0)).then(True).otherwise(False).alias('zone_contact'),
+            pl.when((pl.col('in_zone_final') == False) & (pl.col('swings') == 1)).then(True).otherwise(False).alias('ozone_swing'),
+            pl.when((pl.col('in_zone_final') == False) & (pl.col('swings') == 1) & (pl.col('whiffs') == 0)).then(True).otherwise(False).alias('ozone_contact'),
+            pl.when(pl.col('event_type').str.contains('strikeout')).then(True).otherwise(False).alias('k'),
+            pl.when(pl.col('event_type').is_in(['walk', 'intent_walk'])).then(True).otherwise(False).alias('bb'),
+            pl.when(pl.col('attack_zone').is_null()).then(pl.col('attack_zone_predict')).otherwise(pl.col('attack_zone')).alias('attack_zone_final'),
+            
+
+        ])
+
+        df = df.with_columns([
+            (df['k'].cast(pl.Float32) - df['bb'].cast(pl.Float32)).alias('k_minus_bb'),
+            (df['bb'].cast(pl.Float32) - df['k'].cast(pl.Float32)).alias('bb_minus_k'),
+            (df['launch_speed'] > 0).alias('bip_div'),
+            (df['attack_zone_final'] == 0).alias('heart'),
+            (df['attack_zone_final'] == 1).alias('shadow'),
+            (df['attack_zone_final'] == 2).alias('chase'),
+            (df['attack_zone_final'] == 3).alias('waste'),
+            ((df['attack_zone_final'] == 0) & (df['swings'] == 1)).alias('heart_swing'),
+            ((df['attack_zone_final'] == 1) & (df['swings'] == 1)).alias('shadow_swing'),
+            ((df['attack_zone_final'] == 2) & (df['swings'] == 1)).alias('chase_swing'),
+            ((df['attack_zone_final'] == 3) & (df['swings'] == 1)).alias('waste_swing'),
+            ((df['attack_zone_final'] == 0) & (df['whiffs'] == 1)).alias('heart_whiff'),
+            ((df['attack_zone_final'] == 1) & (df['whiffs'] == 1)).alias('shadow_whiff'),
+            ((df['attack_zone_final'] == 2) & (df['whiffs'] == 1)).alias('chase_whiff'),
+            ((df['attack_zone_final'] == 3) & (df['whiffs'] == 1)).alias('waste_whiff')
+        ])
+
+
+        [0, 0.689, 0.720, 0.881, 1.254, 1.589, 2.048]
+
+        df = df.with_columns([
+                pl.Series(
+                    [sum(x) for x in xwoba_model.predict_proba(df[['launch_angle', 'launch_speed']].fill_null(0).to_numpy()[:]) * ([0, 0.881, 1.254, 1.589, 2.048])]
+                ).alias('woba_pred_predict')
+            ])
+
+        df = df.with_columns([
+            pl.when(pl.col('event_type').is_in(['walk'])).then(0.689)
+            .when(pl.col('event_type').is_in(['hit_by_pitch'])).then(0.720)
+            .when(pl.col('event_type').is_in(['strikeout', 'strikeout_double_play'])).then(0)
+            .otherwise(pl.col('woba_pred_predict')).alias('woba_pred_predict')
+        ])
+
+        df = df.with_columns([
+            pl.when(pl.col('woba_codes').is_null()).then(None).otherwise(pl.col('woba_pred_predict')).alias('woba_pred'),
+            pl.when(pl.col('bip')!=1).then(None).otherwise(pl.col('woba_pred_predict')).alias('woba_pred_contact'),
+        ])
+
+        df = df.with_columns([
+            pl.when(pl.col('trajectory').is_in(['bunt_popup'])).then(pl.lit('popup'))
+            .when(pl.col('trajectory').is_in(['bunt_grounder'])).then(pl.lit('ground_ball'))
+            .when(pl.col('trajectory').is_in(['bunt_line_drive'])).then(pl.lit('line_drive'))
+            .when(pl.col('trajectory').is_in([''])).then(pl.lit(None))
+            .otherwise(pl.col('trajectory')).alias('trajectory')
+        ])
+
+
+        # Create one-hot encoded columns for the trajectory column
+        dummy_df = df.select(pl.col('trajectory')).to_dummies()
+
+        # Rename the one-hot encoded columns
+        dummy_df = dummy_df.rename({
+            'trajectory_fly_ball': 'trajectory_fly_ball',
+            'trajectory_ground_ball': 'trajectory_ground_ball',
+            'trajectory_line_drive': 'trajectory_line_drive',
+            'trajectory_popup': 'trajectory_popup'
+        })
+
+        # Ensure the columns are present in the DataFrame
