Skip to content

About

A modern, interactive Single Page Application (SPA) for learning and practicing English verb tenses, built with Alpine.js, Bootstrap 5, and AI-powered predictions.

Topics

Resources

Stars

1 star

Watchers

1 watching

Forks

Latest commit

Β 

History

68 Commits

Folders and files

NameName
Last commit message
Last commit date
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

Verb Tenses Table (VTT)

A modern, interactive Single Page Application (SPA) for learning and practicing English verb tenses, built with Alpine.js, Bootstrap 5, and AI-powered predictions.

VTT Screenshot

🌟 Features

Interactive Verb Conjugation

  • Dynamic Tense Table: View all English verb tenses (Simple, Continuous, Perfect, Perfect Continuous) for any verb
  • Real-time Updates: Instantly see verb forms as you type
  • Active/Passive Voice: Examples in both active and passive voice
  • Smart Forms: Automatic detection of regular/irregular verbs

AI-Powered Features

  • Brain.js Integration: Machine learning predictions for verb classification
  • Confidence Scoring: AI indicates prediction reliability
  • Model Training: Train the neural network with custom data

Data Visualization

  • D3.js Charts: Visual representation of tense usage frequency
  • Interactive Graphs: Dynamic updating based on selected verb

Modern UI/UX

  • Responsive Design: Mobile-first approach with Bootstrap 5
  • Dark/Light Mode: Toggle between themes
  • Compact Layout: Optimized for small screens (sm breakpoint)
  • Font Awesome Icons: Visual enhancements throughout
  • Google Fonts: Clean, modern typography with Inter font

Comprehensive Verb Database

  • Irregular Verbs: Complete list of irregular English verbs
  • Quick Selection: One-click access to common verbs
  • Verb Statistics: Detailed breakdown of verb forms

πŸ› οΈ Technologies Used

  • Alpine.js - Lightweight reactive framework
  • Bootstrap 5 - Responsive CSS framework
  • Brain.js - Neural networks for AI predictions
  • D3.js - Data visualization library
  • Font Awesome 6 - Icon toolkit
  • Google Fonts - Web font service
  • CDN Delivery - All dependencies via Content Delivery Networks

πŸš€ Quick Start

Option 1: Direct File Open

Simply open index.html in any modern web browser. No installation required!

Option 2: Live Demo

Visit the live demo (if deployed)

Option 3: Local Development

  1. Clone the repository:
git clone https://github.com/yourusername/verb-tenses-table.git
cd verb-tenses-table
  1. Open index.html in your browser or use a local server:
# Using Python
python -m http.server 8000

# Using Node.js
npx serve .
  1. Navigate to http://localhost:8000 in your browser

πŸ“– How to Use

Basic Usage

  1. Enter a Verb: Type any English verb in the input field (e.g., "walk", "see", "go")
  2. View Conjugations: The table automatically updates with all tense forms
  3. Explore Examples: See affirmative, interrogative, and negative examples for each tense

Advanced Features

  • AI Prediction: Click the robot icon for AI classification of verb type
  • Train AI Model: Improve prediction accuracy by training the neural network
  • Visualize Data: Check the Usage Frequency chart for graphical insights
  • Toggle Theme: Switch between dark and light modes using the moon/sun icon
  • Quick Verbs: Use the quick-select buttons for common verbs

Verb Examples

  • Regular verbs: walk, talk, play, help
  • Irregular verbs: go, see, take, make, be, have, do
  • Mixed forms: dream, learn, burn (both regular and irregular forms)

πŸ“± Responsive Design

The application is fully responsive and optimized for:

  • Mobile: Compact tables with sm-optimized Bootstrap classes
  • Tablet: Two-column layout for medium screens
  • Desktop: Full-featured three-column interface
  • Print: Clean formatting for printing verb tables

πŸ€– AI Implementation

Neural Network

  • Architecture: 2 hidden layers (6-4 neurons)
  • Training: 2000 iterations on labeled verb data
  • Features: Character encoding, word length, ending patterns
  • Accuracy: Typically achieves 85-95% confidence on trained verbs

Training Data

The model is pre-trained on:

  • 8 regular verb patterns
  • 8 irregular verb patterns
  • Extensible with custom training

πŸ“Š D3.js Visualization

Features

  • Bar Charts: Visual representation of tense usage frequency
  • Dynamic Updates: Charts refresh with verb changes
  • Theme Support: Adapts to dark/light mode
  • Responsive: Scales with screen size

