search

fundamentalvision / Deformable-DETR

Deformable DETR: Deformable Transformers for End-to-End Object Detection.
Apache License 2.0
3.15k stars 513 forks source link

two_stage #182

Open liangjxiong opened 1 year ago

liangjxiong commented 1 year ago

The following error occurred when I used parameter --two_stage during training. What should I do?

enc_outputs_coord_unact = self.decoder.bbox_embed[self.decoder.num_layers](output_memory) + output_proposals

TypeError: 'NoneType' object is not subscriptable

How should I define self.decoder.class_embed and self.decoder.bbox_embed?

chenhbo commented 1 year ago

I meet the same issue. Did you solve it?

chenhbo commented 1 year ago

I found the problem. When we set the --two_stage as True, we also need to set --with_box_refine as True. Thx