Skip to content
#

drn

Here are 3 public repositories matching this topic...

shaibagon
shaibagon commented Jan 7, 2019

When training, the augmentation RandomScaleCrop may downscale the image and the target label image. It then pads the image and the label with [self.fill][1] which is ZERO.
This is in contrast to the "ignore value" of the loss [that is set to 255][2].
This way the loss treats the padded region as valid "class 0" pixels and compute loss for it.

self.fill of the augmentation functions

Improve this page

Add a description, image, and links to the drn topic page so that developers can more easily learn about it.

Curate this topic

Add this topic to your repo

To associate your repository with the drn topic, visit your repo's landing page and select "manage topics."

Learn more

You can’t perform that action at this time.