Add minimun features

CustomDataset/Text/IMDB.py

@@ -6,7 +6,7 @@ from Utils.Utils import group_themes
 from CustomDataset.Text.Base import Base
 
 class IMDB(Base):
-    def __init__(self, path='rawData/imdb/movie_metadata.csv', bag_of_words_size=300, remove_all_false_values=False):
+    def __init__(self, path='rawData/imdb/movie_metadata.csv', bag_of_words_size=300, min_features=1):
         """
         Initialize the IMDB dataset.
 
@@ -14,7 +14,8 @@ class IMDB(Base):
             path (str): Path to the IMDB directory.
         """
         
-        super().__init__(path=path, bag_of_words_size=bag_of_words_size, remove_all_false_values=remove_all_false_values)
+        super().__init__(path=path, bag_of_words_size=bag_of_words_size)
+        self.min_features=min_features
         self._initialize()
 
     def _initialize(self):
@@ -32,7 +33,7 @@ class IMDB(Base):
         )
 
         # Filter out movies without the desired genres or zero bag-of-words features
-        valid_movie_indices = torch.where(movie_tensor.sum(dim=1) >=2)[0]
+        valid_movie_indices = torch.where(movie_tensor.sum(dim=1) >= self.min_features)[0]
         filtered_index_map = {i.item(): idx for idx, i in enumerate(valid_movie_indices)}
 
         # Apply the filtering logic