`StrixDataset` error with `Dataset` type

[ChenglongWang]

StrixDataset(
filelist,
loader=LoadImageD(keys=img_lbl_key, dtype=np.float32),
channeler=None,
orienter=None,
spacer=None,
rescaler=None,
resizer=None,
cropper=None,
caster=CastToTypeD(keys=img_lbl_key, dtype=[np.float32, np.uint8]),
to_tensor=ToTensorD(keys=img_lbl_key),
is_supervised=True,
dataset_type=Dataset, # CacheDataset
dataset_kwargs=None, #{"cache_rate": opts.get("preload", 0)}
additional_transforms=None,
check_data=True,
)

<class 'torch.utils.data.dataset.Dataset'> <class 'type'>
Exception trace:
Exception entry is:
  File "/homes/clwang/Code/Strix/strix/data_checker.py", line 205, in get_train_valid_datasets
    train_ds = dataset_fn(files_train, Phases.TRAIN, auxilary_params)

    ↑ Input arguments: () {}
Exception occured at:
  File "/homes/clwang/Code/Strix/strix/data_io/base_dataset/basic_dataset.py", line 79, in __new__
    return self.dataset(self.input_data, transform=self.transforms)

TypeError: object.__new__() takes exactly one argument (the type to instantiate)