ML Platform - No-Code Machine Learning Platform

A comprehensive no-code machine learning platform that allows users to upload datasets, train various ML models, and deploy them with ease. Built with Flask backend and modern web technologies.

Access the project

Try Live App on https://luft-artificial-intelligence.vercel.app/

Web Interface on https://reaishma.github.io/LUFT-Artificial-Intelligence-/

🚀 Features

Core Functionality

• Easy Data Upload: Support for CSV, Excel (.xlsx), and JSON formats
• Automated Data Preprocessing: Handle missing values, normalization, and feature engineering
• Multiple ML Algorithms:
  • Regression: Linear Regression, Decision Tree, Random Forest, SVM, Neural Networks
  • Classification: Logistic Regression, Decision Tree, Random Forest, SVM, Neural Networks
  • Clustering: K-Means, Hierarchical Clustering
  • Dimensionality Reduction: Principal Component Analysis (PCA)

• Model Training: Background training with real-time progress tracking
• Model Evaluation: Comprehensive performance metrics and visualizations
• Model Deployment: One-click deployment to cloud platforms (AWS, GCP, Azure)

User Interface

• Landing Page: Platform introduction and benefits
• Model Builder: Step-by-step model creation wizard
• Model Gallery: Browse and manage trained models

• Data Uploader: Drag-and-drop data upload with analysis
• Model Trainer: Monitor training progress and logs
• Model Evaluator: Comprehensive performance analysis
• Model Deployer: Deploy models as REST APIs
• Documentation: Complete user guides and API documentation
• Support: Help center and troubleshooting

📁 Project Structure

ML-Platform/
├── Frontend Files
│   ├── index.html              # Landing page
│   ├── model-builder.html      # Model builder interface
│   ├── model-gallery.html      # Model gallery
│   ├── data-uploader.html      # Data upload interface
│   ├── model-trainer.html      # Training interface
│   ├── model-evaluator.html    # Evaluation interface
│   ├── model-deployer.html     # Deployment interface
│   ├── documentation.html      # Documentation
│   ├── support.html            # Support center
│   ├── style.css               # Styling
│   └── script.js               # Frontend logic
├── Backend Files
│   ├── app.py                  # Main Flask application
│   ├── models.py               # ML model management
│   ├── data.py                 # Data processing
│   ├── train.py                # Model training
│   ├── evaluate.py             # Model evaluation
│   └── deploy.py               # Model deployment
├── Configuration
│   ├── requirements.txt        # Python dependencies
│   └── README.md              # This file

🛠 Installation & Setup

Prerequisites

• Python 3.11 or higher
• Modern web browser
• 4GB+ RAM recommended
• Internet connection for cloud deployments

Quick Start

1. Clone the repository (or ensure all files are in your project directory)

2. Install dependencies:

   pip install -r requirements.txt

3. Start the Flask server:

   python app.py

4. Open your browser and navigate to:

   http://localhost:5000
   

Development Setup

For development with auto-reload:

export FLASK_ENV=development
export FLASK_DEBUG=1
python app.py

🎯 Quick Usage Guide

1. Upload Data

• Go to Data Uploader page
• Drag & drop or click to upload CSV, Excel, or JSON files
• View automatic data analysis and quality assessment
• Configure preprocessing options

2. Build Model

• Navigate to Model Builder
• Select your algorithm type (Regression, Classification, Clustering, etc.)
• Choose specific algorithm and configure parameters
• Select dataset and target column
• Create your model

3. Train Model

• Monitor training progress in Model Trainer
• View real-time logs and status updates
• Control training process (stop, pause, restart)
• Get notified when training completes

4. Evaluate Model

• Use Model Evaluator for comprehensive performance analysis
• View metrics like accuracy, F1-score, R² score
• Generate confusion matrices and ROC curves
• Compare multiple models

5. Deploy Model

• Go to Model Deployer
• Select trained model and deployment configuration
• Choose cloud provider (AWS, GCP, Azure) or deploy locally
• Get REST API endpoints for your model

📊 Supported Algorithms

Regression

• Linear Regression: Simple linear relationships
• Decision Tree Regressor: Non-linear patterns
• Random Forest Regressor: Ensemble method, robust
• Support Vector Regressor: High-dimensional data
• Neural Network Regressor: Complex patterns

Classification

• Logistic Regression: Linear classification
• Decision Tree Classifier: Interpretable rules
• Random Forest Classifier: High accuracy ensemble
• Support Vector Classifier: Margin-based classification
• Neural Network Classifier: Deep learning patterns

