Skip to content

Conversation

@sdalvi-quic
Copy link
Contributor

Motivation : Unary ops calculate the value during runtime which is costly. Since we are handling quantized data hence, it is better to pre compute the value during compile time (for 0 -256 value) and store in a Look Up Table which is accessed during the runtime.

Implementation : This pass generates the Look Up Table for each unary op during compile time. It iterates over the original graph and replaces the unary ops call node, (example tanh, sqrt) to a call node that accesses a Look Up Table corresponding to the unary op, and return the value.

Testing : The test_take checks for both value correctness and structural correctness of the output generated after the pass.

@tvm-bot
Copy link
Collaborator

tvm-bot commented Aug 23, 2023

Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews from Reviewers by @-ing them in a comment.

  • No users to auto-tag found, no teams are specified in PR title See #10317 for details

Generated by tvm-bot

@sdalvi-quic
Copy link
Contributor Author

sdalvi-quic commented Aug 23, 2023

@jverma-quic @quic-sanirudh @kparzysz-quic
Please review the PR for upstream.

@sdalvi-quic sdalvi-quic changed the title [UPSTREAM] Implementation to replace unary ops with take op Replacing unary ops with LookUpTable and Take op to improve performance. Aug 25, 2023
@quic-sanirudh
Copy link
Contributor

@sdalvi-quic Can you fix the lint errors, and maybe also update the title to add [Unity] tag so that its obvious that this patch goes into unity branch?

@sdalvi-quic sdalvi-quic changed the title Replacing unary ops with LookUpTable and Take op to improve performance. [Unity] Replacing unary ops with LookUpTable and Take op to improve performance. Aug 29, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants