About the metrics in training and testing

yucornetto / MGMatting

This repository includes the official project of Mask Guided (MG) Matting, presented in our paper: Mask Guided Matting via Progressive Refinement Network

Other

332 stars 48 forks source link

About the metrics in training and testing #37

Open crystraldo opened 1 year ago

crystraldo commented 1 year ago

Hi, I found that the sad/mse tested during training is different from the sad/mse tested with inference, what is the specific reason? Is it because of the dist problem?