Formally adding CPI to weekly job

erikhall6373 · Oct 16, 2023 · cfc509f · cfc509f
1 parent 2924147
commit cfc509f
Show file tree

Hide file tree

Showing 4 changed files with 27 additions and 6 deletions.
diff --git a/heisman_project/Weekly_Data_Job/Weekly_Job.py b/heisman_project/Weekly_Data_Job/Weekly_Job.py
@@ -26,12 +26,12 @@
 
 qb_stats_df = Quarterback_Stats.get_cfb_ref_passing_data(2023)
 qb_stats_df = qb_stats_df[qb_stats_df['School'] != 'School']
-#cpi_df = CPI_Ratings.get_cpi_df(2022)
+cpi_df = CPI_Ratings.current_cpi_df(2023)
 
-#cpi_df = model_data_clean.cpi_cleaning(cpi_df)
+cpi_df = model_data_clean.cpi_cleaning(cpi_df)
 
-cpi_df = Get_Weekly_Win_Percent.calculate_winning_percent(2023)
-cpi_df['CPI'] = (cpi_df['Winning_Percent'] * 0.25) * 100
+#cpi_df = Get_Weekly_Win_Percent.calculate_winning_percent(2023)
+#cpi_df['CPI'] = (cpi_df['Winning_Percent'] * 0.25) * 100
 cpi_df = cpi_df[['School', 'Year', 'CPI']]
 
 weekly_df = qb_stats_df.merge(cpi_df, how = 'inner', on = ['School', 'Year'])

diff --git a/heisman_project/heisman_model/Data_Wrangling/CPI_Ratings.py b/heisman_project/heisman_model/Data_Wrangling/CPI_Ratings.py
@@ -55,6 +55,28 @@ def get_cpi_df(year):
 
     return cpi_df
 
+def current_cpi_df(current_year):
+
+    cpi_tables = pd.read_html(f"http://www.cpiratings.com/archives/current_table.html")
+    cpi_df = cpi_tables[0]
+
+    cpi_df = set_col_names(cpi_df)
+    cpi_df = calc_unrounded_ratings(cpi_df)
+
+    cpi_df = cpi_df[['Team', 'CPI', 'W', 'L', 'OW', 'OL', 'OOW', 'OOL']].reset_index(drop = True)
+    cpi_df['Year'] = current_year
+    cpi_df['Team'] = cpi_df['Team'].apply(lambda x: x.replace('CFP', ""))
+    cpi_df['Team'] = cpi_df['Team'].apply(lambda x: x.replace('AP', ""))
+    cpi_df['Team'] = cpi_df['Team'].apply(lambda x: x.replace('BCS', ""))
+    cpi_df['Team'] = cpi_df['Team'].apply(lambda x: x.replace('(', ""))
+    cpi_df['Team'] = cpi_df['Team'].apply(lambda x: x.replace(')', ""))
+    cpi_df['Team'] = cpi_df['Team'].apply(lambda x: x.replace(',', ""))
+    cpi_df['Team'] = cpi_df['Team'].apply(lambda x: x.strip())
+
+    return cpi_df
+
+
+
 def get_cpi_by_years(years):
 
     cpi_list = [get_cpi_df(year) for year in years]
@@ -66,5 +88,4 @@ def get_cpi_by_years(years):
 
 #test = get_cpi_by_years(range(2012, 2020, 1))
 
-#test[test['Year'] == 2012].head()
-
+#test[test['Year'] == 2012].head()
diff --git a/heisman_project/heisman_model/Data_Wrangling/__pycache__/CPI_Ratings.cpython-311.pyc b/heisman_project/heisman_model/Data_Wrangling/__pycache__/CPI_Ratings.cpython-311.pyc
diff --git a/...n_project/heisman_model/Data_Wrangling/__pycache__/Get_Weekly_Win_Percent.cpython-311.pyc b/...n_project/heisman_model/Data_Wrangling/__pycache__/Get_Weekly_Win_Percent.cpython-311.pyc