TFTRT: LOG(ERROR) --> VLOG(1) when use_calibration=True with fp32/fp16 #24302
Conversation
This file contains 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
pooyadavoodi
requested review from
aaroey,
azaks2 and
smit-hinsu
as code owners
pooyadavoodi
force-pushed
the
tftrt_change_log_use_calibration
branch
from
8165173
to
ecb1d04
Compare
|
Can we use |
smit-hinsu
requested review from
aaroey
and removed request for
aaroey and
azaks2
@aaroey WARNING is definitely better than ERROR. But I still wouldn't print a warning just because it would confuse users when they see the message. My guess is that your concern is about users who actually want to do calibration with fp32/fp16, but I don't think anyone would be interested in that functionality. Please let me know if you still want to change it to WARNING. |
|
@pooyadavoodi fair enough, thanks! |
aaroey
approved these changes
Dec 12, 2018
tensorflow-copybara
pushed a commit
that referenced
this pull request
Dec 13, 2018
…bration PiperOrigin-RevId: 225400493
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
This is just to keep the default LOG cleaner.
The API mentions that
use_calibrationis ignored in case of precisions fp32/fp16.My preference would be to change the default to
use_calibration = Falsebut that's a change in API behavior, and @aaroey asked to postpone that to TF2.0.