Clustering

• K-Means: Partition-based clustering
• Hierarchical Clustering: Tree-based clustering

Dimensionality Reduction

• PCA: Principal Component Analysis

🔧 API Reference

Health Check

GET /api/health

List Algorithms

GET /api/algorithms

Upload Dataset

POST /api/upload
Content-Type: multipart/form-data

file: [your-dataset-file]

Train Model

POST /api/train
Content-Type: application/json

{
  "model_name": "My Model",
  "algorithm": "random_forest_classifier",
  "dataset_id": "uuid-here",
  "target_column": "target",
  "parameters": {
    "n_estimators": 100,
    "max_depth": 10
  }
}

Check Training Status

GET /api/training-jobs/{job_id}

List Models

GET /api/models

List Datasets

GET /api/datasets

🚀 Deployment Options

Local Deployment

• Docker containers
• Direct Flask server
• Development environment

Cloud Deployment

• AWS: ECS, Lambda, EC2
• Google Cloud: Cloud Run, App Engine
• Azure: Container Instances, App Service

Features

• Auto-scaling capabilities
• Load balancing
• Health monitoring
• SSL/TLS encryption
• Cost estimation

📈 Model Performance Tracking

Metrics Supported

• Classification: Accuracy, Precision, Recall, F1-Score, ROC AUC
• Regression: R² Score, RMSE, MAE, MAPE
• Clustering: Silhouette Score, Inertia
• Feature Importance: For tree-based models

Visualizations

• Confusion matrices
• ROC curves
• Precision-recall curves
• Feature importance plots
• Performance comparison charts

🔒 Security Features

• Secure file upload with validation
• Input sanitization
• Error handling and logging
• No exposure of sensitive information
• Cache control for development

🛡 Data Privacy

• Local processing by default
• No data sent to external services without consent
• Secure cloud deployment options
• User control over data storage

📝 Configuration

Environment Variables

• FLASK_ENV: development/production
• FLASK_DEBUG: True/False for debug mode
• MAX_CONTENT_LENGTH: Maximum upload size
• Cloud provider credentials for deployment

Preprocessing Options

• Missing value handling strategies
• Feature scaling methods
• Categorical encoding approaches
• Feature selection techniques

🐛 Troubleshooting

Common Issues

Upload fails:

• Check file size (max 100MB)
• Verify file format (CSV, Excel, JSON)
• Ensure proper file structure

Training fails:

• Verify data quality
• Check target column exists
• Ensure sufficient data samples
• Review algorithm parameters

Deployment fails:

• Check cloud credentials
• Verify network connectivity
• Review deployment configuration
• Check resource limits

Performance Tips

• Clean data before upload
• Remove irrelevant features
• Start with simple algorithms
• Monitor resource usage
• Use appropriate instance sizes for deployment

📞 Support

Documentation

• Complete user guides in the Documentation section
• API reference with examples
• Video tutorials (coming soon)

Community

• GitHub Issues for bug reports
• Feature requests welcome
• Community forum discussions

Contact

• Email: [email protected]
• Emergency support: Available for critical issues

🔮 Roadmap

Upcoming Features

• Deep Learning: CNN and RNN support with TensorFlow/PyTorch
• Time Series: Specialized algorithms for temporal data
• Auto ML: Automated model selection and hyperparameter tuning
• Advanced Visualizations: Interactive charts and dashboards
• Collaborative Features: Team workspaces and model sharing
• Model Versioning: Track model iterations and comparisons
• Data Connectors: Direct database and API integrations
• Real-time Inference: Streaming predictions
• Model Monitoring: Production model performance tracking

Platform Improvements

• Enhanced mobile interface
• Offline capabilities
• Advanced security features
• Multi-language support
• Enterprise SSO integration

📄 License

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

🤝 Contributing

Contributions are welcome! Please read our contributing guidelines and submit pull requests for any improvements.

Development Guidelines

• Follow PEP 8 for Python code
• Use semantic commit messages
• Add tests for new features
• Update documentation
• Ensure cross-browser compatibility

🙏 Acknowledgments

• Built with Flask, scikit-learn, and modern web technologies
• Inspired by the need for accessible machine learning tools
• Thanks to the open-source ML community

---

Made with ❤️ for democratizing machine learning

Transform your data into insights without writing a single line of code!