modify the codes to train all categories

[JieLi1994]

Firstly, thanks for your great job.

I think it's not reasonable for a category to correspond to a model. If I want to modify the code to train all the categories at the same time. Do you have any suggestions? Which part of the code is needed to be modified

[sshaoshuai]

Maybe you should change the classification layers of RCNN stage and the corresponding BCE loss to some other losses like CE loss. Also it's better to have more mean sizes for different classes.

[JieLi1994]

thanks for your answer, I will tell you about the progress and problems in the follow-up.

[sherrygp]

Firstly, thanks for your great job.

I think it's not reasonable for a category to correspond to a model. If I want to modify the code to train all the categories at the same time. Do you have any suggestions? Which part of the code is needed to be modified

Hello,i am also want to train all the categories at the same time.Have you achieved it?

[JieLi1994]

@sherrygp Not yet, i'm working to achieve it.

[sherrygp]

@JieLi1994 Me too, more exchanges.
Can I leave a way to communicate?

[JieLi1994]

@sherrygp You can email me([email protected])

[dingfuzhou]

Hi

@sherrygp You can email me([email protected])

Hi, have you obtained some progress for the modification now?

Cheers,

[lyuweiwang]

I am working on this and I find maybe everyone under this issue is Chinese. Thus I wonder if we could communicate with we-chat. My we-chat is lao5-king and I can group us.