Add ServiceContext to llama_index imports and instantiate it for use in index.query() for more efficient and customizable responses

app.py

@@ -1,8 +1,7 @@
-from llama_index import Document, SimpleDirectoryReader, GPTListIndex, GPTSimpleVectorIndex, LLMPredictor, PromptHelper
+from llama_index import Document, SimpleDirectoryReader, GPTListIndex, GPTSimpleVectorIndex, LLMPredictor, PromptHelper, ServiceContext
 from llama_index import download_loader
-from langchain import OpenAI
+from langchain.chat_models import ChatOpenAI
 from pathlib import Path
-import openai
 import gradio as gr
 import sys
 import os
@@ -11,6 +10,10 @@ dataFiles = ["RetroApril","RetroMarch", "Snowflake", "Datadog", "Databricks", "S
 
 cache = {}
 
+prompt_helper = PromptHelper(4096, 256, 20)
+llm_predictor = LLMPredictor(llm=ChatOpenAI(temperature=0, model_name="gpt-3.5-turbo"))
+service_context = ServiceContext.from_defaults(llm_predictor=llm_predictor, prompt_helper=prompt_helper)
+
 def indexFile(filePath):
     PandasCSVReader = download_loader("PandasCSVReader")
     loader = PandasCSVReader()
@@ -34,7 +37,7 @@ def chatbot(indexName, input_text):
     Chatbot function that takes in a prompt and returns a response
     """
     index = cache[indexName]
-    response = index.query(input_text, response_mode="compact")
+    response = index.query(input_text, response_mode="compact", service_context=service_context)
     return response.response
 
 

requirements.txt

@@ -1,2 +1,3 @@
 llama-index==0.5.25
+openai==0.27.4
 transformers