facebookresearch / GuidedDistillation

Official implementation of the paper "Guided Distillation for Semi-Supervised Instance Segmentation".
Other
34 stars 4 forks source link

Standards for data partitioning #8

Closed CStar-777 closed 7 months ago

CStar-777 commented 7 months ago

Hello, I have observed that when dividing training data, unlabeled data is used for all training data. Does this not comply with the semi-supervised task setting? If I misunderstood, please provide an explanation. Thank you.