+        for col in ['trajectory_fly_ball', 'trajectory_ground_ball', 'trajectory_line_drive', 'trajectory_popup']:
+            if col not in dummy_df.columns:
+                dummy_df = dummy_df.with_columns(pl.lit(0).alias(col))
+
+        # Join the one-hot encoded columns back to the original DataFrame
+        df = df.hstack(dummy_df)
+
+        # Check if 'trajectory_null' column exists and drop it
+        if 'trajectory_null' in df.columns:
+            df = df.drop('trajectory_null')
+
+
+        pitch_cat = {'FA': None,
+                            'FF': 'Fastball',
+                            'FT': 'Fastball',
+                            'FC': 'Fastball',
+                            'FS': 'Off-Speed',
+                            'FO': 'Off-Speed',
+                            'SI': 'Fastball',
+                            'ST': 'Breaking',
+                            'SL': 'Breaking',
+                            'CU': 'Breaking',
+                            'KC': 'Breaking',
+                            'SC': 'Off-Speed',
+                            'GY': 'Off-Speed',
+                            'SV': 'Breaking',
+                            'CS': 'Breaking',
+                            'CH': 'Off-Speed',
+                            'KN': 'Off-Speed',
+                            'EP': 'Breaking',
+                            'UN': None,
+                            'IN': None,
+                            'PO': None,
+                            'AB': None,
+                            'AS': None,
+                            'NP': None}
+        df = df.with_columns(
+            df["pitch_type"].map_elements(lambda x: pitch_cat.get(x, x)).alias("pitch_group")
+        )
+
+        df = df.with_columns([
+            
+            (-(pl.col('vy0')**2 - (2 * pl.col('ay') * (pl.col('y0') - 17/12)))**0.5).alias('vy_f'),
+            ])
+
+        df = df.with_columns([
+            ((pl.col('vy_f') - pl.col('vy0')) / pl.col('ay')).alias('t'),
+            ])
+
+        df = df.with_columns([
+            (pl.col('vz0') + (pl.col('az') * pl.col('t'))).alias('vz_f'),
+            (pl.col('vx0') + (pl.col('ax') * pl.col('t'))).alias('vx_f')
+            ])
+
+        df = df.with_columns([
+                (-np.arctan(pl.col('vz_f') / pl.col('vy_f')) * (180 / np.pi)).alias('vaa'),
+                (-np.arctan(pl.col('vx_f') / pl.col('vy_f')) * (180 / np.pi)).alias('haa')
+            ])
+
+        # Mirror horizontal break for left-handed pitchers
+        df = df.with_columns(
+            pl.when(pl.col('pitcher_hand') == 'L')
+            .then(-pl.col('ax'))
+            .otherwise(pl.col('ax'))
+            .alias('ax')
+        )
+
+        # Mirror horizontal break for left-handed pitchers
+        df = df.with_columns(
+            pl.when(pl.col('pitcher_hand') == 'L')
+            .then(-pl.col('hb'))
+            .otherwise(pl.col('hb'))
+            .alias('hb')
+        )
+
+        # Mirror horizontal release point for left-handed pitchers
+        df = df.with_columns(
+            pl.when(pl.col('pitcher_hand') == 'L')
+            .then(pl.col('x0'))
+            .otherwise(-pl.col('x0'))
+            .alias('x0')
+        )
+
+        df = df.with_columns([
+            pl.when(df['swings'].is_null()).then(None).otherwise(df['swings']).alias('is_swing'),
+            pl.when(df['bip'].is_null()).then(None).otherwise(df['bip']).alias('is_bip')])
+
+
+        return df
+
+    # Assuming df is your Polars DataFrame
+    def update_summary(self, df: pl.DataFrame, pitcher: bool = True) -> pl.DataFrame:
+        """
+        Update summary statistics for pitchers or batters.
+
+        Parameters:
+        df (pl.DataFrame): The input Polars DataFrame containing player statistics.
+        pitcher (bool): A flag indicating whether to calculate statistics for pitchers (True) or batters (False).
+
+        Returns:
+        pl.DataFrame: A Polars DataFrame with aggregated and calculated summary statistics.
