Skip to content

A chatbot built using LangChain and Gemini (or any LLMs) that answers user queries from documents and includes a conversational form for collecting user information. Features date extraction and validation for user inputs.

Notifications You must be signed in to change notification settings

AnjaanKhadka/DocsChatBot

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

DocsChatBot2.mp4

DocsChatBot

A chatbot built using LangChain and Gemini that answers user queries from documents and includes a conversational form for collecting user information. Features date extraction and validation for user inputs.

Functionalities

Document QA part

It is a implementation of document retrival type chatbot that can answer based on information available in a python file. Currently I am using Attention is all you need paper. This file is kept in sample_docs folder as attention_is_all_you_need.pdf.

You can Change the content of file or use multiple files with simple tweek in the code.

Booking part (Conversation Form)

Now Booking ststem is also implemented. Booking session is initiated by simply asking to make a booking in Natural language.

Booking system can extract information from the natural conversation between user and ChatBot. This System can efficiently collect information in conversational Form.

Installation

Clone the repository in a folder.

git clone https://github.com/AnjaanKhadka/DocsChatBot.git

Get in the direcory

cd DocsChatBot

All dependencies are kept in requirements.txt file.

pip install -r requirements.txt

Beside requirements, you need to have Gemini API key. You can create a free Gemini API key for testing purpose.

After getting the key, Copy your API key and paste it in your API_KEY.txt file in the primary direcory.

echo Your_API_KEY > API_KEY.txt

Execution

I am using Streamlit to create interactive chatbot. Run the main.py with streamlit.

streamlit run main.py

Demo Run

Document QA System

DocsChatBot.mp4

Triggering the Booking system

DocsChatBot2.mp4

Collecting information

DocsChatBot3.mp4

About

A chatbot built using LangChain and Gemini (or any LLMs) that answers user queries from documents and includes a conversational form for collecting user information. Features date extraction and validation for user inputs.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages