作者您好，运行visualize.py时报错

[niuyixing]

latest.pth Traceback (most recent call last): File "D:/domain/Transfer-Learning-Library-master/examples/domain_adaptation/object_detection/visualize.py", line 142, in launch( File "d:\domain\detectron2-main\detectron2\engine\launch.py", line 82, in launch main_func(args) File "D:/domain/Transfer-Learning-Library-master/examples/domain_adaptation/object_detection/visualize.py", line 97, in main DetectionCheckpointer(model, save_dir=cfg.OUTPUT_DIR).resume_or_load( File "C:\Users\xingze\AppData\Roaming\Python\Python38\site-packages\fvcore\common\checkpoint.py", line 229, in resume_or_load return self.load(path, checkpointables=[]) File "d:\domain\detectron2-main\detectron2\checkpoint\detection_checkpoint.py", line 52, in load ret = super().load(path, args, **kwargs) File "C:\Users\xingze\AppData\Roaming\Python\Python38\site-packages\fvcore\common\checkpoint.py", line 158, in load incompatible = self._load_model(checkpoint) File "d:\domain\detectron2-main\detectron2\checkpoint\detection_checkpoint.py", line 96, in _load_model self._convert_ndarray_to_tensor(checkpoint["model"]) File "C:\Users\xingze\AppData\Roaming\Python\Python38\site-packages\fvcore\common\checkpoint.py", line 371, in _convert_ndarray_to_tensor raise ValueError( ValueError: Unsupported type found in checkpoint! netG_S2T: <class 'collections.OrderedDict'>

进程已结束，退出代码 1

报错位置：

model could be an OrderedDict with _metadata attribute

    # (as returned by Pytorch's state_dict()). We should preserve these
    # properties.
    for k in list(state_dict.keys()):
        v = state_dict[k]
        if not isinstance(v, np.ndarray) and not isinstance(v, torch.Tensor):
            raise ValueError(
                "Unsupported type found in checkpoint! {}: {}".format(k, type(v))
            )
        if not isinstance(v, torch.Tensor):
            state_dict[k] = torch.from_numpy(v)

还要问下如何将自己的数据、图片根据训练权重全部进行转化，生成所有每张原图像对应的结果图片

[thucbx99]

根据上面的报错信息来看是使用detectron2读取模型参数时出现的bug，具体原因建议参考detectron2有关这一部分的文档。

使用自己的数据也是类似，detectron2默认提供了两种格式，即coco和pascal voc，其中pascal voc的实现可以参考库里的代码。