docs: Update architecture documentation and add auto-match load analysis

This commit updates the `ARCHITECTURE.md` file to reflect the transition to a web-based application, including new features and system overview. Additionally, it introduces `AUTOMATCH_LOAD_ANALYSIS.md`, detailing performance issues with the Auto-Match page and recommendations for optimizations. A new document, `CONFIDENCE_CALIBRATION_SUMMARY.md`, is also added to explain the implementation of a confidence calibration system for face recognition, ensuring more accurate match probabilities. These updates enhance the project's documentation and provide insights for future improvements.
This commit is contained in:
Tanya
2026-01-06 13:11:30 -05:00
parent 3ec0da1573
commit 845b3f3b87
25 changed files with 7544 additions and 74 deletions
+129
View File
@@ -0,0 +1,129 @@
# PunimTag - Quick Start Guide
## 🚀 Running the Application
### Start Dashboard
```bash
source venv/bin/activate
python run_dashboard.py
```
### Run CLI Tool
```bash
source venv/bin/activate
python -m src.photo_tagger --help
```
---
## 📁 Project Structure
```
punimtag/
├── src/
│ ├── core/ # Business logic
│ ├── gui/ # GUI components
│ └── utils/ # Utilities
├── tests/ # Test suite
├── docs/ # Documentation
├── .notes/ # Project planning
└── run_dashboard.py # Main launcher
```
---
## 📚 Key Documentation
- **README.md** - Main documentation
- **CONTRIBUTING.md** - How to contribute
- **docs/ARCHITECTURE.md** - System design
- **RESTRUCTURE_SUMMARY.md** - Restructure details
- **IMPORT_FIX_SUMMARY.md** - Import fixes
---
## 🔧 Common Tasks
### Add Photos
1. Open dashboard: `python run_dashboard.py`
2. Click "Scan Photos" in menu
3. Select folder with photos
### Process Faces
1. Open dashboard
2. Click "Process Photos" button
3. Wait for face detection to complete
### Identify People
1. Open "Identify" tab
2. View unidentified faces
3. Enter person name or select existing
4. Click "Identify"
### Search Photos
1. Open "Search" tab
2. Enter search criteria (name, date, tags)
3. View results
---
## 🐛 Troubleshooting
### ModuleNotFoundError
**Solution**: Use `run_dashboard.py` launcher, not direct file execution
### Import Errors
**Solution**: Make sure you're in the venv:
```bash
source venv/bin/activate
```
### PIL/ImageTk Error
**Solution**: Install Pillow in venv:
```bash
pip install Pillow
```
---
## 💡 Tips
- Always activate venv before running
- Use `run_dashboard.py` for GUI
- Use `python -m src.photo_tagger` for CLI
- Check `.notes/` for planning docs
- Read `docs/ARCHITECTURE.md` for system design
---
## 📞 Need Help?
1. Check documentation in `docs/`
2. Read `.notes/` for planning info
3. See `CONTRIBUTING.md` for guidelines
---
**Quick Command Reference:**
```bash
# Activate environment
source venv/bin/activate
# Run dashboard
python run_dashboard.py
# Run CLI
python -m src.photo_tagger
# Run tests
python -m pytest tests/
# Deactivate environment
deactivate
```
---
**Last Updated**: October 15, 2025