Update app.py

app.py

@@ -398,7 +398,7 @@ def server(input, output, session):
     @reactive.event(input.refresh)
     def table_all():
 
-        df_spring = spring_data()
+        df_spring = spring_data().unique(subset=['play_id'])
 
         # df_year_old = stuff_apply.stuff_apply(fe.feature_engineering(pl.concat([df_mlb,df_aaa,df_a,df_afl])))
         # df_year_2old = stuff_apply.stuff_apply(fe.feature_engineering(pl.concat([df_mlb_2023])))
@@ -608,7 +608,7 @@ def server(input, output, session):
     @reactive.event(input.refresh)
     def table_daily():
 
-        df_spring = spring_data()
+        df_spring = spring_data().unique(subset=['play_id'])
 
         # df_year_old = stuff_apply.stuff_apply(fe.feature_engineering(pl.concat([df_mlb,df_aaa,df_a,df_afl])))
         # df_year_2old = stuff_apply.stuff_apply(fe.feature_engineering(pl.concat([df_mlb_2023])))
@@ -619,13 +619,13 @@ def server(input, output, session):
         import polars as pl
 
         # Compute total pitches for each pitcher
-        df_pitcher_totals = df_spring_stuff.group_by(["pitcher_id",'game_date']).agg(
+        df_pitcher_totals = df_spring_stuff.group_by(["pitcher_id",'game_id','game_date']).agg(
             pl.col("start_speed").count().alias("pitcher_total")
         )
 
         df_pitcher_totals_hands = (
             df_spring_stuff
-            .group_by(["pitcher_id", "batter_hand",'game_date'])
+            .group_by(["pitcher_id", "batter_hand",'game_id','game_date'])
             .agg(pl.col("start_speed").count().alias("pitcher_total"))
             .pivot(
                 values="pitcher_total",
@@ -638,7 +638,7 @@ def server(input, output, session):
         )
                 
         
-        df_spring_group = df_spring_stuff.group_by(['pitcher_id', 'pitcher_name', 'pitch_type','game_date']).agg([
+        df_spring_group = df_spring_stuff.group_by(['pitcher_id', 'pitcher_name', 'pitch_type','game_id','game_date']).agg([
             pl.col('start_speed').count().alias('count'),
             pl.col('start_speed').mean().alias('start_speed'),
             pl.col('start_speed').max().alias('max_start_speed'),