Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Update some PyTorch examples for new API #191

Merged
merged 10 commits into from Dec 22, 2022
Merged

Update some PyTorch examples for new API #191

merged 10 commits into from Dec 22, 2022

Conversation

PenghuiCheng
Copy link
Collaborator

Signed-off-by: Cheng, Penghui penghui.cheng@intel.com

Type of Change

Update PyTroch examples(3d_unet, CNN-2, MobileNet_v2)
No API changed

Description

JIRA ticket: https://jira.devtools.intel.com/secure/RapidBoard.jspa?rapidView=32982&projectKey=ILITV&view=detail&selectedIssue=ILITV-2354

@chensuyue
Copy link
Contributor

  1. example readme need to update.
  2. benchmark part need update into Config Class as well.
  3. run_benchmark.sh input parameter update.
  4. json param missing main_script.

Details pls check INC newAPI examples.pptx

@chensuyue chensuyue added this to the v2.0 milestone Nov 30, 2022
@chensuyue
Copy link
Contributor

Pls rebase the PR.

@chensuyue
Copy link
Contributor

extension test
Please fix issues.

@PenghuiCheng
Copy link
Collaborator Author

extension test: https://inteltf-jenk.sh.intel.com/job/intel-lpot-validation-top-mr-extension/3789/

@chensuyue
Copy link
Contributor

extension test: https://inteltf-jenk.sh.intel.com/job/intel-lpot-validation-top-mr-extension/3789/

retest with expend tuning time for 3dunet,bert_large_SQuAD

@chensuyue
Copy link
Contributor

Please check accuracy gap for gpt_j_wikitext.
image

@chensuyue
Copy link
Contributor

extension test: https://inteltf-jenk.sh.intel.com/job/intel-lpot-validation-top-mr-extension/3789/

retest with expend tuning time for 3dunet,bert_large_SQuAD

image
bert_large_SQuAD evaluate too slow.

@PenghuiCheng
Copy link
Collaborator Author

extension test: https://inteltf-jenk.sh.intel.com/job/intel-lpot-validation-top-mr-extension/3817/

@chensuyue
Copy link
Contributor

extension test: https://inteltf-jenk.sh.intel.com/job/intel-lpot-validation-top-mr-extension/3817/

retest the model failed with connect issue: https://inteltf-jenk.sh.intel.com/job/intel-lpot-validation-top-mr-extension/3824/

@chensuyue
Copy link
Contributor

extension test: https://inteltf-jenk.sh.intel.com/job/intel-lpot-validation-top-mr-extension/3817/

retest the model failed with connect issue: https://inteltf-jenk.sh.intel.com/job/intel-lpot-validation-top-mr-extension/3824/

PLease check 3dunet issue.

@chensuyue
Copy link
Contributor

https://inteltf-jenk.sh.intel.com/job/intel-lpot-validation-top-mr-extension/3824/

image
What's wrong with the distilbert_base_ipex tuning result?

@PenghuiCheng
Update some PyTorch examples for new API
a6ad7e1 
Signed-off-by: Cheng, Penghui <penghui.cheng@intel.com>
@PenghuiCheng
Update some PyTorch examples to new API
743fb99 
Signed-off-by: Cheng, Penghui <penghui.cheng@intel.com>
@PenghuiCheng
Change nlp qa pipeline example with new API
30b57cd 
Signed-off-by: Cheng, Penghui <penghui.cheng@intel.com>
@PenghuiCheng
Update examples of PyTorch with new APIs
f1aa252 
Signed-off-by: Cheng, Penghui <penghui.cheng@intel.com>
@PenghuiCheng
Fixed benchmark error and distillation error for nlp question-answering
daa2294 
Signed-off-by: Cheng, Penghui <penghui.cheng@intel.com>
@PenghuiCheng
Update examples test config file
152f4de 
Signed-off-by: Cheng, Penghui <penghui.cheng@intel.com>
@PenghuiCheng
Change benchmark mode to performance
8ddfc72 
Signed-off-by: Cheng, Penghui <penghui.cheng@intel.com>
@PenghuiCheng
Add num_of_instance for examples benchmark
642c2a5 
Signed-off-by: Cheng, Penghui <penghui.cheng@intel.com>
@PenghuiCheng
Fixed examples test error
cd40ec3 
Signed-off-by: Cheng, Penghui <penghui.cheng@intel.com>
@PenghuiCheng
Copy link
Collaborator Author

extension test: https://inteltf-jenk.sh.intel.com/job/intel-lpot-validation-top-mr-extension/3879/

@PenghuiCheng
Update README
225427d 
Signed-off-by: Cheng, Penghui <penghui.cheng@intel.com>
@chensuyue chensuyue merged commit b3e4bf0 into master Dec 22, 2022
@chensuyue chensuyue deleted the penghuic/examples_new_API branch December 22, 2022 06:15
@chensuyue
Copy link
Contributor

Please check accuracy gap for gpt_j_wikitext. image

Have you check this issue, it's still in the latest test.
image

VincyZhang pushed a commit that referenced this pull request Feb 12, 2023
@zhenwei-intel
fix accuracy bug when enabling multi-stream. (#191)
f9a2f6a 
* fix thread pool of multi-stream

* fix race of memory

* update way of initailizing thread pool
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@chensuyue chensuyue chensuyue approved these changes

@xin3he xin3he Awaiting requested review from xin3he

Assignees
No one assigned
Labels
examples extension test
Projects
None yet
Milestone
v2.0
Development

Successfully merging this pull request may close these issues.

None yet
2 participants
@PenghuiCheng @chensuyue