ChenDirk

> try this > > ``` > from mmcv.runner import get_dist_info, init_dist, load_checkpoint > from mmrazor.models.builder import build_algorithm > > def split_student_model(cls_cfg_path, cls_model_path, device='cuda', save_path): > """ > :param: cls_cfg_path:...

> Hi, Thank you for your code. I try your method in classification and segmentation model, both of them can split the student model successfully . BTW, I suggest the...

> Hi @johnwlambert , I have a copy of v1.2 and v2.0 but cannot find v1.1 on the official website. Is there a problem with using v1.2? I have the...

> Thank you for your reply, I check the code, the implementation method is amazing! I also find the LayerNorm is very slow in inference, and the batchnorm can merge...