您好,我使用了readme中的KITTI_Eigen (Swin-L)模型,加载的时候报错
Traceback (most recent call last):
File "iebins/train.py", line 526, in
main()
File "iebins/train.py", line 519, in main
main_worker(ngpus_per_node=ngpus_per_node,args = args)
File "iebins/train.py", line 199, in main_worker
model = NewCRFDepth(version=args.encoder, inv_depth=False, max_depth=args.max_depth, pretrained=args.pretrain)
File "/IEBins-main/iebins/networks/NewCRFDepth.py", line 96, in init
self.init_weights(pretrained=pretrained)
File /IEBins-main/iebins/networks/NewCRFDepth.py", line 106, in init_weights
self.backbone.init_weights(pretrained=pretrained)
File "/IEBins-main/iebins/networks/swin_transformer.py", line 584, in init_weights
load_checkpoint(self, pretrained, strict=False)
File "/IEBins-main/iebins/networks/newcrf_utils.py", line 248, in load_checkpoint
table_current = model.state_dict()[table_key]
KeyError: 'backbone.layers.0.blocks.0.attn.relative_position_bias_table'
您好,我使用了readme中的KITTI_Eigen (Swin-L)模型,加载的时候报错 Traceback (most recent call last): File "iebins/train.py", line 526, in
main()
File "iebins/train.py", line 519, in main
main_worker(ngpus_per_node=ngpus_per_node,args = args)
File "iebins/train.py", line 199, in main_worker
model = NewCRFDepth(version=args.encoder, inv_depth=False, max_depth=args.max_depth, pretrained=args.pretrain)
File "/IEBins-main/iebins/networks/NewCRFDepth.py", line 96, in init
self.init_weights(pretrained=pretrained)
File /IEBins-main/iebins/networks/NewCRFDepth.py", line 106, in init_weights
self.backbone.init_weights(pretrained=pretrained)
File "/IEBins-main/iebins/networks/swin_transformer.py", line 584, in init_weights
load_checkpoint(self, pretrained, strict=False)
File "/IEBins-main/iebins/networks/newcrf_utils.py", line 248, in load_checkpoint
table_current = model.state_dict()[table_key]
KeyError: 'backbone.layers.0.blocks.0.attn.relative_position_bias_table'