[ETHOSN] Transpose fully connected weights #12970
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
asparkhi
reviewed
Oct 3, 2022
Contributor
asparkhi
left a comment
asparkhi
left a comment
There was a problem hiding this comment.
Thanks Luke for the fix. Surprised that none of the common networks picked it up. Just a few nits and questions.
The NPU driver stack expects weights in IO (HWIO) format, however, Relay uses an OI representation. Although the shape of the weight tensor was correctly changed during codegen, the values in the weights tensor were not being transposed. This lead to an output mismatch when the output "units" was > 1. The tests didn't catch this due to using a weights tensor of all 1's. Change-Id: I51b2bcd14b677280ef3b6a6845d56b7dfacc7d6a
lhutton1
force-pushed
the
fully-connected-fix
branch
from
e14bfcf to
66972bf
Compare
* Refactor use of weight transpose to common file between contrib codegens. * Make function areguments more explicit. * Update network hashes. Change-Id: Ib53bc7d2837b62908b92fd09062cbe9a8bb4ab30
lhutton1
force-pushed
the
fully-connected-fix
branch
from
66972bf to
53b3383
Compare
Change-Id: Ie89e429da222ffe17bc8faf831bf59217008a68a
Contributor
Author
|
Thanks for the review @ashutosh-arm, this is ready for another look :) |
asparkhi
approved these changes
Oct 6, 2022
Contributor
asparkhi
left a comment
asparkhi
left a comment
There was a problem hiding this comment.
Thanks Luke for making the changes. Looks much cleaner with a separate API. I've left few nits to follow up in the future.
Change-Id: Ie5ded2db3024b9e2c5095f01adea65798fc1da55
xinetzone
pushed a commit
to daobook/tvm
that referenced
this pull request
Nov 25, 2022
* [ETHOSN] Transpose fully connected weights The NPU driver stack expects weights in IO (HWIO) format, however, Relay uses an OI representation. Although the shape of the weight tensor was correctly changed during codegen, the values in the weights tensor were not being transposed. This lead to an output mismatch when the output "units" was > 1. The tests didn't catch this due to using a weights tensor of all 1's. Change-Id: I51b2bcd14b677280ef3b6a6845d56b7dfacc7d6a * Address comments * Refactor use of weight transpose to common file between contrib codegens. * Make function areguments more explicit. * Update network hashes. Change-Id: Ib53bc7d2837b62908b92fd09062cbe9a8bb4ab30 * Fix lint Change-Id: I6a1d9ffa8e3a747b7c77c9b27aa1b1c0d4c5cbff * Fix cmsis-nn weights transpose Change-Id: Ie89e429da222ffe17bc8faf831bf59217008a68a * Address comments Change-Id: Ie5ded2db3024b9e2c5095f01adea65798fc1da55
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.
The NPU driver stack expects weights in IO (HWIO) format, however, Relay uses an OI representation. Although the shape of the weight tensor was correctly changed during codegen, the values in the weights tensor were not being transposed. This lead to an output mismatch when the output "units" was > 1. The tests didn't catch this due to using a weights tensor of all 1's.
cc @leandron @ashutosh-arm