Upload app.py

app.py

@@ -21,14 +21,11 @@ import pandas as pd
 import shinyswatch
 import inflect
 from matplotlib.pyplot import text
-import rolling_batter_functions as rbf
+from functions import rolling_batter_functions as rbf
 import joblib
 import polars as pl
-
-in_zone_model = joblib.load('joblib_model/in_zone_model_knn_20240410.joblib')
-px_model = joblib.load('joblib_model/linear_reg_model_x.joblib')
-pz_model = joblib.load('joblib_model/linear_reg_model_z.joblib')
-xwoba_model = joblib.load('joblib_model/xwoba_model.joblib')
+from functions import df_update
+update = df_update.df_update()
 
 def percentile(n):
     def percentile_(x):
@@ -42,133 +39,42 @@ colour_palette = ['#FFB000','#648FFF','#785EF0',
 
 
 print('Starting Everything:')
-# exit_velo_df = milb_a_ev_df.append([triple_a_ev_df,double_a_ev_df,a_high_a_ev_df,single_a_ev_df]).reset_index(drop=True)
-# player_df_all = mlb_a_player_df.append([triple_a_player_df,double_a_player_df,a_high_a_player_df,single_a_player_df]).reset_index(drop=True)
-# exit_velo_df = pd.read_csv('exit_velo_df_all.csv',index_col=[0])
-# player_df_all = pd.read_csv('player_df_all.csv',index_col=[0])
-
-# pa_df = pd.read_csv('pa_df_all.csv',index_col=[0])
-# pa_df_full_na = pa_df.dropna()
 
 
-exit_velo_df_mlb = pl.read_parquet(f"hf://datasets/TJStatsApps/mlb_data/data/mlb_pitch_data_2025.parquet").to_pandas()
+exit_velo_df_mlb = pl.read_parquet(f"hf://datasets/TJStatsApps/mlb_data/data/mlb_pitch_data_2025.parquet")
 #print(df_2023)
-exit_velo_df_mlb['level'] = 'MLB'
+exit_velo_df_mlb = exit_velo_df_mlb.with_columns(pl.lit('MLB').alias('level'))
 
 
-exit_velo_df_aaa = pl.read_parquet(f"hf://datasets/TJStatsApps/mlb_data/data/aaa_pitch_data_2025.parquet").to_pandas()
+exit_velo_df_aaa = pl.read_parquet(f"hf://datasets/TJStatsApps/mlb_data/data/aaa_pitch_data_2025.parquet")
 #print(df_2023)
-exit_velo_df_aaa['level'] = 'AAA'
+exit_velo_df_aaa = exit_velo_df_aaa.with_columns(pl.lit('AAA').alias('level'))
 
 
-exit_velo_df_aa = pl.read_parquet(f"hf://datasets/TJStatsApps/mlb_data/data/aa_pitch_data_2025.parquet").to_pandas()
+exit_velo_df_aa = pl.read_parquet(f"hf://datasets/TJStatsApps/mlb_data/data/aa_pitch_data_2025.parquet")
 #print(df_2023)
-exit_velo_df_aa['level'] = 'AA'
+exit_velo_df_aa = exit_velo_df_aa.with_columns(pl.lit('AA').alias('level'))
 
 
-exit_velo_df_ha = pl.read_parquet(f"hf://datasets/TJStatsApps/mlb_data/data/hi_a_pitch_data_2025.parquet").to_pandas()
+exit_velo_df_ha = pl.read_parquet(f"hf://datasets/TJStatsApps/mlb_data/data/hi_a_pitch_data_2025.parquet")
 #print(df_2023)
-exit_velo_df_ha['level'] = 'A+'
+exit_velo_df_ha = exit_velo_df_ha.with_columns(pl.lit('A+').alias('level'))
 
