Mask Image Classification

Code for solution in Mask Image Classification hosted by Naver Boostcamp AI Tech.

To learn more detail about the competition, please, refer to the AI Stage post

Archive contents

minibatch28/ ├── data/ | ├── image/ | | ├── train/ │ | | ├── 00001_male_Asian_40/ │ | | | ├── mask.jpg │ | | | ├── mask2.jpg │ | | | ├── incorrect.png │ | | | └── normal.jpeg │ | | ├── {Number}_{Gender}_{Race}_{Age}/ │ | | | ├── ... │ | | | └── ... | | | └── 99999_female_Asian_150/ │ | | ├── mask.jpg │ | | ├── incorrect.jpg │ | | └── normal.jpg | | └── eval/ │ | ├── abcde.jpg │ | ├── {Any_image_name}.jpg │ | └── lorem_ipsum.jpeg │ ├── train.csv │ └── info.csv ├── output/ │ └── ensemble/ ├── models/ ├── dataset.py ├── loss.py ├── inference.py ├── train.py └── train.sh

• data/ : contains raw data dir and label data (should contain 'train.csv', 'info.csv')
• data/image/ : raw image dir of the competition
• data/eval/ : evaluation image dir of the competition
• output/ : inference result csv files will be created
• output/ensemble/ : ensemble result csv files will be created
• models/ : contains trained state_dict of each model

Requirements

• Ubuntu 18.04.5 LTS
• Python 3.8.5
• Pytorch 1.7.1
• CUDA 11.0

You can use the pip install -r requirements.txt to install the necessary packages.