app.py

# https://towardsai.net/p/machine-learning/deploying-a-langchain-large-language-model-llm-with-streamlit-pinecone?amp=1

"""Python file to serve as the frontend"""
import os
import pinecone
import streamlit as st
from streamlit_chat import message
from langchain.llms import OpenAI
from langchain.embeddings.openai import OpenAIEmbeddings
from langchain.vectorstores import Pinecone
from langchain.chains import ConversationChain
from langchain.retrievers.self_query.base import SelfQueryRetriever
from langchain.chains.query_constructor.base import AttributeInfo

pinecone.init(
     api_key=str(os.environ['PINECONE_API_KEY']), 
     environment=str(os.environ['PINECONE_ENV']))

llm = OpenAI(temperature=0)
embeddings = OpenAIEmbeddings()
vectorstore = Pinecone.from_existing_index("impromptu", embeddings)
document_content_description = ("Excerpts from the book Impromptu, "
                                "jointly authored by Reid Hoffman, " 
                                "GPT-3, and GPT 4.")
metadata_field_info=[
    AttributeInfo(
        name="author",
        description="The author of the excerpt", 
        type="string or list[string]", 
    ),
    AttributeInfo(
        name="chapter_number",
        description="The chapter number of excerpt", 
        type="integer", 
    ),
    AttributeInfo(
        name="chapter_name",
        description="The chapter name of the excerpt", 
        type="string", 
    ),
]


# metadata_field_info=[
#     AttributeInfo(
#         name="author",
#         description="The author of the excerpt", 
#         type="string or list[string]", 
#     ),
#     AttributeInfo(
#         name="chapter_number",
#         description="The chapter number of excerpt", 
#         type="integer", 
#     ),
#     AttributeInfo(
#         name="chapter_name",
#         description="The chapter name of the excerpt", 
#         type="string", 
#     ),
# ]

# document_content_description = "Excerpt's from Reid Hoffman's book Impromptu"
# embeddings = OpenAIEmbeddings()

# index_name = str(os.environ['PINECONE_INDEX_NAME'])

def load_chain():
    # vectorstore = Pinecone.from_existing_index(index_name, embeddings)
    # llm = OpenAI(temperature=0)
    retriever = SelfQueryRetriever.from_llm(
         llm, 
         vectorstore, 
         document_content_description, 
         metadata_field_info, 
         verbose=True)
    return retriever
    
 # docsearch = Pinecone.from_existing_index(index_name, embeddings)
 # return docsearch

# def load_chain():
#     """Logic for loading the chain you want to use should go here."""
#     llm = OpenAI(temperature=0)
#     chain = ConversationChain(llm=llm)
#     return chain

chain = load_chain()

# From here down is all the StreamLit UI.
st.set_page_config(page_title="LangChain Demo", page_icon=":robot:")
st.header("LangChain Demo")

if "generated" not in st.session_state:
    st.session_state["generated"] = []

if "past" not in st.session_state:
    st.session_state["past"] = []


def get_text():
    input_text = st.text_input("You: ", "What are the four types of hallucinations?", key="input")
    return input_text


user_input = get_text()

if user_input:
    docs = chain.get_relevant_documents(user_input)
    print('docs')
    print(docs)
    print('docs[0]')
    print(docs[0])
    print('docs[0].page_content')
    print(docs[0].page_content)

    output = docs[0].page_content
    # output = 'nothing right now'

    st.session_state.past.append(user_input)
    st.session_state.generated.append(output)

if st.session_state["generated"]:

    for i in range(len(st.session_state["generated"]) - 1, -1, -1):
        message(st.session_state["generated"][i], key=str(i))
        message(st.session_state["past"][i], is_user=True, key=str(i) + "_user")