Allow trivial slice/concat #3014
Merged
Conversation
This file contains hidden or 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
Member
|
LGTM :) I'll take a closer look tomorrow. |
Contributor
|
Looks good to me. This is the correct behavior IMO. |
db56f86 to
784dfdd
Compare
Member
|
Took another pass today and looks ready to merge. Thanks @jeffdonahue |
myfavouritekk
added a commit
to myfavouritekk/caffe
that referenced
this pull request
Aug 15, 2016
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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Outdated suggestions cannot be applied.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.
I'm pulling this out from #2033 for separate review. This allows trivial (identity) operation of the slice & concat layers by changing their
MinTopBlobsorMinBottomBlobs(respectively) from 2 to 1. For performance, it is special cased to useShareDataandShareDiffin this case, as done for other operations that correspond to the identity when the input & output Blobs are viewed as a flat data array (e.g. Reshape and Flatten).This is used in #2033 to allow single-timestep operation of recurrent layers without an annoying special case.
@ronghanghu you recently commented that this would make your work easier as well -- maybe you could review this?