search

qibao77 / CFSNet

pytorch code of "CFSNet: Toward a Controllable Feature Space for Image Restoration"（ICCV2019）
62 stars 4 forks source link

α_in #5

Open marnie007 opened 4 years ago

marnie007 commented 4 years ago

In training step 1, we keep α_in= 0, while in training step 2, we keep α_in = 1. But in the paper , the experiment use different α_in from 0 to 1, what's the mean? Hope you can reply me, thank you.

qibao77 commented 4 years ago

Thank you for your attention. In training stage, we set α_in to 0 and 1 in the two training steps to establish the correspondence between control variables and optimization objectives. Therefore, during the test, we can change α_in from 0 to 1 to achieve continuous changes between the two states.

marnie007 commented 4 years ago

Thanks your reply. I know about it.