+        """
+
+        # Determine the position based on the pitcher flag
+        if pitcher:
+            position = 'pitcher'
+        else:
+            position = 'batter'
+
+        # Group by position_id and position_name, then aggregate various statistics
+        df_summ = df.group_by([f'{position}_id', f'{position}_name']).agg([
+            pl.col('pa').sum().alias('pa'),
+            pl.col('ab').sum().alias('ab'),
+            pl.col('obp').sum().alias('obp_pa'),
+            pl.col('hits').sum().alias('hits'),
+            pl.col('on_base').sum().alias('on_base'),
+            pl.col('k').sum().alias('k'),
+            pl.col('bb').sum().alias('bb'),
+            pl.col('bb_minus_k').sum().alias('bb_minus_k'),
+            pl.col('csw').sum().alias('csw'),
+            pl.col('bip').sum().alias('bip'),
+            pl.col('bip_div').sum().alias('bip_div'),
+            pl.col('tb').sum().alias('tb'),
+            pl.col('woba').sum().alias('woba'),
+            pl.col('woba_contact').sum().alias('woba_contact'),
+            pl.col('woba_pred').sum().alias('xwoba'),
+            pl.col('woba_pred_contact').sum().alias('xwoba_contact'),
+            pl.col('woba_codes').sum().alias('woba_codes'),
+            pl.col('xwoba_codes').sum().alias('xwoba_codes'),
+            pl.col('hard_hit').sum().alias('hard_hit'),
+            pl.col('barrel').sum().alias('barrel'),
+            pl.col('sweet_spot').sum().alias('sweet_spot'),
+            pl.col('launch_speed').max().alias('max_launch_speed'),
+            pl.col('launch_speed').quantile(0.90).alias('launch_speed_90'),
+            pl.col('launch_speed').mean().alias('launch_speed'),
+            pl.col('launch_angle').mean().alias('launch_angle'),
+            pl.col('is_pitch').sum().alias('pitches'),
+            pl.col('swings').sum().alias('swings'),
+            pl.col('in_zone').sum().alias('in_zone'),
+            pl.col('out_zone').sum().alias('out_zone'),
+            pl.col('whiffs').sum().alias('whiffs'),
+            pl.col('zone_swing').sum().alias('zone_swing'),
+            pl.col('zone_contact').sum().alias('zone_contact'),
+            pl.col('ozone_swing').sum().alias('ozone_swing'),
+            pl.col('ozone_contact').sum().alias('ozone_contact'),
+            pl.col('trajectory_ground_ball').sum().alias('ground_ball'),
+            pl.col('trajectory_line_drive').sum().alias('line_drive'),
+            pl.col('trajectory_fly_ball').sum().alias('fly_ball'),
+            pl.col('trajectory_popup').sum().alias('pop_up'),
+            pl.col('attack_zone').count().alias('attack_zone'),
+            pl.col('heart').sum().alias('heart'),
+            pl.col('shadow').sum().alias('shadow'),
+            pl.col('chase').sum().alias('chase'),
+            pl.col('waste').sum().alias('waste'),
+            pl.col('heart_swing').sum().alias('heart_swing'),
+            pl.col('shadow_swing').sum().alias('shadow_swing'),
+            pl.col('chase_swing').sum().alias('chase_swing'),
+            pl.col('waste_swing').sum().alias('waste_swing'),
+            pl.col('heart_whiff').sum().alias('heart_whiff'),
+            pl.col('shadow_whiff').sum().alias('shadow_whiff'),
+            pl.col('chase_whiff').sum().alias('chase_whiff'),
+            pl.col('waste_whiff').sum().alias('waste_whiff')
+        ])
+
+        # Add calculated columns to the summary DataFrame
+        df_summ = df_summ.with_columns([
+            (pl.col('hits') / pl.col('ab')).alias('avg'),
+            (pl.col('on_base') / pl.col('obp_pa')).alias('obp'),
+            (pl.col('tb') / pl.col('ab')).alias('slg'),
