How to run SuperSimpleNet in unsupervised manner?

[tuguldur102]

Hello, the SuperSimpleNet SOTA model was added in the latest version. Thank you for your team effort :). I have datasets such as normal and abnormal images without annotation boxes or segmentations. When I'm running SuperSimpleNet on my dataset the error given was an Attribute error on "the mask function" which is desired for segmentation handling. I have found the source repository GitHub of SuperSimpleNet. There are datasets such as KSDD2 similar to my dataset, such as not containing segmentation annotations and structured in normal and abnormal folders. Therefore, how to run SuperSimpleNet in this unsupervised manner without annotation? Thank you.