Mask detection model using Pytorch
Compose_data.py used data from
Extract both file in 'data' folder
- with_mask
- without_mask
- yolo
Then, execute compose_data.py
to create json file contain bounding box (left x, top y, width, height) and label
Initialize dataloader using MaskDataLoader
Model, Loss function, and other utility functions are modified from source: https://github.com/sgrvinod/a-PyTorch-Tutorial-to-Object-Detection