 ### Import Datasets
-exit_velo_df_a = pl.read_parquet(f"hf://datasets/TJStatsApps/mlb_data/data/lo_a_pitch_data_2025.parquet").to_pandas()
+exit_velo_df_a = pl.read_parquet(f"hf://datasets/TJStatsApps/mlb_data/data/lo_a_pitch_data_2025.parquet")
 #print(df_2023)
-exit_velo_df_a['level'] = 'A'
-
-exit_velo_df = pd.concat([exit_velo_df_mlb,exit_velo_df_aaa,exit_velo_df_aa,exit_velo_df_ha,exit_velo_df_a])
-# exit_velo_df = pd.concat([exit_velo_df_mlb,exit_velo_df_aaa])
-# exit_velo_df_copy = exit_velo_df.copy()
-
-# exit_velo_df = exit_velo_df_copy.copy()
-
-
-from batting_update import df_update,df_update_summ_avg,df_update_summ,df_summ_batter_pitch_up,df_summ_changes,df_summ_filter_out
-
-def percentile(n):
-    def percentile_(x):
-        return np.nanpercentile(x, n)
-    percentile_.__name__ = 'percentile_%s' % n
-    return percentile_
-
-exit_velo_df_codes = df_update(exit_velo_df)
-
-
-exit_velo_df_codes_summ = exit_velo_df_codes.groupby(['batter_id','batter_name','level']).agg(
-    pa = ('pa','sum'), 
-    k = ('k','sum'), 
-    bb = ('bb','sum'), 
-    bb_minus_k = ('bb_minus_k','sum'), 
-    csw = ('csw','sum'), 
-    bip = ('bip','sum'),
-    tb = ('tb','sum'),
-    woba = ('woba','sum'),
-    woba_codes = ('woba_codes','sum'),
-    xwoba = ('xwoba','sum'),
-    xwoba_contact = ('xwoba_contact','sum'),       
-    #woba_codes = ('woba_codes','sum'),
-    hard_hit = ('hard_hit','sum'), 
-    barrel = ('barrel','sum'), 
-    sweet_spot = ('sweet_spot','sum'), 
-    max_launch_speed = ('launch_speed','max'),
-    launch_speed_90 =  ('launch_speed',percentile(90)),
-    launch_speed = ('launch_speed','mean'),
-    launch_angle = ('launch_angle','mean'),
-    pitches = ('pitches','sum'),
-    swings =  ('swings','sum'),
-    in_zone = ('in_zone','sum'),
-    out_zone = ('out_zone','sum'),
-    whiffs = ('whiffs','sum'),
-    zone_swing = ('zone_swing','sum'),
-    zone_contact = ('zone_contact','sum'),
-    ozone_swing = ('ozone_swing','sum'),
-    ozone_contact = ('ozone_contact','sum'),
-    ).reset_index()
-
-#exit_velo_df_codes_summ['out_zone'] = ~exit_velo_df_codes_summ.in_zone
-#bip_min_input = int(input())
-#bip_min = min(bip_min_input,50)
-#exit_velo_df_codes_summ =  exit_velo_df_codes_summ[exit_velo_df_codes_summ.balls_in_play>=bip_min]
-
-
-exit_velo_df_codes_summ['k_percent'] = [exit_velo_df_codes_summ.k[x]/exit_velo_df_codes_summ.pa[x] if exit_velo_df_codes_summ.pa[x] != 0 else np.nan for x in range(len(exit_velo_df_codes_summ))]
-exit_velo_df_codes_summ['bb_percent'] =[exit_velo_df_codes_summ.bb[x]/exit_velo_df_codes_summ.pa[x] if exit_velo_df_codes_summ.pa[x] != 0 else np.nan for x in range(len(exit_velo_df_codes_summ))]
-exit_velo_df_codes_summ['bb_minus_k_percent'] =[exit_velo_df_codes_summ.bb_minus_k[x]/exit_velo_df_codes_summ.pa[x] if exit_velo_df_codes_summ.pa[x] != 0 else np.nan for x in range(len(exit_velo_df_codes_summ))]
-
-exit_velo_df_codes_summ['csw_percent'] =[exit_velo_df_codes_summ.csw[x]/exit_velo_df_codes_summ.pitches[x] if exit_velo_df_codes_summ.pitches[x] != 0 else np.nan for x in range(len(exit_velo_df_codes_summ))]
-
-
-exit_velo_df_codes_summ['sweet_spot_percent'] = [exit_velo_df_codes_summ.sweet_spot[x]/exit_velo_df_codes_summ.bip[x] if exit_velo_df_codes_summ.bip[x] != 0 else np.nan for x in range(len(exit_velo_df_codes_summ))]
-
-exit_velo_df_codes_summ['woba_percent'] = [exit_velo_df_codes_summ.woba[x]/exit_velo_df_codes_summ.woba_codes[x] if exit_velo_df_codes_summ.woba_codes[x] != 0 else np.nan for x in range(len(exit_velo_df_codes_summ))]
-#exit_velo_df_codes_summ['hard_hit_percent'] = [exit_velo_df_codes_summ.sweet_spot[x]/exit_velo_df_codes_summ.bip[x] if exit_velo_df_codes_summ.bip[x] != 0 else np.nan for x in range(len(exit_velo_df_codes_summ))]
-exit_velo_df_codes_summ['hard_hit_percent'] = [exit_velo_df_codes_summ.hard_hit[x]/exit_velo_df_codes_summ.bip[x] if exit_velo_df_codes_summ.bip[x] != 0 else np.nan for x in range(len(exit_velo_df_codes_summ))]
-
-
-exit_velo_df_codes_summ['barrel_percent'] = [exit_velo_df_codes_summ.barrel[x]/exit_velo_df_codes_summ.bip[x] if exit_velo_df_codes_summ.bip[x] != 0 else np.nan for x in range(len(exit_velo_df_codes_summ))]
-
-exit_velo_df_codes_summ['zone_contact_percent'] = [exit_velo_df_codes_summ.zone_contact[x]/exit_velo_df_codes_summ.zone_swing[x] if exit_velo_df_codes_summ.zone_swing[x] != 0 else np.nan for x in range(len(exit_velo_df_codes_summ))]
-
-exit_velo_df_codes_summ['zone_swing_percent'] = [exit_velo_df_codes_summ.zone_swing[x]/exit_velo_df_codes_summ.in_zone[x] if exit_velo_df_codes_summ.pitches[x] != 0 else np.nan for x in range(len(exit_velo_df_codes_summ))]
-
-exit_velo_df_codes_summ['zone_percent'] = [exit_velo_df_codes_summ.in_zone[x]/exit_velo_df_codes_summ.pitches[x] if exit_velo_df_codes_summ.pitches[x] != 0 else np.nan for x in range(len(exit_velo_df_codes_summ))]
-
-exit_velo_df_codes_summ['chase_percent'] = [exit_velo_df_codes_summ.ozone_swing[x]/(exit_velo_df_codes_summ.pitches[x] - exit_velo_df_codes_summ.in_zone[x]) if (exit_velo_df_codes_summ.pitches[x]- exit_velo_df_codes_summ.in_zone[x]) != 0 else np.nan for x in range(len(exit_velo_df_codes_summ))]
-
-exit_velo_df_codes_summ['chase_contact'] = [exit_velo_df_codes_summ.ozone_contact[x]/exit_velo_df_codes_summ.ozone_swing[x] if exit_velo_df_codes_summ.ozone_swing[x] != 0 else np.nan for x in range(len(exit_velo_df_codes_summ))]
-
-exit_velo_df_codes_summ['swing_percent'] = [exit_velo_df_codes_summ.swings[x]/exit_velo_df_codes_summ.pitches[x] if exit_velo_df_codes_summ.pitches[x] != 0 else np.nan for x in range(len(exit_velo_df_codes_summ))]
-
-exit_velo_df_codes_summ['whiff_rate'] = [exit_velo_df_codes_summ.whiffs[x]/exit_velo_df_codes_summ.swings[x] if exit_velo_df_codes_summ.swings[x] != 0 else np.nan for x in range(len(exit_velo_df_codes_summ))]
+exit_velo_df_a = exit_velo_df_a.with_columns(pl.lit('A').alias('level'))
 
