Repository containing files and source code for the CS4445 AI module's final capstone project.

• Python 3.8 or higher
• pip package manager

pip install -r requirements.txt

1. Clone the repository:

   git clone https://github.com/your-username/ai-project-team.git
   cd ai-project-team

2. Install dependencies as mentioned above

3. If you are planning to train a model run synthesise_data.py

4. Select the options:

• --model: Path to a custom trained model
• --image: Path to the image you would like to run the defined model on
• --labels: Path to defined labels json file
• --heatmap: Define whether you want the heatmap of focus points or not

Example:

python model/client.py --model model/saved_model --image G.png --heatmap 
python model/client.py --model model/saved_model --image test.png --train
python model/client.py --model model/saved_model --image test.png --heatmap    
python model/client.py --model model/saved_model --image test.png --train --heatmap   
python model/client.py --model model/saved_model --image test.png --labels model/label_to_letter.json #only used if labels cannot be found 

The project includes an interactive drawing interface that allows you to draw Ogham characters and get real-time predictions using the hybrid model:

python model/draw_interface.py

Features:

• Draw Ogham characters using your mouse
• Click "Predict" to get model predictions
• Shows confidence scores and top 3 predictions
• "Clear" button to reset the canvas
• Automatically resizes drawings to match model input size

Note: Ensure you have trained the model first using the commands above, as the interface requires model/saved_model and model/label_to_letter.json to be present.