https://github.com/darpa-4-aui/paper_data

Used to store necessary public data

This score indicates how likely this project is to be science-related based on various indicators:

○
CITATION.cff file
✓
codemeta.json file
Found codemeta.json file
○
.zenodo.json file
○
DOI references
✓
Academic publication links
Links to: ieee.org
○
Academic email domains
○
Institutional organization owner
○
JOSS paper metadata
○
Scientific vocabulary similarity
Low similarity (3.7%) to scientific vocabulary

Last synced: 10 months ago · JSON representation

Repository

Used to store necessary public data

Created over 3 years ago · Last pushed about 2 years ago

Readme

To help readers better understand DARPA and our paper, we provide necessary data mentioned at our paper in this project .

Our Yolov5 models is trained based on Yolov5

./dark_pattern_UI_data_set
- All training dataset is in coco farmat
- We added a few more data after the paper was published, so its number might be little different from our source paper DARPA
- Totally 3 types: cross, skip and close. Cross and skip are AGO in paper, and close are UPO.
./pytorch_model
- The Yolov5 model in Pytorch format
- We use yolov5s as the base model
./ncnn _model
- The Yolov5 model in ncnn format which is inferenced from ./pytorch_modelby ncnn