This repository has been archived by the owner on Aug 11, 2020. It is now read-only.
[RFC] add ignore-label for image segmentation #87
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
@piiswrong can you review this? |
|
Please make a consistent CPU implementation as well, and document this behavior carefully. As this is not same as normal SoftmaxGrad, but something like special unknown label |
|
LGTM |
tqchen
added a commit
that referenced
this pull request
Dec 20, 2015
[RFC] add ignore-label for image segmentation
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
hello,
i have refactored the example fcn-xs in https://github.com/tornadomeet/mxnet/tree/seg/example/fcn-xs, and in image segmentation, the label image is like this http://host.robots.ox.ac.uk/pascal/VOC/voc2012/segexamples/index.html
as you can see, the index 255 corresponds to 'void' or unlabelled for ambiguity in the edge of object, so i add one SoftmaxGrad function for this problem, i just follow the caffe implentation https://github.com/longjon/caffe/blob/future/src/caffe/layers/softmax_loss_layer.cpp#L103
and we can debat this, maybe there is a better way do it.
thanks~