[Semi-Auto] Add cross_entropy_with_softmax infer_backward rule #56507
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
|
你的PR提交成功，感谢你对开源项目的贡献! |
|
Sorry to inform you that 516c727's CIs have passed for more than 7 days. To prevent PR conflicts, you need to re-run all CIs manually. |
pkuzyc
force-pushed
the
cross_entropy_backward
branch
from
516c727
to
3bf6fc8
Compare
JZ-LIANG
reviewed
Sep 7, 2023
| std::string alphabet = | ||
| "abcdefghijlmnopqrstuvwxyz"; // k for softmax_normalize axis | ||
| std::string x_axes = GetBroadcastAxes(x_ndim, x_ndim, alphabet); | ||
| x_axes[axis] = 'k'; |
There was a problem hiding this comment.
should check if softmax_normalize is the last axis, if it is not, it could not be sharded.
and if softmax_normalize is the last axis,but if soft_label == true, it could not be sharded.
| ) # loss | ||
|
|
||
| # GPT MP case, shard normalized axis | ||
| # [-1, -1, 0], [-1, -1, -1] (outputs) --> |
There was a problem hiding this comment.
distinguish
outputs: loss (last axis is 1) & softmax_out
inputs: label(last axis maybe 1) & logits
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Outdated suggestions cannot be applied.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.
PR types
Function optimization
PR changes
Others
Description
Pcard-70448
Add infer_backward rule for cross_entropy_with_softmax to infer inputs' dims mappings from outputs'.