About Categories model

[ytzfhqs]

Before Asking

• [X] I have read the README carefully. 我已经仔细阅读了README上的操作指引。

• [X] I want to train my custom dataset, and I have read the tutorials for finetune on your data carefully and organize my dataset correctly; 我想训练自定义数据集，我已经仔细阅读了训练自定义数据的教程，以及按照正确的目录结构存放数据集。

• [X] I have pulled the latest code of main branch to run again and the problem still existed. 我已经拉取了主分支上最新的代码，重新运行之后，问题仍不能解决。

Search before asking

• [X] I have searched the DAMO-YOLO issues and found no similar questions.

Question

我想使用阅读文档中的通用检测模型（80-categories-DAMO-YOLO-S和701-categories-DAMO-YOLO-S）但在运行demo.py文件时发现configs文件夹中没有对应配置文件，我尝试使用damoyolo_tinynasL25_S.py文件作为配置文件使用80-categories-DAMO-YOLO-S，但是出现报错：

Inference with torch engine!
2023-05-11 23:17:45.885 | ERROR    | __main__:<module>:367 - An error has been caught in function '<module>', process 'MainProcess' (7828), thread 'MainThread' (13596):
Traceback (most recent call last):

> File "C:\Users\lenovo\Desktop\DAMO-YOLO-master\tools\demo.py", line 367, in <module>
    main()
    └ <function main at 0x000001F5DC15ED30>

  File "C:\Users\lenovo\Desktop\DAMO-YOLO-master\tools\demo.py", line 326, in main
    infer_engine = Infer(config, infer_size=args.infer_size, device=args.device,
                   │     │                  │    │                  │    └ 'cpu'
                   │     │                  │    │                  └ Namespace(input_type='image', config_file='./configs/damoyolo_tinynasL25_S.py', path='./assets/dog.jpg', camid=0, engine='./d...
                   │     │                  │    └ [640, 640]
                   │     │                  └ Namespace(input_type='image', config_file='./configs/damoyolo_tinynasL25_S.py', path='./assets/dog.jpg', camid=0, engine='./d...
                   │     └ ╒═════════╤═══════════════════════════════════════════════════════════════════════════════════════════════════╕
                   │       │ keys    │ v...
                   └ <class '__main__.Infer'>

  File "C:\Users\lenovo\Desktop\DAMO-YOLO-master\tools\demo.py", line 54, in __init__
    self.model = self._build_engine(self.config, self.engine_type)
    │            │    │             │    │       │    └ 'torch'
    │            │    │             │    │       └ <__main__.Infer object at 0x000001F5DC1673D0>
    │            │    │             │    └ ╒═════════╤═══════════════════════════════════════════════════════════════════════════════════════════════════╕
    │            │    │             │      │ keys    │ v...
    │            │    │             └ <__main__.Infer object at 0x000001F5DC1673D0>
    │            │    └ <function Infer._build_engine at 0x000001F5DC15E8B0>
    │            └ <__main__.Infer object at 0x000001F5DC1673D0>
    └ <__main__.Infer object at 0x000001F5DC1673D0>

  File "C:\Users\lenovo\Desktop\DAMO-YOLO-master\tools\demo.py", line 76, in _build_engine
    model.load_state_dict(ckpt['model'], strict=True)
    │     │               └ {'model': OrderedDict([('backbone.block_list.0.conv.conv.weight', tensor([[[[-2.9148e-03,  8.9314e-03, -8.4367e-03],
    │     │                         ...
    │     └ <function Module.load_state_dict at 0x000001F5D81AB310>
    └ Detector(
        (backbone): TinyNAS(
          (block_list): ModuleList(
            (0): Focus(
              (conv): ConvBNAct(
                (conv):...

  File "D:\Python\lib\site-packages\torch\nn\modules\module.py", line 1482, in load_state_dict
    raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(

RuntimeError: Error(s) in loading state_dict for Detector:
    Unexpected key(s) in state_dict: "backbone.block_list.1.block_list.0.residual_proj.conv1.weight", "backbone.block_list.1.block_list.0.residual_proj.bn1.weight", "backbone.block_list.1.block_list.0.residual_proj.bn1.bias", "backbone.block_list.1.block_list.0.residual_proj.bn1.running_mean", "backbone.block_list.1.block_list.0.residual_proj.bn1.running_var", "backbone.block_list.1.block_list.0.residual_proj.bn1.num_batches_tracked", "backbone.block_list.3.block_list.0.residual_proj.conv1.weight", "backbone.block_list.3.block_list.0.residual_proj.bn1.weight", "backbone.block_list.3.block_list.0.residual_proj.bn1.bias", "backbone.block_list.3.block_list.0.residual_proj.bn1.running_mean", "backbone.block_list.3.block_list.0.residual_proj.bn1.running_var", "backbone.block_list.3.block_list.0.residual_proj.bn1.num_batches_tracked", "backbone.block_list.5.block_list.0.residual_proj.conv1.weight", "backbone.block_list.5.block_list.0.residual_proj.bn1.weight", "backbone.block_list.5.block_list.0.residual_proj.bn1.bias", "backbone.block_list.5.block_list.0.residual_proj.bn1.running_mean", "backbone.block_list.5.block_list.0.residual_proj.bn1.running_var", "backbone.block_list.5.block_list.0.residual_proj.bn1.num_batches_tracked". 
    size mismatch for head.gfl_cls.0.weight: copying a param with shape torch.Size([81, 128, 3, 3]) from checkpoint, the shape in current model is torch.Size([80, 128, 3, 3]).
    size mismatch for head.gfl_cls.0.bias: copying a param with shape torch.Size([81]) from checkpoint, the shape in current model is torch.Size([80]).
    size mismatch for head.gfl_cls.1.weight: copying a param with shape torch.Size([81, 256, 3, 3]) from checkpoint, the shape in current model is torch.Size([80, 256, 3, 3]).
    size mismatch for head.gfl_cls.1.bias: copying a param with shape torch.Size([81]) from checkpoint, the shape in current model is torch.Size([80]).
    size mismatch for head.gfl_cls.2.weight: copying a param with shape torch.Size([81, 512, 3, 3]) from checkpoint, the shape in current model is torch.Size([80, 512, 3, 3]).
    size mismatch for head.gfl_cls.2.bias: copying a param with shape torch.Size([81]) from checkpoint, the shape in current model is torch.Size([80]).

期待该问题得到解决。

Additional

No response

[KotovNikitaStudent]

Try to use load_state_dict(torch.load("your_model.pth"), strict=False).