🎨 UI Components

Main Components

  1. Verb Input Card: Central control panel with AI features
  2. Tense Table: Comprehensive verb conjugation matrix
  3. Irregular Verbs List: Scrollable database of irregular verbs
  4. AI Prediction Panel: Machine learning insights and controls
  5. Usage Chart: Data visualization of tense frequencies

Design Features

  • Card-based Layout: Clean, organized information presentation
  • Color Coding: Visual distinction between tense types
  • Interactive Elements: Hover effects, transitions, and feedback
  • Accessibility: Semantic HTML and ARIA attributes

πŸ“ Project Structure

verb-tenses-table/
β”œβ”€β”€ index.html          # Main application file (contains all code)
β”œβ”€β”€ README.md           # This documentation file
└── (Optional assets)
    β”œβ”€β”€ screenshots/
    └── examples/

πŸ”§ Customization

Modifying Verb Database

Edit the irregularVerbs object in the script section to add/remove verbs:

irregularVerbs: {
    "yourverb": { 
        baseForm: "yourverb", 
        pastSimple: "pastform", 
        pastParticiple: "participleform" 
    },
    // ... existing verbs
}

Styling Customization

Modify the CSS variables in the <style> section:

:root {
    --primary-color: #your-color;
    --secondary-color: #your-color;
    --accent-color: #your-color;
}

Adding New Features

The modular Alpine.js structure makes it easy to extend:

  • Add new verb analysis features
  • Implement additional visualization types
  • Integrate with external APIs (dictionary, pronunciation, etc.)

🌐 Browser Compatibility

Browser Version Support
Chrome 60+ βœ… Full
Firefox 55+ βœ… Full
Safari 12+ βœ… Full
Edge 79+ βœ… Full
Opera 50+ βœ… Full

Note: Requires JavaScript and modern ES6+ support.

πŸ“ˆ Performance

  • Single File: All code in one HTML file for easy distribution
  • CDN Dependencies: Fast loading of external libraries
  • Lazy Loading: AI model trains on-demand
  • Optimized Assets: Minimal external requests

🀝 Contributing

  1. Fork the repository
  2. Create a feature branch: git checkout -b feature-name
  3. Make your changes
  4. Test thoroughly
  5. Commit your changes: git commit -m 'Add feature'
  6. Push to the branch: git push origin feature-name
  7. Submit a pull request

Areas for Contribution

  • Add more irregular verbs to the database
  • Improve AI training data and accuracy
  • Enhance visualization with more chart types
  • Add pronunciation guides (audio)
  • Implement spaced repetition for learning
  • Add multi-language support

πŸ› Troubleshooting

Common Issues

  1. AI not working

    • Ensure Brain.js loaded correctly from CDN
    • Check browser console for errors
    • Click "Train AI Model" button to initialize
  2. Chart not displaying

    • Verify D3.js loaded from CDN
    • Check if browser blocks external scripts
    • Try refreshing the page
  3. Verb forms incorrect

    • Ensure verb is in base form (infinitive)
    • Check irregular verbs database
    • Verify spelling
  4. Layout issues on mobile

    • Clear browser cache
    • Ensure viewport meta tag is present
    • Check for conflicting browser extensions

Browser Console Commands

For advanced users, the application exposes:

  • window.verbApp() - Alpine.js component instance
  • Access to D3.js and Brain.js objects

πŸ“š Learning Resources

English Verb Tenses

Technology Documentation

πŸ“„ License

This project is licensed under the MIT License - see the LICENSE file for details.

πŸ™ Acknowledgements

  • AngularJS Team: Original inspiration from Angular-based verb tenses table
  • Alpine.js Creators: For the lightweight reactive framework
  • Bootstrap Team: For the responsive CSS framework
  • Brain.js Contributors: For the browser-based neural network library
  • D3.js Community: For the powerful data visualization library

πŸ“ž Support

For support, questions, or feedback:


Made with ❀️ for English learners and developers

This tool is designed for educational purposes. While we strive for accuracy, always verify verb forms with authoritative sources.

About

A modern, interactive Single Page Application (SPA) for learning and practicing English verb tenses, built with Alpine.js, Bootstrap 5, and AI-powered predictions.

Topics

Resources

Stars

1 star

Watchers

1 watching

Forks

Releases

Packages

Contributors

Languages