Update custom_utils.py

custom_utils.py

@@ -25,6 +25,7 @@ def rag_retrieval_naive(openai_api_key,
                         collection, 
                         vector_index="vector_index"):
     # Naive RAG: Semantic search
+    
     retrieval_result = vector_search_naive(
         openai_api_key, 
         prompt, 
@@ -36,9 +37,7 @@ def rag_retrieval_naive(openai_api_key,
     if not retrieval_result:
         return "No results found."
 
-    #print("###")
-    #print(retrieval_result)
-    #print("###")
+    print(retrieval_result)
     
     return retrieval_result
 
@@ -52,12 +51,14 @@ def rag_retrieval_advanced(openai_api_key,
     # Advanced RAG: Semantic search plus...
     
     # 1a) Pre-retrieval processing: index filter (accomodates, bedrooms) plus...
-    
     # 1b) Post-retrieval processing: result filter (accomodates, bedrooms) plus...
-    
-    # 2) Weighted average review, sorted in descending order
+    # 2)  Weighted average review, sorted in descending order
 
-    additional_stages = [get_average_review_stage(), get_weighting_stage(), get_sorting_stage()]
+    additional_stages = [
+        get_average_review_and_review_count_stage(),
+        get_weighting_stage(),
+        get_sorting_stage()
+    ]
     
     retrieval_result = vector_search_advanced(
         openai_api_key, 
@@ -73,9 +74,7 @@ def rag_retrieval_advanced(openai_api_key,
     if not retrieval_result:
         return "No results found."
 
-    #print("###")
-    #print(retrieval_result)
-    #print("###")
+    print(retrieval_result)
     
     return retrieval_result
 
@@ -140,7 +139,7 @@ def vector_search_naive(openai_api_key,
         }
     }
 
-    pipeline = [vector_search_stage, get_remove_embedding_stage()]
+    pipeline = [vector_search_stage, get_remove_fields_stage()]
 
     return invoke_search(db, collection, pipeline)
 
@@ -157,7 +156,7 @@ def vector_search_advanced(openai_api_key,
     if query_embedding is None:
         return "Invalid query or embedding generation failed."
 
-    vector_search_filter_stage = {
+    vector_search_and_filter_stage = {
         "$vectorSearch": {
             "index": vector_index,
             "queryVector": query_embedding,
@@ -173,16 +172,38 @@ def vector_search_advanced(openai_api_key,
         }
     }
   
-    pipeline = [vector_search_filter_stage, get_remove_embedding_stage()] + additional_stages
+    pipeline = [vector_search_and_filter_stage, get_remove_fields_stage()] + additional_stages
 
     return invoke_search(db, collection, pipeline)
 
-def get_remove_embedding_stage():
+def get_remove_fields_stage():
     return {
         "$unset": "description_embedding"
     }
+    
+def get_project_fields_stage():
+    return {
+        "$project": {
+            "_id": 0,  
+            "name": 1,
+            "accommodates": 1,
+            "address.street": 1,
+            "address.government_area": 1, 
+            "address.market": 1,
+            "address.country": 1, 
+            "address.country_code": 1, 
+            "address.location.type": 1, 
+            "address.location.coordinates": 1,  
+            "address.location.is_location_exact": 1,
+            "summary": 1,
+            "space": 1,  
+            "neighborhood_overview": 1, 
+            "notes": 1, 
+            "score": {"$meta": "vectorSearchScore"} 
+        }
+    }
 
-def get_result_filter_stage():
+def get_filter_result_stage():
     return {
         "$match": {
             "accommodates": { "$eq": 2},
@@ -190,7 +211,7 @@ def get_result_filter_stage():
         }
     }
 
-def get_average_review_stage():
+def get_average_review_and_review_count_stage():
     return {
         "$addFields": {
             "averageReview": {