Do not fail on lack of default precision set. #6139
Merged
+16
−3
Conversation
This file contains hidden or 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
I discovered some models from the suite do not have the precision set so instead of failing the script we just log the case, and use the default precision, as no additional machinery should run for the Inductor anyway. Additionally I wrapped the exceptions with the ValueError so the logging message will not pollute with info about str not inheriting from Exception class. ecg@, note that needs to be hooked "somewhere". Not sure where, as there was a revert in #6134.
|
Thanks Greg -- I had missed that your change was reverted as part of another revert. Please land this (including the PR description when merging, otherwise annoyingly we lose that info in the repo) and I'll try to re-land your reverted change. |
cota
approved these changes
Dec 13, 2023
frgossen
approved these changes
Dec 13, 2023
Given that Jack is looking into this at the bridge level, I'll wait for him to finish that investigation since that's the right place to fix this. |
chunnienc
pushed a commit
to chunnienc/xla
that referenced
this pull request
Dec 14, 2023
I discovered some models from the suite do not have the default precision set so instead of failing the script we just log the case, and do nothing, as no additional machinery should run for the Inductor anyway. Additionally I wrapped the exceptions with the ValueError so the logging message will not pollute with info about str not inheriting from Exception class. @cota , note that needs to be hooked "somewhere". Not sure where, as there was a revert in pytorch#6134, but in general it can be done prior to moving the model to the device safely.
golechwierowicz
added a commit
that referenced
this pull request
Jan 12, 2024
I discovered some models from the suite do not have the default precision set so instead of failing the script we just log the case, and do nothing, as no additional machinery should run for the Inductor anyway. Additionally I wrapped the exceptions with the ValueError so the logging message will not pollute with info about str not inheriting from Exception class. @cota , note that needs to be hooked "somewhere". Not sure where, as there was a revert in #6134, but in general it can be done prior to moving the model to the device safely.
ysiraichi
reviewed
Jan 23, 2024
Comment on lines
228
to
+230
| elif precision == "amp": | ||
| raise f"AMP for PT/XLA:GPU is not implemented yet for torchbench models" | ||
| raise ValueError( | ||
| f"AMP for PT/XLA:GPU is not implemented yet for torchbench models") |
There was a problem hiding this comment.
@miladm @frgossen @golechwierowicz Isn't AMP supported on PT/XLA:GPU?
bhavya01
pushed a commit
that referenced
this pull request
Apr 22, 2024
I discovered some models from the suite do not have the default precision set so instead of failing the script we just log the case, and do nothing, as no additional machinery should run for the Inductor anyway. Additionally I wrapped the exceptions with the ValueError so the logging message will not pollute with info about str not inheriting from Exception class. @cota , note that needs to be hooked "somewhere". Not sure where, as there was a revert in #6134, but in general it can be done prior to moving the model to the device safely.
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.
I discovered some models from the suite do not have the default precision set so instead of failing the script we just log the case, and do nothing, as no additional machinery should run for the Inductor anyway. Additionally I wrapped the exceptions with the ValueError so the logging message will not pollute with info about str not inheriting from Exception class.
@cota , note that needs to be hooked "somewhere". Not sure where, as there was a revert in #6134, but in general it can be done prior to moving the model to the device safely.