-
Notifications
You must be signed in to change notification settings - Fork 74k
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
[TF-TRT] Cast Converter Re-Engineered #55491
[TF-TRT] Cast Converter Re-Engineered #55491
Conversation
e4599fb
to
769b2a7
Compare
I have two general comment on this PR: |
Hi @DEKHTIARJonathan Can you please check @bixia1's comments and keep us posted ? Thank you! |
@DEKHTIARJonathan Any update on this PR? Please. Thank you! |
ad942a9
to
4b7d551
Compare
What I mean by
Yes absolutely, it is intentional. This PR changes the way |
4b7d551
to
41b8fd4
Compare
@bixia1 please review. This PR is good to go ;) |
This PR changes the way TF-TRT deal with
Cast
nodes. TensorRT engineers advised us to treatCast
as an Identity node and let TensorRT decides on which compute precision to use according toprecision_mode=...
.This behavior can be deactivated using
TF_TRT_EXPERIMENTAL_FEATURES=reject_fp32_fp16_cast
if needed to work around any unforeseen issueThis PR also adds
converter.summary()
to TF-TRT test files in order to ease test debugability