Skip to content

[New operator] SparseLengthsSumFused8BitRowwise #1698

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Closed
artemrakhov-glow opened this issue Sep 24, 2018 · 5 comments
Closed

[New operator] SparseLengthsSumFused8BitRowwise #1698

artemrakhov-glow opened this issue Sep 24, 2018 · 5 comments
Assignees
@jfix71

Comments

@artemrakhov-glow
Copy link
Contributor

We need to support SparseLengthsSumFused8BitRowwise operator from Caffe2. Spec can be found here:
https://caffe2.ai/docs/operators-catalogue.html#sparselengthssumfused8bitrowwise

Also there is weighted version: SparseLengthsWeightedSumFused8BitRowwise
We already support float version: SparseLengthsSum / SparseLengthsWeightedSum
@rdzhabarov
Copy link
Contributor

cc: @beicy ^

@beicy beicy self-assigned this Oct 2, 2018
@artemrakhov-glow
Copy link
Contributor Author

@beicy : any updates on this? It is the most important among our unimplemented ops. Feel free to unassign if its not on your radar.

@beicy beicy removed their assignment Jan 8, 2019
@beicy
Copy link
Contributor

beicy commented Jan 8, 2019

@artemrakhov Sorry, forget that it is still assigned to me... Unsigned it for now.

@jfix71
Copy link
Contributor

jfix71 commented Jan 9, 2019

@artemrakhov Were you going to take this? Otherwise I can work on it.

@jfix71 jfix71 self-assigned this Jan 9, 2019
@artemrakhov-glow
Copy link
Contributor Author

No, I wasn't. Thanks for taking this, @jfix71

@jfix71 jfix71 mentioned this issue Jan 18, 2019
Merged
jfix71 added a commit that referenced this issue Jan 22, 2019
@jfix71
Implement RowwiseQuantizedSparseLengthsWeightedSum (#2282)
da381d5 
*Description*: Implement RowwiseQuantizedSparseLengthsWeightedSum. Added support for both Interpreter and CPU backends.

*Testing*: Added unit tests for both RowwiseQuantizedSparseLengthsWeightedSum and RowwiseQuantizedSparseLengthsSum (implemented as the first but with weights as a float splat of 1.0).

*Documentation*: Added to Quantization.md

Related to #1698
@jfix71 jfix71 mentioned this issue Jan 23, 2019
Merged
@jfix71 jfix71 mentioned this issue Feb 9, 2019
Merged
jfix71 added a commit that referenced this issue Feb 13, 2019
@jfix71
[New Operator] FusedRowwiseQuantizedSparseLengthsWeightedSumNode (#2368)
b975212 
*Description*: As noted in #2292, we decided to implement both fused and unfused versions of rowwise-quantized SLWS.

*Testing*: Added OperatorTests and Caffe2ImporterTests.

*Documentation*: Added.

Closes #1698
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@jfix71 jfix71

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
4 participants
@jfix71 @artemrakhov-glow @rdzhabarov @beicy