Update kb2.py

kb2.py

@@ -18,15 +18,22 @@ RATE_LIMIT = 3
 #@limits(calls=RATE_LIMIT, period=1)
 def create_service_context():
 
-    # Constraint parameters
+    # Constraint parameters ORIGINAL
+#    max_input_size = 4096
+#    num_outputs = 512
+#    max_chunk_overlap = 20
+#    chunk_size_limit = 600
+
+    # Allows the user to explicitly set certain constraint parameters
+#    prompt_helper = PromptHelper(max_input_size, num_outputs, max_chunk_overlap, chunk_size_limit=chunk_size_limit)
     max_input_size = 4096
     num_outputs = 512
     max_chunk_overlap = 20
-    chunk_size_limit = 600
-
-    # Allows the user to explicitly set certain constraint parameters
-    prompt_helper = PromptHelper(max_input_size, num_outputs, max_chunk_overlap, chunk_size_limit=chunk_size_limit)
+    chunk_size_limit = 600 
+    prompt_helper = PromptHelper(max_input_size, num_outputs, chunk_overlap_ratio= 0.1, chunk_size_limit=chunk_size_limit)
+#    llm_predictor = LLMPredictor(llm=ChatOpenAI(temperature=0.7, model_name="gpt-4", max_tokens=num_outputs))
 
+    
     # LLMPredictor is a wrapper class around LangChain's LLMChain that allows easy integration into LlamaIndex
     llm_predictor = LLMPredictor(llm=ChatOpenAI(temperature=0.5, model_name="gpt-3.5-turbo", max_tokens=num_outputs))