Skip to content

Latest commit

 

History

History
79 lines (52 loc) · 2.67 KB

README.md

File metadata and controls

79 lines (52 loc) · 2.67 KB

Chatbot with Streamlit Interface

This project implements a chatbot capable of answering questions about me. The chatbot leverages a custom-trained model and is deployed using a Streamlit web application for an intuitive user interface. Check it out here:

Cypher

Features

  • Custom Intent Recognition: Understands user intents defined in intents.json.
  • Deep Learning Model: Uses a neural network for processing and responding to user inputs.
  • Streamlit Integration: Provides a user-friendly web interface for interaction.
  • Expandable: Easily customizable intents and responses for additional functionality.

Repository Structure

File/Folder Description
app.py The main file for running the Streamlit-based chatbot application.
chat.py Contains core logic for processing user queries and generating responses.
data.pth Saved PyTorch model file containing the trained chatbot model.
intents.json JSON file defining chatbot intents and corresponding responses.
model.py Script for building and training the neural network model.
nltk_utils.py Utility functions for text preprocessing, including tokenization and stemming.
requirements.txt List of required Python libraries for setting up the environment.

Installation

Follow the steps below to set up the project on your local machine:

Prerequisites

Ensure you have Python 3.8+ installed.

Steps

  1. Clone the repository:

    git clone https://github.com/VanshajR/ChatBot.git
    cd ChatBot
  2. Install dependencies: Use the provided requirements.txt to install all necessary packages:

    pip install -r requirements.txt
  3. Run the app: Start the chatbot interface with Streamlit:

    streamlit run app.py
  4. Interact with the bot: Open the Streamlit app URL in your browser and start chatting with the bot.


Usage

  • Modify the intents.json file to customize the chatbot's understanding of queries and responses.
  • Train the chatbot model using model.py to suit your data.

Deployment

The Streamlit app can be deployed using platforms like Streamlit Community Cloud, Heroku, or Docker.


License

This project is open-source and available under the MIT License.