Skip to content
This repository has been archived by the owner on Oct 25, 2024. It is now read-only.

[Optimization] Text-generation support qwen #513

Merged
merged 43 commits into from
Oct 23, 2023
Merged

[Optimization] Text-generation support qwen #513

merged 43 commits into from
Oct 23, 2023

Conversation

changwangss
Copy link
Contributor

@changwangss changwangss commented Oct 20, 2023

Type of Change

Qwen/Qwen-7B, Qwen/Qwen-14B, Qwen/Qwen-7B-Chat, Qwen/Qwen-14B-Chat pass,

optimum: huggingface/optimum#1470

optimum-intel: huggingface/optimum-intel#458

Description

detail description
JIRA ticket: xxx

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

zhewang1-intc and others added 15 commits October 19, 2023 13:22
Signed-off-by: Wenxin Zhang <[email protected]>
Signed-off-by: Wenxin Zhang <[email protected]>
* Support StreamingLLM on CPU

Signed-off-by: zhenwei-intel <[email protected]>
Signed-off-by: changwangss <[email protected]>
Signed-off-by: Wenxin Zhang <[email protected]>
Signed-off-by: Wang, Chang <[email protected]>
Co-authored-by: Wenxin Zhang <[email protected]>
Signed-off-by: changwangss <[email protected]>
* Fix ChatGLM2 model loading issue

Signed-off-by: lvliang-intel <[email protected]>
Signed-off-by: Haihao Shen <[email protected]>
Signed-off-by: lvliang-intel <[email protected]>
Co-authored-by: VincyZhang <[email protected]>
Signed-off-by: changwangss <[email protected]>
* support Memcpy2D

* support gelu fusion

---------

Co-authored-by: luoyu-intel <[email protected]>
xin3he and others added 10 commits October 20, 2023 16:18
* Update README.md to new added docker setup session

Signed-off-by: Louie Tsai <[email protected]>
Signed-off-by: Haihao Shen <[email protected]>
Signed-off-by: Wang, Chang <[email protected]>
Signed-off-by: Haihao Shen <[email protected]>
Signed-off-by: Haihao Shen <[email protected]>
Signed-off-by: Haihao Shen <[email protected]>
Signed-off-by: Haihao Shen <[email protected]>
Signed-off-by: Haihao Shen <[email protected]>
louie-tsai and others added 3 commits October 21, 2023 18:07
convertion -> conversion

Signed-off-by: Ikko Eltociear Ashimine <[email protected]>
Signed-off-by: Haihao Shen <[email protected]>
@VincyZhang
Copy link
Contributor

Unit Test failed with lines coverage decrease -0.064%
Unit Test failed with branches coverage decrease -0.158%

@changwangss
Copy link
Contributor Author

changwangss commented Oct 23, 2023

Unit Test failed with lines coverage decrease -0.064% Unit Test failed with branches coverage decrease -0.158%

yes,it is as expected. qwen doesn't have tiny model to add ut. After qwen is officially included by transformers, the newly added code in generate dummy past-kv func can be deleted, the coverage will improve.
PR is ready, please merge. @VincyZhang

@hshen14 hshen14 changed the title Text-generation support qwen [WIP] Text-generation support qwen Oct 23, 2023
@VincyZhang VincyZhang changed the title [WIP] Text-generation support qwen [Optimize] Text-generation support qwen Oct 23, 2023
@VincyZhang VincyZhang changed the title [Optimize] Text-generation support qwen [Optimization] Text-generation support qwen Oct 23, 2023
@VincyZhang VincyZhang merged commit 8f41d49 into main Oct 23, 2023
15 of 16 checks passed
@VincyZhang VincyZhang deleted the wangchang/qwen branch October 23, 2023 14:50
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.