Is the validate in training necessary?

poppinace / indexnet_matting

(ICCV'19) Indices Matter: Learning to Index for Deep Image Matting

Other

389 stars 90 forks source link

Open QWERDF007 opened 4 years ago

QWERDF007 commented 4 years ago

the validate is too slow, its batch size is 1. If the size of test set is very large, it will take a lot of time.

poppinace commented 4 years ago

It is necessary. The best performance does not always appear in the final epoch. To save time, you can enable validation after 20 epochs.

QWERDF007 commented 4 years ago

thanks.