Learning ML/Python Chatbot 🤖

This is a learning project designed to understand the basics of machine learning and natural language processing using Python. It implements a simple chatbot that can be trained to recognise intents and carry out basic conversations.

🧠 What It Does

The project consists of two main components:

1. Training Module (train.py)

• Processes training data from intents.json
• Uses a neural network to learn patterns
• Creates a model that can recognise user intents
• Implements Bag of Words (BoW) approach for text processing

2. Chatbot Module (chatbot.py)

• Uses the trained model to understand user input
• Manages conversation flow
• Extracts entities (like amounts, names)
• Provides contextual responses

The chatbot can handle:

• Basic conversations
• Money transfers
• Balance enquiries
• Loan enquiries
• Multi-step conversations
• Context maintenance

📋 Requirements

• Python 3.8+
• TensorFlow (macOS optimized)
• NLTK (Natural Language Toolkit)
• NumPy

🛠 Environment Setup (macOS)

1. Create and Activate Conda Environment

# Create new environment
conda create -n tensorflow python=3.8

# Activate environment
conda activate tensorflow

2. Install TensorFlow Dependencies

# Install base dependencies
conda install tensorflow-deps

# Install macOS optimized TensorFlow
pip install tensorflow-macos

# Install Metal acceleration (Apple Silicon)
pip install tensorflow-metal

3. Install Required Packages

# Install NLTK
pip install nltk

# Install NumPy
pip install numpy

🎯 Training the Model

1. Prepare Training Data

• Review intents.json
• Add/modify intents as needed
• Format:

  jsonCopy{
    "intents": [
      {
        "tag": "greeting",
        "patterns": ["hi", "hello"],
        "responses": ["Hello!", "Hi there!"]
      }
    ]
  }

2. Run Training

python train.py

This will:

• Process the intents
• Train the neural network
• Save the model as 'chatbot_model.h5'
• Save vocabulary as 'words.pkl'
• Save classes as 'classes.pkl'

🤖 Running the Chatbot

1. Start the Chatbot

python chatbot.py

2. Interact with the Bot

• Type your messages
• Bot will respond based on trained intents
• Type "quit" to exit