TrashSort

1. Install

python -m venv venv
. venv/bin/activate
pip install -r requirements.txt

2. Train model

python training_pipeline.py

Output model (default):

<output-dir>/runs/train_py/weights/best.pt

3. Run inference demo

Set API key (or create .env in this folder):

export GROQ_API_KEY=your_key_here

Run:

python execution_pipeline.py

You can resize the inference window while it is running. To start larger:

python execution_pipeline.py --window-width 1280 --window-height 720

Optional custom weights:

python execution_pipeline.py --weights /path/to/best.pt