-exit_velo_df_codes_summ['swstr_rate'] = [exit_velo_df_codes_summ.whiffs[x]/exit_velo_df_codes_summ.pitches[x] if exit_velo_df_codes_summ.pitches[x] != 0 else np.nan for x in range(len(exit_velo_df_codes_summ))]
 
-exit_velo_df_codes_summ['xwoba_percent'] = [exit_velo_df_codes_summ.xwoba[x]/exit_velo_df_codes_summ.woba_codes[x] if exit_velo_df_codes_summ.woba_codes[x] != 0 else np.nan for x in range(len(exit_velo_df_codes_summ))]
-exit_velo_df_codes_summ['xwoba_percent_contact'] = [exit_velo_df_codes_summ.xwoba_contact[x]/exit_velo_df_codes_summ.bip[x] if exit_velo_df_codes_summ.bip[x] != 0 else np.nan for x in range(len(exit_velo_df_codes_summ))]
 
+exit_velo_df = pl.concat([exit_velo_df_mlb,exit_velo_df_aaa,exit_velo_df_aa,exit_velo_df_ha,exit_velo_df_a])
+# exit_velo_df = pl.concat([exit_velo_df_mlb])#,exit_velo_df_aaa,exit_velo_df_aa,exit_velo_df_ha,exit_velo_df_a])
 
-exit_velo_df_codes_summ = exit_velo_df_codes_summ.dropna(subset=['bip'])
+exit_velo_df_codes = update.update(exit_velo_df)
+exit_velo_df_codes_summ = update.update_summary_select(df=exit_velo_df_codes,selection=['batter_id','batter_name','level']).to_pandas()
+exit_velo_df_codes = exit_velo_df_codes.to_pandas()
 
