This repository allows you to have a conversation with your own data!

• Support for Multiple File Types: The system can process both .pdf and .txt files, allowing a wide range of documents to be used.
• Real-time Responses: Using OpenAI's GPT API, the chatbot provides quick and accurate responses based on the content of your documents.
• Interactive UI: A user-friendly interface built using Gradio's Blocks allows for a seamless chatting experience.

1. create_index.py:

   • This script is responsible for creating a Faiss index from the text present in the files stored in the uploads directory.
   • It can process both .pdf and .txt files.
   • All Faiss indices are saved in a indices.pkl file.
   • Important: This script should be run only once or whenever there are new documents you want the chatbot to reference.

2. bot.py:

   • This script initializes an interactive chatbot interface.
   • The chatbot uses the Faiss indices to find documents similar to the user's input and generate responses.
   • The chatbot uses OpenAI's GPT-3.5 model for generating responses.
   • An interactive UI is built using Gradio's Blocks, which allows users to interact with the chatbot in real-time.

1. Prerequisites:

   • Make sure you have Python 3.x installed.
   • Set the OpenAI API key as an environment variable: export OPENAI_API_KEY=your_api_key_here.

2. Installation:

   • Clone this repository: git clone https://github.com/your_repository_link.git
   • Navigate to the directory: cd path_to_directory
   • Install the required packages: pip install -r requirements.txt (Note: You need to have a requirements.txt file with all necessary libraries listed.)

3. Usage:

   • Place the documents you want the chatbot to reference in the uploads directory.
   • Run the create_index.py script to create the Faiss indices: python create_index.py
   • Launch the chatbot interface: python bot.py
   • Interact with the chatbot using the provided UI.

Please direct any questions or comments to Andre.

This project is licensed under the MIT License.