Speed up _create_topic_vectors by replacing DataFrame .loc with NumPy masking (#2406)

Author: jinsolp

bertopic/_bertopic.py

@@ -4315,11 +4315,11 @@ def _create_topic_vectors(
         if embeddings is not None and documents is not None:
             topic_embeddings = []
             topics = documents.sort_values("Topic").Topic.unique()
+            topic_ids = documents["Topic"].values
+            doc_ids = documents["ID"].values.astype(int)
             for topic in topics:
-                indices = documents.loc[documents.Topic == topic, "ID"].values
-                indices = [int(index) for index in indices]
-                topic_embedding = np.mean(embeddings[indices], axis=0)
-                topic_embeddings.append(topic_embedding)
+                mask = topic_ids == topic
+                topic_embeddings.append(embeddings[doc_ids[mask]].mean(axis=0))
             self.topic_embeddings_ = np.array(topic_embeddings)
 
         # Topic embeddings when merging topics