zhenyuw16

OK, the README has been updated.

This repo is implemented with mmdetection v2.18, alreadly with swin transformer. You can add the corresponding config file for vision transformer experiments. We can consider add the code of swin...

For 1% to 10%, we use the prepare_coco_data.py to generate data. Besides this setting, we also use the 35k subset of the COCO 2014 validation set as labeled images. This...

As I know, this dividion is firstly proposed in "Data Distillation: Towards Omni-Supervised Learning".