Skip to content
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.

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

[Tracking issue] Quantized TFLite operators #9187

Closed
13 tasks done
leandron opened this issue Oct 4, 2021 · 5 comments
Closed
13 tasks done

[Tracking issue] Quantized TFLite operators #9187

leandron opened this issue Oct 4, 2021 · 5 comments

Comments

@leandron
Copy link
Contributor

leandron commented Oct 4, 2021

As part of #9165, we recently enabled support for quantized operators in the TFLite frontend "unary element-wise" function, that does generic support for a few operators.

That opens the door for more operators to be supported easily, just pending the addition of tests and logically unlocking the conditional block that is keeping the quantised version of the operator to be unsupported. See the example as implemented in #9165.

This is a tracking issue for new unary element-wise operators to be supported:

For each operator above, there are actions to be taken:

  1. Remove the conditional block if self.is_quantized(op): ...
  2. Implement unit tests
  3. ask any committer in the project to update this issue, reflecting the PR to support the operators on the list above

#9165 can be used as an example.

Resources:

@onkar-sima-ai
Copy link
Contributor

Hi @leandron can you please update this issue and if possible review the PR for qunatized SIN operator?
I have followed the instruction in this issue to write the test cases.

Please let me know if I need to do any changes.
Thanks for posting, Happy to contribute.

@leandron
Copy link
Contributor Author

leandron commented Nov 9, 2021

Hi @leandron can you please update this issue and if possible review the PR for qunatized SIN operator? I have followed the instruction in this issue to write the test cases.

Please let me know if I need to do any changes. Thanks for posting, Happy to contribute.

Done! thanks @onkar-sima-ai

@psigillito
Copy link

Hi @leandron I am interested in getting to know the source code better. Is the list of operators up to date? I might try implementing the GREATER operator using #9165 as a template. Thanks!

@leandron
Copy link
Contributor Author

Hi @leandron I am interested in getting to know the source code better. Is the list of operators up to date? I might try implementing the GREATER operator using #9165 as a template. Thanks!

Sorry I missed your message. I've just updated the list of operator.
To your question: yes, you can use #9165 as a template.

dchauhan-arm added a commit to dchauhan-arm/tvm that referenced this issue Jun 1, 2022
Support GREATER quantization operation conversion as part of issue apache#9187
dchauhan-arm added a commit to dchauhan-arm/tvm that referenced this issue Jun 1, 2022
Support EQUAL quantization operation conversion as part of issue apache#9187
leandron pushed a commit that referenced this issue Jun 22, 2022
* [TFLite] Support quantized EQUAL op in TFLite frontend

Support EQUAL quantization operation conversion as part of issue #9187

* [TFLite] Support quantized EQUAL op in TFLite frontend

Update elementwise quantized test for EQUAL op
Change-Id: I3897d1ac07051ebfc10356ad45397117b592f878
blackkker pushed a commit to blackkker/tvm that referenced this issue Jul 7, 2022
* [TFLite] Support quantized EQUAL op in TFLite frontend

Support EQUAL quantization operation conversion as part of issue apache#9187

* [TFLite] Support quantized EQUAL op in TFLite frontend

Update elementwise quantized test for EQUAL op
Change-Id: I3897d1ac07051ebfc10356ad45397117b592f878
dchauhan-arm added a commit to dchauhan-arm/tvm that referenced this issue Sep 12, 2022
Support GREATER quantization operation conversion as part of issue apache#9187
Continuation of apache#11519
lhutton1 pushed a commit that referenced this issue Sep 12, 2022
Support GREATER quantization operation conversion as part of issue #9187 Continuation of #11519.
@lhutton1
Copy link
Contributor

Thanks @dchauhan-arm @leeexyz @onkar-sima-ai @elinx @ophirfrish for the contributions towards this effort! Closing as it is now complete.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

4 participants