Notely - AI-Powered Note Taking Application

A modern Django-based web application for creating, organizing, and managing notes with AI-powered automatic summarization using the T5 transformer model.

✨ Features

Core Functionality

• Create & Edit Notes: Rich note creation with titles, content, and AI summarization
• AI Summarization: Automatically generates summaries using T5-small transformer model
• Smart Fallback: Works even on low-memory environments with simple summarization

🎨 Organization & Visual

• Color Coding: 7 color options for visual organization
• Categories: Organize notes with custom categories
• Pinned Notes: Pin important notes to the top
• Favorites: Mark and filter favorite notes
• Dark Mode: Toggle between light and dark themes

🔍 Discovery

• Full-Text Search: Search across titles, content, and categories
• Multi-Filter: Filter by category, color, or favorites
• Pagination: Efficient browsing with 12 notes per page

💡 User Experience

• Modern UI: Card-based layout with smooth animations
• Responsive Design: Works perfectly on desktop, tablet, and mobile
• Keyboard Navigation: Fully keyboard accessible
• Time Tracking: Shows when notes were last updated

For detailed feature documentation, see FEATURES.md

Technology Stack

Core Framework

• Django 5.2.4 - Web framework
• asgiref 3.9.1 - ASGI server reference implementation
• sqlparse 0.5.3 - SQL parsing for Django

AI/ML Stack

• transformers 4.53.2 - Hugging Face transformers library
• torch 2.2.2 - PyTorch deep learning framework
• torchaudio 2.2.2 - Audio processing for PyTorch
• torchvision 0.17.2 - Computer vision for PyTorch
• tokenizers 0.21.2 - Fast tokenizers for transformers
• safetensors 0.5.3 - Safe tensor serialization
• huggingface-hub 0.33.4 - Hugging Face model hub integration
• hf-xet 1.1.5 - Hugging Face XET support
• T5-small model - Lightweight transformer for text summarization

Scientific Computing

• numpy 1.26.4 - Numerical computing
• sympy 1.14.0 - Symbolic mathematics
• mpmath 1.3.0 - Arbitrary-precision arithmetic

HTTP & Networking

• requests 2.32.4 - HTTP library
• urllib3 2.5.0 - HTTP client
• certifi 2025.7.9 - SSL certificates
• charset-normalizer 3.4.2 - Character encoding detection
• idna 3.10 - Internationalized domain names

Utilities

• regex 2024.11.6 - Regular expressions
• tqdm 4.67.1 - Progress bars
• filelock 3.18.0 - File locking
• fsspec 2025.5.1 - Filesystem spec
• networkx 3.4.2 - Network/graph algorithms
• packaging 25.0 - Package version handling
• typing_extensions 4.14.1 - Typing backports
• Jinja2 3.1.6 - Template engine
• MarkupSafe 3.0.2 - Safe string handling
• PyYAML 6.0.2 - YAML parser
• pillow 11.3.0 - Image processing

Database

• SQLite (default Django database)

Project Structure

notely/
├── manage.py                 # Django management script
├── notely/                   # Main project directory
│   ├── __init__.py
│   ├── settings.py          # Django settings
│   ├── urls.py              # Main URL configuration
│   ├── asgi.py              # ASGI configuration
│   └── wsgi.py              # WSGI configuration
└── notes/                    # Notes application
    ├── __init__.py
    ├── admin.py             # Django admin configuration
    ├── apps.py              # App configuration
    ├── models.py            # Note model definition
    ├── views.py             # View functions
    ├── forms.py             # NoteForm for creating/editing notes
    ├── urls.py              # URL routing for notes app
    ├── utils.py             # AI summarization utility
    ├── templates/           # HTML templates
    └── migrations/          # Database migrations

Installation

Prerequisites

• Python 3.8 or higher (tested with Python 3.10-3.13)
• pip (Python package installer)
• Approximately 500MB of free disk space (for ML models and dependencies)

