Upload 16 files

app.py

@@ -371,7 +371,7 @@ if document_type == "Single-Document":
             query_text, sparse_retriever_model, sparse_retriever_tokenizer
         )
         dense_query_embedding, sparse_query_embedding = hybrid_score_norm(
-            dense_query_embedding, sparse_query_embedding, 0.1
+            dense_query_embedding, sparse_query_embedding, 0.3
         )
         query_results = query_pinecone_sparse(
             dense_query_embedding,
@@ -430,7 +430,7 @@ else:
                 query_text, sparse_retriever_model, sparse_retriever_tokenizer
             )
             dense_query_embedding, sparse_query_embedding = hybrid_score_norm(
-                dense_query_embedding, sparse_query_embedding, 0.1
+                dense_query_embedding, sparse_query_embedding, 0.3
             )
             year_quarter_list = year_quarter_range(
                 start_quarter, start_year, end_quarter, end_year
@@ -503,7 +503,7 @@ else:
                 query_text, sparse_retriever_model, sparse_retriever_tokenizer
             )
             dense_query_embedding, sparse_query_embedding = hybrid_score_norm(
-                dense_query_embedding, sparse_query_embedding, 0.1
+                dense_query_embedding, sparse_query_embedding, 0.3
             )
             year_quarter_list = year_quarter_range(
                 start_quarter, start_year, end_quarter, end_year

utils/vector_index.py

@@ -42,8 +42,10 @@ def hybrid_score_norm(dense, sparse, alpha: float):
     """
     if alpha < 0 or alpha > 1:
         raise ValueError("Alpha must be between 0 and 1")
-    hs = {
+    hsparse = {
         "indices": sparse["indices"],
         "values": [v * (1 - alpha) for v in sparse["values"]],
     }
-    return [v * alpha for v in dense], hs
+    
+    hdense = [[v * alpha for v in dense[0]]]
+    return hdense, hsparse