Skip to content

This is AI ChatBot. I make it By AI(NLP). I know sometime it's show wrong answer because i use less data to train my model because i haven't big dataset. you can read my code and you can learn many thing. This chatbot is not bad at all. You can have fun with it in your spare time if you want.

License

Notifications You must be signed in to change notification settings

shukur-alom/AI-ChatBot

Repository files navigation

AI-ChatBot

This project implements a simple chatbot using Keras with a pre-trained model. The chatbot is trained on a dataset of intents and responses. The model predicts the intent of user input and generates an appropriate response.

Overview

This chatbot script uses a neural network model (chatbot_model.h5), a tokenizer (Tokenizer.p), and a label encoder (LabelEncoder.p) to understand user input and generate responses. The model is trained on a dataset of intents and responses provided in the intents.json file.

Prerequisites

  • Python 3.x
  • Required Python packages: Keras, pandas, numpy

Installation

  1. Clone the repository:

    https://github.com/shukur-alom/AI-ChatBot.git
    
  2. Install the required dependencies:

pip install -r requirements.txt
  1. Run the chatbot script:
python main.py
  1. Enter your messages when prompted, and the chatbot will generate responses based on the trained model.

Model Details

  • The model is a neural network loaded from chatbot_model.h5.
  • Tokenization is performed using the tokenizer saved in Tokenizer.p.
  • Intent labels are encoded and decoded using the label encoder saved in LabelEncoder.p.
  • The intents and responses are defined in the intents.json file.

About

This is AI ChatBot. I make it By AI(NLP). I know sometime it's show wrong answer because i use less data to train my model because i haven't big dataset. you can read my code and you can learn many thing. This chatbot is not bad at all. You can have fun with it in your spare time if you want.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published