How to Create a Smart Chatbot with Streamlit, Python, and ChatGPT

Create your own intelligent memory-enabled chatbot in 80 lines of code

AI technologies have progressed to a point where humans are now engaging in chat-like interactions with chatbots, creating an experience that closely resembles human conversation.

At the core of this development lies Large Language Models (LLMs), and one such model called GPT (on which ChatGPT is based) has gained considerable popularity in recent months.

There have been chatbots present over the Internet for quite some time. But they have an inherent limitation of not remembering the context of the conversation. This happens because they are stateless (or lack memory).

Let’s look at an example of a stateless chatbot.

A Typical Example of a Stateless Chatbot | Image by Author

Today, we will look at the other side of the coin. We will create a chatbot that remembers the context and past conversations using ChatGPT’s model GPT-3.5.

Flow for The Chatbot

The workflow of the chatbot | Image by Author

Packages/Libraries

1. Streamlit: To create a web interface that will run our chatbot. Install it using pip install streamlit
2. OpenAI: They are a research and deployment company that created the GPT-3, GPT-3.5, and GPT-4.0 models. Install it using pip install openai
3. LangChain: It is a Python package that offers a consistent interface for memory and a range of memory implementations for chatbots. Install it using pip install langchain

import streamlit as st
from langchain.chains import ConversationChain
from langchain.chains.conversation.memory import ConversationEntityMemory
from langchain.chains.conversation.prompt import ENTITY_MEMORY_CONVERSATION_TEMPLATE
from langchain.chat_models import ChatOpenAI