is there any ImageNet pre-training config of Topformer?
wmkai opened this issue · comments
I want to change some modules when using Topformer to do classification on ImageNet. Could you please provide the ImageNet Pre-training config? Thanks!
Please look into the close issue. You could obtain the classification model by removing SIM in the Seg model and adding a Linear layer on the top of the Transformer block (SASE) as the classification head.
If it is convenient for you, can I have your wechat and share the code of imagenet pre-training?