Glorainow

[Docs] Request to supply some comment or documentation

4

### Branch master branch https://mmdetection.readthedocs.io/en/latest/ ### 📚 The doc issue I found no comment or documentations about config files, which is inconvient. For example, what's the main difference between files...

YOLO-Worldv2-XL is not provided

5

The link to the model at huggingface is actually for YOLO-Worldv2-X instead of YOLO-Worldv2-XL.

it will be appricated if the training code is completed!

12

The evaluate_crowd_no_overlap function is not implemented. When would this be completed?  Look forward to try APGCC!

I have test your inference code with few problems solved. The mae and mse value is great! But I ran into some problems when I tred to reproduce model. ...

cache_position is missing for input of function prepare_inputs_for_generation in class VITAQwen2ForCausalLM

3

as said in the title of this issue and showed in the picture below:  the envirnment is: V100, cuda 12.4, torch 2.3.1