Qualcomm AI Engine Direct - LLM multi-batch quantization and evaluation#20488
Open
DannyYuyang-quic wants to merge 1 commit into
Open
Qualcomm AI Engine Direct - LLM multi-batch quantization and evaluation#20488DannyYuyang-quic wants to merge 1 commit into
DannyYuyang-quic wants to merge 1 commit into
Conversation
Summary: - Support multi-batch calibration - Support multi-batch evaluation - Add CLI flag to specify batch size during calibration - Fix the shape of attention mask
DannyYuyang-quic
requested review from
abhinaykukkadapu and
psiddh
as code owners
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/20488
Note: Links to docs will display an error until the docs builds have been completed. ❗ 1 Active SEVsThere are 1 currently active SEVs. If your PR is affected, please view them below: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
Bot
added
the
CLA Signed
Contributor
Author
|
@pytorchbot label "release notes: qualcomm" |
pytorch-bot
Bot
added
the
release notes: qualcomm
Contributor
Author
|
@psiddh Hi, this PR mainly adds support for multi-batch quantization and evaluation. cc: @shewu-quic, @haowhsu-quic |
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Details about fix the shape of attention mask (Optimize runtime graph)
The attention mask without the head dimension, the model had to broadcast the shape from [1, 1, S] to [1, 1, 1, S]. This introduced an extra
view_copynode in the runtime graph, which is unnecessary.With head dim, no broadcast needed, redundant
view_copynode removedTest plan