[BUG] when training with pure background images, loss tend to be inf.

MCG-NJU / AdaMixer

[CVPR 2022 Oral] AdaMixer: A Fast-Converging Query-Based Object Detector

MIT License

236 stars 24 forks source link

[BUG] when training with pure background images, loss tend to be inf. #13

Closed wuterry closed 2 years ago

wuterry commented 2 years ago

In file _mmdet/models/roi_heads/bbox_heads/adamixer_decoderstage.py line 353，avg_factor = reduce_mean(num_pos) should be modified as 'avg_factor = max(reduce_mean(num_pos), 1)' to support training pure background images.