+            (pl.col('on_base') / pl.col('obp_pa') + pl.col('tb') / pl.col('ab')).alias('ops'),
+            (pl.col('k') / pl.col('pa')).alias('k_percent'),
+            (pl.col('bb') / pl.col('pa')).alias('bb_percent'),
+            (pl.col('bb_minus_k') / pl.col('pa')).alias('bb_minus_k_percent'),
+            (pl.col('bb') / pl.col('k')).alias('bb_over_k_percent'),
+            (pl.col('csw') / pl.col('pitches')).alias('csw_percent'),
+            (pl.col('sweet_spot') / pl.col('bip_div')).alias('sweet_spot_percent'),
+            (pl.col('woba') / pl.col('woba_codes')).alias('woba_percent'),
+            (pl.col('woba_contact') / pl.col('bip')).alias('woba_percent_contact'),
+            (pl.col('hard_hit') / pl.col('bip_div')).alias('hard_hit_percent'),
+            (pl.col('barrel') / pl.col('bip_div')).alias('barrel_percent'),
+            (pl.col('zone_contact') / pl.col('zone_swing')).alias('zone_contact_percent'),
+            (pl.col('zone_swing') / pl.col('in_zone')).alias('zone_swing_percent'),
+            (pl.col('in_zone') / pl.col('pitches')).alias('zone_percent'),
+            (pl.col('ozone_swing') / (pl.col('pitches') - pl.col('in_zone'))).alias('chase_percent'),
+            (pl.col('ozone_contact') / pl.col('ozone_swing')).alias('chase_contact'),
+            (pl.col('swings') / pl.col('pitches')).alias('swing_percent'),
+            (pl.col('whiffs') / pl.col('swings')).alias('whiff_rate'),
+            (pl.col('whiffs') / pl.col('pitches')).alias('swstr_rate'),
+            (pl.col('ground_ball') / pl.col('bip')).alias('ground_ball_percent'),
+            (pl.col('line_drive') / pl.col('bip')).alias('line_drive_percent'),
+            (pl.col('fly_ball') / pl.col('bip')).alias('fly_ball_percent'),
+            (pl.col('pop_up') / pl.col('bip')).alias('pop_up_percent'),
+            (pl.col('heart') / pl.col('attack_zone')).alias('heart_zone_percent'),
+            (pl.col('shadow') / pl.col('attack_zone')).alias('shadow_zone_percent'),
+            (pl.col('chase') / pl.col('attack_zone')).alias('chase_zone_percent'),
+            (pl.col('waste') / pl.col('attack_zone')).alias('waste_zone_percent'),
+            (pl.col('heart_swing') / pl.col('heart')).alias('heart_zone_swing_percent'),
+            (pl.col('shadow_swing') / pl.col('shadow')).alias('shadow_zone_swing_percent'),
+            (pl.col('chase_swing') / pl.col('chase')).alias('chase_zone_swing_percent'),
+            (pl.col('waste_swing') / pl.col('waste')).alias('waste_zone_swing_percent'),
+            (pl.col('heart_whiff') / pl.col('heart_swing')).alias('heart_zone_whiff_percent'),
+            (pl.col('shadow_whiff') / pl.col('shadow_swing')).alias('shadow_zone_whiff_percent'),
+            (pl.col('chase_whiff') / pl.col('chase_swing')).alias('chase_zone_whiff_percent'),
+            (pl.col('waste_whiff') / pl.col('waste_swing')).alias('waste_zone_whiff_percent'),
+            (pl.col('xwoba') / pl.col('xwoba_codes')).alias('xwoba_percent'),
+            (pl.col('xwoba_contact') / pl.col('bip')).alias('xwoba_percent_contact')
+        ])
+
+        return df_summ
+    
+
+
+
+
+    
+    # Assuming df is your Polars DataFrame
+    def update_summary_select(self, df: pl.DataFrame, selection: list) -> pl.DataFrame:
+        """
+        Update summary statistics for pitchers or batters.
+
+        Parameters:
+        df (pl.DataFrame): The input Polars DataFrame containing player statistics.
+        pitcher (bool): A flag indicating whether to calculate statistics for pitchers (True) or batters (False).
+
+        Returns:
+        pl.DataFrame: A Polars DataFrame with aggregated and calculated summary statistics.
