移植到mmengine 2.*, model加载问题 #93
Description
我想将代码移植到mmengine 2.*版本，但是mmengine中取消了MMDataParallel, 我换成了torch.nn.DataParallel.
在加载模型的时候会报错： File "./tools/train.py", line 104, in
main()
File "./tools/train.py", line 100, in main
runner.train()
File "/root/miniconda3/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1748, in train
self._init_model_weights()
File "/root/miniconda3/lib/python3.8/site-packages/mmengine/runner/runner.py", line 923, in _init_model_weights
model.init_weights()
File "/root/miniconda3/lib/python3.8/site-packages/mmengine/model/base_module.py", line 136, in init_weights
m.init_weights()
File "/root/mmseg/mmseg/models/backbones/vit_bimla.py", line 339, in init_weights
load_pretrained(self, num_classes=self.num_classes, in_chans=self.in_chans, pos_embed_interp=self.pos_embed_interp, num_patches=self.patch_embed.num_patches, align_corners=self.align_corners)
File "/root/mmseg/mmseg/models/backbones/helpers.py", line 60, in load_pretrained
state_dict = torch.load(cfg['pretrained_finetune'])
File "/root/miniconda3/lib/python3.8/site-packages/torch/serialization.py", line 713, in load
return _legacy_load(opened_file, map_location, pickle_module, **pickle_load_args)
File "/root/miniconda3/lib/python3.8/site-packages/torch/serialization.py", line 938, in _legacy_load
typed_storage._storage._set_from_file(
RuntimeError: storage has wrong byte size: expected %ld got %ld04096
加载的是pretrain/jx_vit_large_p16_384-b3be5167.pth