For encoded convolution, add check for when min_width would have been larger than in_width #610
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
jmitrevs
changed the title
|
This has been tested for 1D, and could be merged in the more restricted sense, but it has not been tested for 2D. I think planning should take into account the future of encoded. |
jmitrevs
changed the title
|
I will restrict this PR to the 1D case because I think that's the only one that would be important. If we fix it for 2D, we can make a different PR. |
vloncar
requested changes
Oct 14, 2022
|
There are some issues still, so I am converting to a draft till I get a chance to look. |
jmitrevs
added
please test
vloncar
reviewed
Jan 10, 2023
jmitrevs
added
please test
jmitrevs
added
please test
|
This should now also fix the same issue with Conv2D. |
|
@vloncar, are you happy with the changes here as they are? |
vloncar
reviewed
Jan 13, 2023
jmitrevs
added
please test
vloncar
reviewed
Jan 13, 2023
jmitrevs
added
please test
|
Is there anything left to do here before it's merged? |
vloncar
approved these changes
Feb 17, 2023
calad0i
pushed a commit
to calad0i/hls4ml
that referenced
this pull request
Jul 1, 2023
… larger than in_width (fastmachinelearning#610) * add special case when W == min_width. * add test that fails without this fix * add comments * Make simple bug fixes before reworking structure * Try to cleanup how we handle the narrow convolutions * fix hardcoded returns * update convolution templates for narrow encoded * flake8 formatting * update narrow for conv2d, fix test * change narrow to unscaled, always do unscaled when in_W == min_W * pre-commit fixes for nnet_conv1d_stream.h and nnet_conv2d_stream.h * remove unneeded attributes * change hardcoded instructions condition
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.
Description
In the encoded convolution, the case when the in_width is smaller than the nominal min_width needs to be handled as a special case. This attempts to fix this situation. It has been been minimally tested in a conv1d case where in_width = 11 while the kernel width is 9. Additional tests need to be made.
Type of change
For a new feature or function, please create an issue first to discuss it
with us before submitting a pull request.
Note: Please delete options that are not relevant.
Tests
test_conv1d_narrow.pyfails in main, but passes after this fix. Similarly withtest_conv2d_narrow.py.Checklist