Partial convolutions for Multiple-*SSA #235
Open
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.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.
When window length in N-dimensional SSA ends with 1 (for example in Multiple N-d SSA), it is possible to replace full N-d FFT's with N-1 dimensional ones.
Changes are limited to convolution routine, thus are applicable for shaped and weight extensions of SSA; test-cases with applicable window sizes already exist in testthat.
Time measurements for decomposing MNIST (28x28x60000 input, 14x14x1 window; OMP_NUM_THREADS=1, R version 3.4.1):
(usr system wall-clock)
This change is