Add log1p elementwise op #993
Open
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
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D54176180 |
22quinn
force-pushed
the
export-D54176180
branch
from
b7a9b04
to
c938fba
Compare
|
This pull request was exported from Phabricator. Differential Revision: D54176180 |
22quinn
force-pushed
the
export-D54176180
branch
from
c938fba
to
63ad837
Compare
|
This pull request was exported from Phabricator. Differential Revision: D54176180 |
22quinn
force-pushed
the
export-D54176180
branch
from
63ad837
to
0902435
Compare
|
This pull request was exported from Phabricator. Differential Revision: D54176180 |
Summary: `log1p(x)` is more precise than `log(1+x)` when `x` is close to 0. We utilize cuda `log1pf` implementation for fp32. For other precision types, input is first converted to float, then `log1pf` is computed, finally output is converted back to original precision. CUDA log1pf function for float and double: https://docs.nvidia.com/cuda/cuda-math-api/group__CUDA__MATH__SINGLE.html Reviewed By: frank-wei Differential Revision: D54176180
22quinn
force-pushed
the
export-D54176180
branch
from
0902435
to
23e93f2
Compare
|
This pull request was exported from Phabricator. Differential Revision: D54176180 |
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Outdated suggestions cannot be applied.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.
Summary:
log1p(x)is more precise thanlog(1+x)whenxis close to 0. We utilize cudalog1pfimplementation for fp32. For other precision types, input is first converted to float, thenlog1pfis computed, finally output is converted back to original precision.CUDA log1pf function for float and double: https://docs.nvidia.com/cuda/cuda-math-api/group__CUDA__MATH__SINGLE.html
Differential Revision: D54176180