+        """
+
+        # Group by position_id and position_name, then aggregate various statistics
+        df_summ = df.group_by(selection).agg([
+            pl.col('pa').sum().alias('pa'),
+            pl.col('ab').sum().alias('ab'),
+            pl.col('obp').sum().alias('obp_pa'),
+            pl.col('hits').sum().alias('hits'),
+            pl.col('on_base').sum().alias('on_base'),
+            pl.col('k').sum().alias('k'),
+            pl.col('bb').sum().alias('bb'),
+            pl.col('bb_minus_k').sum().alias('bb_minus_k'),
+            pl.col('k_minus_bb').sum().alias('k_minus_bb'),
+            pl.col('csw').sum().alias('csw'),
+            pl.col('bip').sum().alias('bip'),
+            pl.col('bip_div').sum().alias('bip_div'),
+            pl.col('tb').sum().alias('tb'),
+            pl.col('woba').sum().alias('woba'),
+            pl.col('woba_contact').sum().alias('woba_contact'),
+            pl.col('woba_pred').sum().alias('xwoba'),
+            pl.col('woba_pred_contact').sum().alias('xwoba_contact'),
+            pl.col('woba_codes').sum().alias('woba_codes'),
+            pl.col('xwoba_codes').sum().alias('xwoba_codes'),
+            pl.col('hard_hit').sum().alias('hard_hit'),
+            pl.col('barrel').sum().alias('barrel'),
+            pl.col('sweet_spot').sum().alias('sweet_spot'),
+            pl.col('launch_speed').max().alias('max_launch_speed'),
+            pl.col('launch_speed').quantile(0.90).alias('launch_speed_90'),
+            pl.col('launch_speed').mean().alias('launch_speed'),
+            pl.col('launch_angle').mean().alias('launch_angle'),
+            pl.col('is_pitch').sum().alias('pitches'),
+            pl.col('swings').sum().alias('swings'),
+            pl.col('in_zone').sum().alias('in_zone'),
+            pl.col('out_zone').sum().alias('out_zone'),
+            pl.col('whiffs').sum().alias('whiffs'),
+            pl.col('zone_swing').sum().alias('zone_swing'),
+            pl.col('zone_contact').sum().alias('zone_contact'),
+            pl.col('ozone_swing').sum().alias('ozone_swing'),
+            pl.col('ozone_contact').sum().alias('ozone_contact'),
+            pl.col('trajectory_ground_ball').sum().alias('ground_ball'),
+            pl.col('trajectory_line_drive').sum().alias('line_drive'),
+            pl.col('trajectory_fly_ball').sum().alias('fly_ball'),
+            pl.col('trajectory_popup').sum().alias('pop_up'),
+            pl.col('attack_zone').count().alias('attack_zone'),
+            pl.col('heart').sum().alias('heart'),
+            pl.col('shadow').sum().alias('shadow'),
+            pl.col('chase').sum().alias('chase'),
+            pl.col('waste').sum().alias('waste'),
+            pl.col('heart_swing').sum().alias('heart_swing'),
+            pl.col('shadow_swing').sum().alias('shadow_swing'),
+            pl.col('chase_swing').sum().alias('chase_swing'),
+            pl.col('waste_swing').sum().alias('waste_swing'),
+            pl.col('heart_whiff').sum().alias('heart_whiff'),
+            pl.col('shadow_whiff').sum().alias('shadow_whiff'),
+            pl.col('chase_whiff').sum().alias('chase_whiff'),
+            pl.col('waste_whiff').sum().alias('waste_whiff'),
+            pl.col('tj_stuff_plus').sum().alias('tj_stuff_plus'),
+            pl.col('start_speed').sum(),
+            pl.col('vb').sum(),
+            pl.col('ivb').sum(),
+            pl.col('hb').sum(),
+            pl.col('x0').sum(),
+            pl.col('z0').sum(),
+            pl.col('vaa').sum(),
+            pl.col('haa').sum(),
+            pl.col('spin_rate').sum(),
+            pl.col('extension').sum(),
+        ])
+
+        # Add calculated columns to the summary DataFrame
+        df_summ = df_summ.with_columns([
+            (pl.col('hits') / pl.col('ab')).alias('avg'),
+            (pl.col('on_base') / pl.col('obp_pa')).alias('obp'),
+            (pl.col('tb') / pl.col('ab')).alias('slg'),
