How to use ManualThreshold and set default_value？

[esas4]

I'm using anomalib 1.2.0.
I use python API method to train Patchcore on my own dataset, it goes well. But when I want to use ManualThreshold and set the default_value, I don't know how to set it.
This is my code:
`from anomalib.models import Patchcore
from anomalib.engine import Engine
from anomalib.data import Folder

datamodule = Folder(
name = 'test',
root='path/to/dataset',
normal_dir='real',
abnormal_dir='fake',
task='classification',
num_workers=0,
)

datamodule.setup()

_, train_data = next(enumerate(datamodule.train_dataloader()))
print(train_data.keys())
# dict_keys(['image_path', 'label', 'image'])

_, val_data = next(enumerate(datamodule.val_dataloader()))
print(val_data.keys())
# dict_keys(['image_path', 'label', 'image'])

_, test_data = next(enumerate(datamodule.test_dataloader()))
print(test_data.keys())
# dict_keys(['image_path', 'label', 'image'])

model = Patchcore()

engine = Engine(task='classification', threshold= "ManualThreshold")

engine.train(datamodule=datamodule, model=model)`

I could find that there is only one argument about threshold in Engine(), but it needs to be a string. I tried to change the default value in the init() of class ManualThreshold directly, but it didn't work. Where can I set the default_value of ManualThreshold?

I searched some issues, like https://github.com/openvinotoolkit/anomalib/issues/2229. In this issue, it seems that Engine() use a threshold of different type which is an object of class ManualThreshold.
I also found that training by using cli may be another choice, but I couldn't find the official config.yaml. If I use a copy of that in https://github.com/openvinotoolkit/anomalib/issues/2586, the error info also told me that threshold is expected to be a string, not the same as that in the copy.

[yelloji]

threshold:
class_path: anomalib.metrics.ManualThreshold #F1AdaptiveThreshold ManualThreshold
init_args:
default_value: 32

[github-actions[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.