-woba_list = ['woba','xwoba']
+woba_list = ['woba']
 pa_list = ['k','bb','bb_minus_k']
 balls_in_play_list = ['hard_hit','launch_speed','launch_speed_90','launch_angle','barrel','sweet_spot']
 pitches_list = ['zone_percent','swing_percent','sw_str','csw']
@@ -184,7 +90,6 @@ plot_dict = {
 'bb_minus_k':{'x_axis':'Plate Appearances','y_axis':'BB-K%','title':'BB-K%','x_value':'bb_minus_k','x_range':[-0.3,-0.2,-0.1,0,0.1,0.2],'percent':True,'percentile_label':'bb_minus_k_percent','flip_p':False,'percentile':False,'avg_adjust':False},
 'csw':{'x_axis':'Pitches','y_axis':'CSW%','title':'CSW%','x_value':'csw','x_range':[.2,.25,.3,.35,.4],'percent':True,'percentile_label':'csw_percent','flip_p':True,'percentile':False,'avg_adjust':False},
 'woba':{'x_axis':'wOBA PA','y_axis':'wOBA','title':'wOBA','x_value':'woba','x_range':[.20,.30,.40,.50],'percent':False,'percentile_label':'woba_percent','flip_p':False,'percentile':False,'avg_adjust':True},
-'xwoba':{'x_axis':'xwOBA PA','y_axis':'xwOBA','title':'xwOBA','x_value':'xwoba','x_range':[.20,.30,.40,.50],'percent':False,'percentile_label':'xwoba_percent','flip_p':False,'percentile':False,'avg_adjust':True},
 'launch_speed':{'x_axis':'Balls In Play','y_axis':'Exit Velocity','title':'Exit Velocity','x_value':'launch_speed','x_range':[85,90,95,100],'percent':False,'percentile_label':'launch_speed','flip_p':False,'percentile':False,'avg_adjust':False},
 'launch_speed_90':{'x_axis':'Balls In Play','y_axis':'90th Percentile Exit Velocity','title':'90th Percentile Exit Velocity','x_value':'launch_speed','x_range':[95,100,105,110,115],'percent':False,'percentile_label':'launch_speed_90','flip_p':False,'percentile':True,'avg_adjust':False},              
 'hard_hit':{'x_axis':'Balls In Play','y_axis':'HardHit%','title':'HardHit%','x_value':'hard_hit','x_range':[0.2,0.3,0.4,0.5,0.6,0.7],'percent':True,'percentile_label':'hard_hit_percent','flip_p':False,'percentile':False,'avg_adjust':False},
@@ -209,11 +114,12 @@ plot_dict = {
 # test_df = test_df.set_index('batter_id')
 # #test_df =  test_df[test_df.pitcher == 'Chris Bassitt'].append(test_df[test_df.pitcher != 'Chris Bassitt'])
 
-batter_dict_mlb = exit_velo_df_mlb.sort_values(['batter_name']).set_index('batter_id')['batter_name'].to_dict()
-batter_dict_aaa = exit_velo_df_aaa.sort_values(['batter_name']).set_index('batter_id')['batter_name'].to_dict()
-batter_dict_aa = exit_velo_df_aa.sort_values(['batter_name']).set_index('batter_id')['batter_name'].to_dict()
-batter_dict_ha = exit_velo_df_ha.sort_values(['batter_name']).set_index('batter_id')['batter_name'].to_dict()
-batter_dict_a = exit_velo_df_a.sort_values(['batter_name']).set_index('batter_id')['batter_name'].to_dict()
+batter_dict_mlb = dict(zip(exit_velo_df_mlb.sort(['batter_name'])['batter_id'],exit_velo_df_mlb.sort(['batter_name'])['batter_name']))
+# batter_dict_mlb = exit_velo_df_mlb.sort(['batter_name']).set_index('batter_id')['batter_name'].to_dict()
+batter_dict_aaa = dict(zip(exit_velo_df_aaa.sort(['batter_name'])['batter_id'],exit_velo_df_aaa.sort(['batter_name'])['batter_name']))
+batter_dict_aa = dict(zip(exit_velo_df_aa.sort(['batter_name'])['batter_id'],exit_velo_df_aa.sort(['batter_name'])['batter_name']))
+batter_dict_ha = dict(zip(exit_velo_df_ha.sort(['batter_name'])['batter_id'],exit_velo_df_ha.sort(['batter_name'])['batter_name']))
+batter_dict_a =dict(zip(exit_velo_df_a.sort(['batter_name'])['batter_id'],exit_velo_df_a.sort(['batter_name'])['batter_name']))
 
 level_dict = {'MLB':'MLB','AAA':'AAA','AA':'AA','A+':'A+','A':'A'}
 
@@ -223,7 +129,6 @@ plot_dict_small = {
 'bb_minus_k':'BB-K%',
 'csw':'CSW%',
 'woba':'wOBA',
-'xwoba':'xwOBA',
 'launch_speed':'Exit Velocity',
 'launch_speed_90':'90th Percentile Exit Velocity',
 'hard_hit':'HardHit%',