Merge pull request #1 from tylerjrichards/main

Updated Streamlit Example, deployed app link added to readme

Author: logan-markewich

.gitignore

@@ -0,0 +1,4 @@
+.envrc
+.direnv
+.streamlit/secrets.toml
+.mypy_cache

README.md

@@ -1,5 +1,5 @@
 # llama_index_starter_pack
-This repository provides very basic flask, streamlit, and docker examples for the [llama_index](https://github.com/jerryjliu/gpt_index) (FKA gpt_index) package.
+This repository provides very basic flask, [Streamlit](https://llama-index.streamlit.app/), and docker examples for the [llama_index](https://github.com/jerryjliu/gpt_index) (FKA gpt_index) package.
 
 If you need to quickly create a POC to impress your boss, start here!
 
@@ -20,13 +20,14 @@ There are two main example files
   - creates a simple api that loads the text from the documents folder
   - The "/query" endpoint accepts requests that contain a "text" parameter, which is used to query the index
   - resturns string response containing the query answer
-    
+
 - streamlit_demo.py (localhost:8501)
   - `streamlit run streamlit_demo.py`
   - creates a simple UI using streamlit
   - loads text from the documents folder (using `st.cache_resource`, so it only loads once)
   - provides an input text-box and a button to run the query
   - The string response is displayed after it finishes
+  - Want to see this example in action? Check it out [here](https://llama-index.streamlit.app/)
 
 
 ## Docker

streamlit_demo.py

@@ -1,10 +1,13 @@
 import os
 
+import streamlit as st
+from llama_index import GPTSimpleVectorIndex, SimpleDirectoryReader
+
 # NOTE: for local testing only, do NOT deploy with your key hardcoded
-os.environ['OPENAI_API_KEY'] = "your_key_here"
+# to use this for yourself, create a file called .streamlit/secrets.toml with your api key
+# Learn more about Streamlit on the docs: https://docs.streamlit.io/
+os.environ["OPENAI_API_KEY"] = st.secrets["openai_api_key"]
 
-import streamlit as st
-from llama_index import SimpleDirectoryReader, GPTSimpleVectorIndex
 
 index_name = "./index.json"
 documents_folder = "./documents"
@@ -32,15 +35,12 @@ def query_index(_index, query_text):
 index = initialize_index(index_name, documents_folder)
 
 
-st.title("Llama Index")
-
-st.header("Welcome to the Llama Index streamlit")
-
-st.text("Please enter a query about Paul Graham's essay?")
+st.title("🦙 Llama Index Demo 🦙")
+st.header("Welcome to the Llama Index Streamlit Demo")
+st.text("Please enter a query about Paul Graham's essays")
 
 text = st.text_input("Query text:")
 
 if st.button("Run Query") and text is not None:
     response = query_index(index, text)
     st.markdown(response)
-