Setup Steps

1. Clone the repository

   git clone git@github.com:Kelvyn2012/Note_AI.git
   cd notely

2. Create a virtual environment (recommended)

   python -m venv venv
   
   # On macOS/Linux:
   source venv/bin/activate
   
   # On Windows:
   venv\Scripts\activate

3. Install dependencies

   pip install -r requirements.txt

   This will install all 30 required packages including:

   • Django and its dependencies
   • PyTorch and AI/ML libraries (transformers, tokenizers, etc.)
   • Scientific computing packages (numpy, sympy)
   • HTTP libraries and utilities

   Note: Installation may take several minutes due to the size of PyTorch and AI libraries.

4. Run database migrations

   python manage.py migrate

5. Create a superuser (optional, for admin access)

   python manage.py createsuperuser

6. Start the development server

   python manage.py runserver

7. Access the application

   • Open your browser and navigate to http://127.0.0.1:8000/
   • Admin panel: http://127.0.0.1:8000/admin/

Usage

Creating a Note

1. Navigate to the home page
2. Click on "Add Note" or navigate to /add-note/
3. Enter a title and content
4. Click "Save" - the AI will automatically generate a summary

Editing a Note

1. From the home page, click "Edit" on the desired note
2. Modify the title or content
3. Click "Save" - a new summary will be generated

Deleting a Note

1. From the home page, click "Delete" on the desired note
2. The note will be removed from the system

AI Summarization

The application uses the T5-small model from Hugging Face for text summarization:

• Model: t5-small (lightweight transformer model)
• Minimum Text Length: 30 words (shorter texts are returned as-is)
• Input Limit: First 512 characters of the note content
• Summary Length: 20-60 words
• First Load: The model downloads on first use (approximately 200MB)
• Fallback Mode: If the model cannot load (e.g., memory constraints on free hosting), the app automatically falls back to simple text truncation

Memory Requirements

• Local Development: AI summarization works with sufficient RAM (2GB+)
• Render Free Tier: May fall back to simple summarization due to 512MB RAM limit
• Render Paid Plans: Full AI summarization available with 1GB+ RAM plans

Database Schema

Note Model

• id: Primary key (auto-generated)
• title: CharField (max 255 characters)
• content: TextField (unlimited length)
• summary: TextField (AI-generated, nullable)
• created_at: DateTimeField (auto-generated timestamp)

Configuration

Settings

Key configuration in notely/settings.py:

• DEBUG = True: Set to False in production
• SECRET_KEY: Change this in production
• ALLOWED_HOSTS: Configure for deployment
• DATABASES: Uses SQLite by default (can be changed to PostgreSQL, MySQL, etc.)

Development

Running Tests

python manage.py test

Creating Migrations

python manage.py makemigrations
python manage.py migrate

Accessing Django Shell

python manage.py shell

Security Considerations

• The current SECRET_KEY in notely/settings.py is for development only
• DEBUG mode is enabled - disable in production
• Configure ALLOWED_HOSTS before deploying
• Consider using environment variables for sensitive settings

Performance Notes

• The T5 model loads into memory on first summarization request
• Model remains cached in memory for subsequent requests
• Initial summarization may take a few seconds; subsequent requests are faster
• Text is truncated to 512 characters before summarization to ensure model compatibility

Troubleshooting

Model Download Issues

If the T5 model fails to download:

• Check your internet connection
• Ensure you have sufficient disk space (approximately 200MB)
• The model will download to the Hugging Face cache directory

Long Load Times

• First summarization request downloads the model (one-time operation)
• Server restart requires model reload
• Consider pre-loading the model in production environments

Future Enhancements

Potential features for future development:

• User authentication and authorization
• Note categories/tags
• Search functionality
• Export notes to PDF/Markdown
• Rich text editor
• Note sharing capabilities
• Multiple AI models for summarization

License

This project is provided as-is for educational and development purposes.

Contributing

Contributions are welcome. Please ensure all tests pass before submitting pull requests.

Support

For issues, questions, or contributions, please open an issue in the repository.