Change CrossEncoder to the smaller model for faster response

backend/semantic_search.py

@@ -36,8 +36,8 @@ def retrieve(query, k):
         raise gr.Error(str(e))
 
 def reranking(query, retrieval_result):
-  model_name = 'BAAI/bge-reranker-large'
-  # model_name = 'cross-encoder/ms-marco-MiniLM-L-6-v2'
+  # model_name = 'BAAI/bge-reranker-large'
+  model_name = 'cross-encoder/ms-marco-MiniLM-L-6-v2'
   model = CrossEncoder(model_name, max_length=512)
 
   # Prepare the list of tuples (query, document) for the model