Skip to content

CHANGELOG.md

Latest commit

 

History

History
210 lines (159 loc) · 5.14 KB

CHANGELOG.md

File metadata and controls

210 lines (159 loc) · 5.14 KB

Changelog

All notable changes to the Custom Image Classifier project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[Unreleased]

Planned

  • Transfer learning with pre-trained models (ResNet, VGG, etc.)
  • Model ensemble support
  • Advanced data augmentation options
  • Model interpretability with Grad-CAM
  • Experiment tracking with MLflow
  • Docker containerization
  • Mobile app for predictions

[1.0.0] - 2025-11-07

Added

  • Initial release of Custom Image Classifier
  • Web-based interface for model training and predictions
  • Command-line interface (CLI) for project management
  • RESTful API for integration
  • Support for custom image datasets
  • CNN model architecture with configurable hyperparameters
  • Project-based organization system
  • Multiple project support
  • Real-time training progress monitoring
  • GPU acceleration support (CUDA)
  • Data augmentation pipeline
  • Webcam integration for predictions
  • ZIP file upload support
  • Individual image upload support
  • Confidence score visualization
  • Training history tracking
  • Dataset statistics and class distribution
  • Model checkpointing
  • Comprehensive documentation (README, GETTING_STARTED, DATASET_GUIDE)

Features

Web Interface

  • Project creation and management
  • Dataset upload (ZIP and individual files)
  • Model training with configurable parameters
  • Real-time prediction interface
  • Webcam capture support
  • Project deletion

CLI Tool

  • list - List all projects
  • info - Show project details
  • create - Create new project
  • train - Train project model

API Endpoints

  • POST /api/create_project - Create a new project
  • POST /api/upload_dataset - Upload dataset
  • POST /api/train - Start model training
  • POST /api/predict - Make predictions
  • GET /api/projects - List all projects
  • GET /api/projects/<name> - Get project details
  • DELETE /api/delete_project/<name> - Delete project

Model Features

  • Custom CNN architecture
  • 3 convolutional layers with ReLU activation
  • MaxPooling for spatial reduction
  • Batch normalization
  • Dropout regularization (50%)
  • Fully connected layers
  • Softmax output for classification
  • Adam optimizer
  • Cross-entropy loss

Data Processing

  • Automatic image resizing (128×128)
  • Random horizontal flips
  • Random rotations (±10°)
  • Color jittering (brightness, contrast, saturation)
  • Normalization
  • Train/validation split

Configuration

  • Configurable epochs (default: 10)
  • Configurable batch size (default: 32)
  • Configurable learning rate (default: 0.001)
  • Configurable image size (default: 128×128)
  • Configurable dropout rate (default: 0.5)
  • Optional MongoDB integration for history tracking

Documentation

  • Comprehensive README with badges and examples
  • Getting Started guide with step-by-step instructions
  • Dataset preparation guide with examples
  • API documentation with curl examples
  • Contributing guidelines
  • MIT License
  • Code of Conduct

Dependencies

  • Flask 3.0.0 - Web framework
  • PyTorch 2.1.0 - Deep learning
  • torchvision 0.16.0 - Computer vision utilities
  • OpenCV 4.8.1.78 - Image processing
  • Pillow 10.1.0 - Image handling
  • NumPy 1.26.2 - Numerical computing
  • Werkzeug 3.0.1 - WSGI utilities
  • requests 2.31.0 - HTTP library
  • pymongo 4.6.0 - MongoDB driver (optional)

[0.9.0] - 2025-10-15

Added

  • Beta release for testing
  • Core functionality implementation
  • Basic web interface
  • Model training pipeline
  • Prediction interface

Known Issues

  • No CLI support
  • Limited documentation
  • No API documentation
  • No error handling for edge cases

[0.5.0] - 2025-09-01

Added

  • Alpha release
  • Proof of concept
  • Basic CNN model
  • Flask application skeleton
  • Simple file upload

Known Issues

  • No project management
  • Single model only
  • No web interface
  • Command-line only

Version History Summary

Version Release Date Status Highlights
1.0.0 2025-11-07 Stable Full release with web UI, CLI, API
0.9.0 2025-10-15 Beta Testing phase
0.5.0 2025-09-01 Alpha Initial proof of concept

Migration Guides

Upgrading from 0.9.0 to 1.0.0

No breaking changes. All 0.9.0 projects are compatible with 1.0.0.

New features available:

  • CLI tool (cli.py)
  • API endpoints
  • Enhanced documentation
  • Better error handling

Contributing

See CONTRIBUTING.md for information on how to contribute to this project.

Links

Legend:

  • Added - New features
  • Changed - Changes in existing functionality
  • Deprecated - Soon-to-be removed features
  • Removed - Removed features
  • Fixed - Bug fixes
  • Security - Vulnerability fixes

This changelog is maintained by the Custom Image Classifier team.