This repository has been archived by the owner on Aug 30, 2024. It is now read-only.
support double quant in BesTLA #60
Merged
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
airMeng
reviewed
Jan 18, 2024
airMeng
approved these changes
Jan 18, 2024
zhewang1-intc
force-pushed
the
double_quant_main
branch
from
757d136 to
579c2e1
Compare
airMeng
reviewed
Jan 18, 2024
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Outdated suggestions cannot be applied.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.
Type of Change
feature or bug fix or documentation or others: feature
API changed or not: Yes
how to enable double-quant:
auto packw=kernel.createStorage(n, k, blocksize, qtype, BTLA_DTYPE::DQ8_BNB);double_quant blocksize will be same as blocksize,
if user want to reset double_quant blocksize just need to call
kernel.setDoubleQuantBlkSize(&packw, BTLA_DTYPE::DQ8_BNB, dq_blksize);Description
detail description
JIRA:https://jira.devtools.intel.com/browse/NLPTOOLKIU-1102
support double-quant feature in BesTLA(scale using dynamic-tree-quantization which def from bisandbytes, for more details pls refer (8-Bit Approximations for Parallelism in Deep Learning)[https://arxiv.org/abs/1511.04561])
Supported Launcher:
LauncerBase(refgetweightimpl),LauncherIntKblock(avx512/avx2getscaleimpl)Expected Behavior & Potential Risk
the expected behavior that triggered by this PR
How has this PR been tested?
how to reproduce the test (including hardware information)
Dependency Change?
any library dependency introduced or removed