how to integrate langchain conversations with react chatbot

Question

I have a chatbot UI built with react that invokes a Python script and displays the script response using fast API. I use gpt 3.5 to format the question into a query in my Python script. I then send the query to an internal API and return the API response to the chatbot. I now want to add conversation to the chatbot. How do I handle this on the react and python/LLM end?

I tried using Langchain's conversation buffer window. I am expecting similar results with the chatbot. I would like to answer any follow-up questions about the last 5 questions asked on the chatbot. How do I store the question and response and add it to the context next time the Python script is invoked?

j3ffyang · Accepted Answer · 2024-02-05 19:31:14Z

0

Here's an example of ConversationChain and ConversationBufferWindowMemory functions that achieves your expectation. Where if k is set to 5, the model will remember the last 5 interactions

from langchain.chains import ConversationChain
from langchain.memory import ConversationBufferWindowMemory
conversation = ConversationChain(
    llm=llm,
    prompt=PROMPT,
    # memory=ConversationBufferMemory(),
    memory=ConversationBufferWindowMemory(k=5),
    verbose=True)

answered Feb 5, 2024 at 19:31

j3ffyang

2,52827 silver badges16 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

saishree godbole Over a year ago

does the llm remember the previous questions if I invoke the script for every question?

j3ffyang Over a year ago

ConversationChain with ConversationBufferWindowMemory will keep the conversation in memory. As the example, k=5 means the last 5 interactions will be remembered.

Collectives™ on Stack Overflow

how to integrate langchain conversations with react chatbot

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related