# Langchain chat bot

course: Academy — 54-Next-Gen-AI-GenAI-Agents-Future-Trends
module: Academy/54-Next-Gen-AI-GenAI-Agents-Future-Trends
type: notebook
source_url: https://personal-learn.armco.dev/files/Academy/54-Next-Gen-AI-GenAI-Agents-Future-Trends/folders/Session_9_Folder/Langchain_chat_bot.ipynb

---
[cell 1 code]
pip install -r C:\Users\DELL\Downloads\OPENAI-API-Tutorials-main\OPENAI-API-Tutorials-main\requirements.txt

[cell 2 code]
pip install langchain_openai

[cell 3 code]
pip install streamlit

[cell 4 code]
from langchain_openai import ChatOpenAI
from langchain_core.prompts import ChatPromptTemplate
from langchain_core.output_parsers import StrOutputParser

[cell 5 code]
import streamlit as st
import os
from dotenv import load_dotenv

[cell 6 code]
os.environ["OPENAI_API_KEY"]=os.getenv("OPENAI_API_KEY")
## Langmith tracking
os.environ["LANGCHAIN_TRACING_V2"]="true"
os.environ["LANGCHAIN_API_KEY"]=os.getenv("LANGCHAIN_API_KEY")

[cell 7 code]
prompt=ChatPromptTemplate.from_messages(
    [
        ("system","You are a helpful assistant. Please response to the user queries"),
        ("user","Question:{question}")
    ]
)

[cell 8 code]
## streamlit framework

st.title('Langchain Demo With OPENAI API')
input_text=st.text_input("Search the topic u want")

[cell 9 code]
# openAI LLm 
llm=ChatOpenAI(model="gpt-3.5-turbo")
output_parser=StrOutputParser()
chain=prompt|llm|output_parser

[cell 10 code]
if input_text:
    st.write(chain.invoke({'question':input_text}))