src/streamlit_app.py

import streamlit as st
import os
from dotenv import load_dotenv

# import pinecone
from pinecone import Pinecone

# import langchain
from langchain_pinecone import PineconeVectorStore
from langchain_groq import ChatGroq
from langchain_community.embeddings import HuggingFaceEmbeddings
from langchain_core.messages import HumanMessage, SystemMessage, AIMessage
# from langchain_huggingface import HuggingFaceEmbeddings

load_dotenv()

st.title("Groq Chatbot with Pinecone (Free Embeddings)")

# initialize pinecone database
pc = Pinecone(api_key=os.environ.get("PINECONE_API_KEY"))
index_name = os.environ.get("PINECONE_INDEX_NAME")
index = pc.Index(index_name)

# initialize embeddings model (local + free)
embeddings = HuggingFaceEmbeddings(
    model_name="sentence-transformers/all-mpnet-base-v2"
)

# create Pinecone vector store
vector_store = PineconeVectorStore(index=index, embedding=embeddings)

# initialize chat history
if "messages" not in st.session_state:
    st.session_state.messages = []
    st.session_state.messages.append(SystemMessage("You are a helpful assistant for answering user questions."))

# display chat messages
for message in st.session_state.messages:
    role = "user" if isinstance(message, HumanMessage) else "assistant"
    with st.chat_message(role):
        st.markdown(message.content)

# chat input
prompt = st.chat_input("Ask me something...")

if prompt:
    with st.chat_message("user"):
        st.markdown(prompt)
        st.session_state.messages.append(HumanMessage(prompt))

    # initialize Groq LLM
    llm = ChatGroq(
        api_key=os.environ.get("GROQ_API_KEY"),
        model="llama-3.3-70b-versatile",
        temperature=0.7
    )

    # retrieve similar documents
    retriever = vector_store.as_retriever(
        search_type="similarity_score_threshold",
        search_kwargs={"k": 3, "score_threshold": 0.5},
    )

    docs = retriever.invoke(prompt)
    docs_text = "".join(d.page_content for d in docs)

    # build system prompt
    system_prompt = f"""
    You are an assistant for question-answering tasks.
    Use the following context to answer the question.
    If you don't know the answer, say you don't know.
    Keep your answer concise and helpful.
    Context: {docs_text}
    """

    # st.session_state.messages.append(SystemMessage(system_prompt))

    # # generate answer using Groq Cloud
    # result = llm.invoke(st.session_state.messages).content
    # build temporary message list for this query
    messages = [SystemMessage(system_prompt)] + st.session_state.messages

# generate answer using Groq Cloud
    result = llm.invoke(messages).content

    # display answer
    with st.chat_message("assistant"):
        st.markdown(result)
        st.session_state.messages.append(AIMessage(result))