This repository has been archived by the owner on Aug 15, 2019. It is now read-only.
Fixing various nan related issues impacted by floating point precision and NaN #1116
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
…s (most likely due to gpu driver changes).
|
Reviewed 3 of 3 files at r2. Comments from Reviewable |
complete! 1 of 1 LGTMs obtained
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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 fixing issue: tensorflow/tfjs#291
On the newer macs we had issues computing pow, mod, acosh, and atanh due to various issues related to NaN not being properly handled at the GPU level.
In the case of acosh and mod the problem stemmed from Div not successfully dividing 3/3 (or 6/6, 7/7, etc). In those cases the result would be .99999 and when floor was applied we got 0 instead of 1.
In the case of pow the problem stemmed from log not catching NaN for negative numbers.
Lastly, atanh failed due to not checking the inputs for values outside of the range [-1, 1]. For those values atanh needed to just return NaN directly.
This change is