last update

projet_en_groupe/algorithme_netflix.py

@@ -378,34 +378,32 @@ def parental_code(data_1):
 
 
 def directors_nationality(data_1):
-    # Check if the 'director' and 'country' columns exist in the dataset
-    if 'director' not in data_1.columns or 'country' not in data_1.columns:
-        print("The dataset does not contain 'director' or 'country' columns.")
-        return
 
     # Extract unique directors and their respective nationalities
     directors_nationality_dict = {}
+    country_nationalities_set = set()
+
     for index, row in data_1.iterrows():
         directors = str(row['director']).split(', ') if pd.notna(row['director']) else []
-        nationality = row['country']
+        nationality = str(row['country']).split(',')
+
+        # Add unique nationalities from 'country' column to the set, excluding 'nan'
+        unique_nationalities = set(filter(lambda x: pd.notna(x) and x.lower() != 'nan', map(str.strip, nationality)))
+        country_nationalities_set.update(unique_nationalities)
 
         for director in directors:
+            director = director.strip()
             if director in directors_nationality_dict:
-                # Ajouter la nationalité seulement si elle n'est pas déjà présente dans la liste
-                if str(nationality) not in directors_nationality_dict[director]['nationalities']:
-                    directors_nationality_dict[director]['nationalities'].add(str(nationality))
+                # Add unique nationalities only if they are not already present
+                directors_nationality_dict[director]['nationalities'].update(unique_nationalities)
                 directors_nationality_dict[director]['number of movies or series'] += 1
             else:
-                directors_nationality_dict[director] = {'nationalities': {str(nationality)}, 'number of movies or series': 1}
-        
-    # Remove duplicates from the nationalities list for each director
-    # Remove consecutive duplicates from the nationalities list for each director
-    for director_info in directors_nationality_dict.values():
-        nationalities_set = set(director_info['nationalities'])
-        director_info['nationalities'] = list(nationalities_set)
+                directors_nationality_dict[director] = {'nationalities': set(unique_nationalities), 'number of movies or series': 1}
+
     # Sort the directors by the number of movies and series produced
     sorted_directors = sorted(directors_nationality_dict.items(), key=lambda x: x[1]['number of movies or series'], reverse=True)
 
+    # Create a DataFrame
     columns = ['director', 'nationalities', 'number of movies or series']
     directors_df = pd.DataFrame([[director, ', '.join(info['nationalities']), info['number of movies or series']] for director, info in sorted_directors], columns=columns)
 
@@ -417,7 +415,7 @@ def directors_nationality(data_1):
         print(f"{i+1}. {director_name}: {nationalities_str} - {info['number of movies or series']} movies/series")
 
     # Save to CSV using the DataFrame
-    save_to_csv(directors_df)  # Commenting this out since the 'save_to_csv' function is not provided
+    save_to_csv(directors_df)
 
     return directors_df