+            (pl.col('on_base') / pl.col('obp_pa') + pl.col('tb') / pl.col('ab')).alias('ops'),
+            (pl.col('k') / pl.col('pa')).alias('k_percent'),
+            (pl.col('bb') / pl.col('pa')).alias('bb_percent'),
+            (pl.col('bb_minus_k') / pl.col('pa')).alias('bb_minus_k_percent'),
+            (pl.col('k_minus_bb') / pl.col('pa')).alias('k_minus_bb_percent'),
+            (pl.col('bb') / pl.col('k')).alias('bb_over_k_percent'),
+            (pl.col('csw') / pl.col('pitches')).alias('csw_percent'),
+            (pl.col('sweet_spot') / pl.col('bip_div')).alias('sweet_spot_percent'),
+            (pl.col('woba') / pl.col('woba_codes')).alias('woba_percent'),
+            (pl.col('woba_contact') / pl.col('bip')).alias('woba_percent_contact'),
+            (pl.col('hard_hit') / pl.col('bip_div')).alias('hard_hit_percent'),
+            (pl.col('barrel') / pl.col('bip_div')).alias('barrel_percent'),
+            (pl.col('zone_contact') / pl.col('zone_swing')).alias('zone_contact_percent'),
+            (pl.col('zone_swing') / pl.col('in_zone')).alias('zone_swing_percent'),
+            (pl.col('in_zone') / pl.col('pitches')).alias('zone_percent'),
+            (pl.col('ozone_swing') / (pl.col('pitches') - pl.col('in_zone'))).alias('chase_percent'),
+            (pl.col('ozone_contact') / pl.col('ozone_swing')).alias('chase_contact'),
+            (pl.col('swings') / pl.col('pitches')).alias('swing_percent'),
+            (pl.col('whiffs') / pl.col('swings')).alias('whiff_rate'),
+            (pl.col('whiffs') / pl.col('pitches')).alias('swstr_rate'),
+            (pl.col('ground_ball') / pl.col('bip')).alias('ground_ball_percent'),
+            (pl.col('line_drive') / pl.col('bip')).alias('line_drive_percent'),
+            (pl.col('fly_ball') / pl.col('bip')).alias('fly_ball_percent'),
+            (pl.col('pop_up') / pl.col('bip')).alias('pop_up_percent'),
+            (pl.col('heart') / pl.col('attack_zone')).alias('heart_zone_percent'),
+            (pl.col('shadow') / pl.col('attack_zone')).alias('shadow_zone_percent'),
+            (pl.col('chase') / pl.col('attack_zone')).alias('chase_zone_percent'),
+            (pl.col('waste') / pl.col('attack_zone')).alias('waste_zone_percent'),
+            (pl.col('heart_swing') / pl.col('heart')).alias('heart_zone_swing_percent'),
+            (pl.col('shadow_swing') / pl.col('shadow')).alias('shadow_zone_swing_percent'),
+            (pl.col('chase_swing') / pl.col('chase')).alias('chase_zone_swing_percent'),
+            (pl.col('waste_swing') / pl.col('waste')).alias('waste_zone_swing_percent'),
+            (pl.col('heart_whiff') / pl.col('heart_swing')).alias('heart_zone_whiff_percent'),
+            (pl.col('shadow_whiff') / pl.col('shadow_swing')).alias('shadow_zone_whiff_percent'),
+            (pl.col('chase_whiff') / pl.col('chase_swing')).alias('chase_zone_whiff_percent'),
+            (pl.col('waste_whiff') / pl.col('waste_swing')).alias('waste_zone_whiff_percent'),
+            (pl.col('xwoba') / pl.col('xwoba_codes')).alias('xwoba_percent'),
+            (pl.col('xwoba_contact') / pl.col('bip')).alias('xwoba_percent_contact'),
+            (pl.col('tj_stuff_plus') / pl.col('pitches')).alias('tj_stuff_plus_avg'),
+            (pl.col('start_speed')/ pl.col('pitches')).alias('start_speed_avg'),
+            (pl.col('vb')/ pl.col('pitches')).alias('vb_avg'),
+            (pl.col('ivb')/ pl.col('pitches')).alias('ivb_avg'),
+            (pl.col('hb')/ pl.col('pitches')).alias('hb_avg'),
+            (pl.col('x0')/ pl.col('pitches')).alias('x0_avg'),
+            (pl.col('z0')/ pl.col('pitches')).alias('z0_avg'),
+            (pl.col('vaa')/ pl.col('pitches')).alias('vaa_avg'),
+            (pl.col('haa')/ pl.col('pitches')).alias('haa_avg'),
+            (pl.col('spin_rate')/ pl.col('pitches')).alias('spin_rate_avg'),
+            (pl.col('extension')/ pl.col('pitches')).alias('extension_avg'),
+            
+        ])
+
         return df_summ