[MultiThreshold] Generalize data layouts for node execution #143
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
This allows node execution of MultiThreshold operators with arbitrary number of dimensions, as long as the channel dimension is last. This is necessary to run some verification steps of attention operators which, at least for some intermediate steps, have 3 dimensional data layouts. This does not change the behavior of execution on the already existing 2d and 4d data layouts.
The relevant aspect of the data layout annotation seems to be which axis is labeled as the channel dimension "C": We do not actually have to care about the total number and ordering of the other axes, as long as we can find the index of the "C" axis and swap to have "C" at index 1 for node execution (and swap it back afterwards). Falls back to the default assumption that "C" is at index 1 if there is no layout annotation, which is equivalent to the "NCHW" or "NC" layouts. This is a rather experimental change which might break existing code and is currently still restricted to the well-known 2-, 3- and 4-dimensional layouts.
As we only really care for the index of the "C" axis there is no need to restrict the set of valid layouts here.
Note: Only covers data layouts for tensors with less than five axes
|
Instead of relying on data layout strings, how about we switch to an attribute |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
The relevant aspect of the data layout annotation seems to be which axis is labeled as the channel dimension "C": We do not actually have to care about the total number and ordering of the other axes, as long as we can find the index of the "C" axis and swap to have "C" at index 1 for node execution (and swap it back afterwards).
Falls back to the default assumption that "C" is at index 1 if there is no layout annotation, which is equivalent to the "NCHW" or "NC" layouts.
This is a rather experimental change which might break existing code and is currently still restricted to the well-known 2-, 3- and 4-dimensional layouts.
This PR is based on #92 which is less experimental and should be merged first (#92 also does not risk breaking existing code as it only adds new special cases).