Co4AI

Results 7 comments of Co4AI

Questions about implementation details

It seems that self.decoder_track is trained under torch.randn(1).item() > 0.0 condition, self.decoder is trained otherwise. There are two questions: 1. when we train self.decoder, the two part feature maps are...

Questions about implementation details

> > It seems that self.decoder_track is trained under torch.randn(1).item() > 0.0 condition, self.decoder is trained otherwise. There are two questions: > > > > 1. when we train self.decoder,...

Eval_coco error: Problem in running contour metrics

good job. I meet the same error

[Reimplementation] YOLOX bbox pred using t,l,r,b format always produce loss_bbox: 5.0

close or open 'with_stride' in get priors function may help you

[Reimplementation] YOLOX bbox pred using t,l,r,b format always produce loss_bbox: 5.0

Do you miss exp(box_pred) after conv? This is done in FCOS' forward but YOLOX' post processing.

[Reimplementation] YOLOX bbox pred using t,l,r,b format always produce loss_bbox: 5.0

FCOS use different loss for different box-encoding settings.

The link to the dataset is not accessible

@yi-ming-qian Thanks for your wonderful work! I am trying to run this code, but the dataset/models links are broken. Could you share these? My email is [email protected]