JingyuLi comments

Repositories
Issues
Comments

Results 15 comments of


                                            JingyuLi

Swin Transformer pre-trained Model?

Hi， thanks for your reply! I want to know the result of using the backbone pre-trained on Regular ImageNet-1K Swin-B.

About preprocess code

Thanks for your great work! In your paper Table 3, Sub-GC-oracle is an upper bound by assuming that we have an oracle ranking function, i.e., always selecting the maximum scored...

About preprocess code

Yes, The upper bound is promising!!! Can you share the code of sub-graph sampling or sub-graph extraction now? Thanks a lot!

about code release time

The same question

训练验证时间太长？

嗯，对的 batchsize设置16,32好像不行，那是什么原因导致训练时验证时间那么长呢？十几分钟？我只有1500张验证图片，在YOLOv3验证时只需要一分钟不到。