aggefa

age, gender & face detector

Install

Dependencies:

• Python 3.10.12
• CUDA 12.6

Python environment:

git clone https://github.com/TigDang/aggefa

cd aggefa

pip install -r requirements.txt

Data:

In this work I use UTKFace dataset from Kaggle. UTKFace folder is used.

Training

For training simply run:

python train.py

Feel free to edit hydra config inside config folder.

Tensorboard server run:

tensorboard --logdir tb_logs/

Test

python test.py ckpt_path=...

Inference

For that case you must get YOLO weights from HF:

wget https://huggingface.co/jaredthejelly/yolov8s-face-detection/resolve/main/YOLOv8-face-detection.pt -P checkpoints/

Thanks jaredthejelly for that model weights!

And aggefa weights:

wget https://huggingface.co/tigdang/aggefa/resolve/main/aggefa.ckpt -P checkpoints/

For json-inference service:

python inference.py

And then take a look at the http://127.0.0.1:8000/docs

Docker

• Docker version 27.1.1
• Docker-compose version 1.29.2

For build container for inference:

docker-compose up --build inference