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.

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

[OTX]Feature/otx mpa integ #1464

Merged
merged 11 commits into from
Dec 30, 2022
Merged

[OTX]Feature/otx mpa integ #1464

merged 11 commits into from
Dec 30, 2022

Conversation

sungmanc
Copy link
Contributor

@sungmanc sungmanc commented Dec 26, 2022

Description

In the previous development environment, we need to merge MPA and OTX separately. And due to the separate PR process, we met difficulties to test w/the OTX environment, and many MPA issues were merged before checking.

Objective

Integrate the latest MPA into the Feature/OTX

Status

  • Local CLI test: Classification (100% PASS except for OPENVINO export accuracy issue)
    image

  • Local CLI test: Detection (100% PASS)
    image

  • Local CLI test: Instance Segmentation (100% PASS)
    image

  • Local CLI test: Semantic Segmentation (100% PASS)
    image

  • Local Pre-commit test (100% PASS)
    image

  • CI test

@sungmanc sungmanc requested a review from a team as a code owner December 26, 2022 02:28
@github-actions github-actions bot added CLI Any changes in OTE CLI ALGO Any changes in OTX Algo Tasks implementation labels Dec 26, 2022
@harimkang
Copy link
Contributor

That's a good progress!
However, I think this PR needs content to remove mpa from requirements.
Also, how about not checking the pre-commit test for the otx/mpa folder and recipe folder temporarily?
(I know that this can probably be adjusted in tox.ini.)

@sungmanc
Copy link
Contributor Author

However, I think this PR needs content to remove mpa from requirements.

Ah, I missed requirements things ! Thanks, I'll edit

@sungchul2
Copy link
Contributor

sungchul2 commented Dec 26, 2022

I'd like to know if this PR is for permanent or temporary integration.
If it's the former, is there any plan to update modules like adapters and remove unused modules or recipes?

@sungmanc
Copy link
Contributor Author

I'd like to know if this PR is for permanent or temporary integration. If it's the former, is there any plan to update modules like adapters and remove unused modules or recipes?

This is the first step to integrating MPA into the OTX. So, this step is a kind of temporary solution.
We'll remove useless files and folders in the near future

@github-actions github-actions bot added the DEPENDENCY Any changes in any dependencies (new dep or its version) should be produced via Change Request on PM label Dec 26, 2022
@sungmanc sungmanc force-pushed the feature/otx-mpa-integ branch from aa54ba0 to 0bba87d Compare December 28, 2022 01:36
@github-actions github-actions bot added TEST Any changes in tests and removed TEST Any changes in tests labels Dec 28, 2022
sungmanc and others added 11 commits December 29, 2022 17:23
* make logfile saved in save-model-to directory

* enable train

* make main process train also

* bugfix

* refactor multi gpu training

* make all processs have same output path

* prevent child process from being termated by fokred main process

* refactor multigpu implementation

* refactor multi gpu implementation

* modify argument help sentence

* add multi gpu test code

* align with pre-commit test

* separate multi GPU manager class

* modify train cli argument 'save-logs-to' to 'output-path'

* remove tray excpet during killing child process

* apply output_path to all tasks

* change print to logger

* skip multi gpu test if number of gpu is insufficient

* fix typo

* multi gpu test bugfix

* isort fix

* test case bugfix

* fix typo and change some variable name

* [OTX] Apply changes in develop to feature/otx branch (#1436)

* Add tiling module (#1200)

* Update submodule branch (#1222)

* Enhance training schedule for multi-label classification (#1212)

* [CVS-88098] Remove initialize from export functions (#1226)

* Train graph added (#1211)

* Add @attrs decorator for base configs (#1229)

* Pretrained weight download error in MobilenetV3-large-1 of deep-object-reid in SC (#1233)

* [Anomaly Task] Revert hpo template (#1230)

* 🐞 [Anomaly Task] Fix progress bar (#1223)

* [CVS-90555] Fix NaN value in classification (#1244)

* update hpo_config.yaml (#1240)

* [CVS-90400, CVS-91015] NNCF pruning supported tweaks (#1248)

* [Anomaly Task] 🐞 Fix inference when model backbone changes (#1242)

* [CVS-91472] Add pruning_supported value (#1263)

* Pruning supported tweaks (#1256)

* [CVS-90400, CVS-91015] NNCF pruning supported tweaks (#1248)

* Revert "[CVS-90400, CVS-91015] NNCF pruning supported tweaks (#1248)" (#1269)

* [OTE-TEST] Disable obsolete test cases (#1220)

* [OTE-TEST] hot-fix for MPA performance tests (#1273)

* [Anomaly Task] ✨ Upgrade anomalib (#1243)

* Expose early stopping hyper-parameters for all tasks (#1241)

* Resolve pre-commit issues (#1272)

* Remove LazyEarlyStopHook in model_multilabel.py (#1281)

* Removed xfail (#1239)

* Implement IB loss for incremental learning in multi-class classification (#1289)

* Edit num_workers and change MPA repo as a latest (#1314)

* fix annotation bug (#1320)

* Valid POT configs for small HRNet models (#1313)

* Disable NNCF optimization for FP16 models (#1312)

* fliter object less than 1 pixel  (#1305)

* Fix some tests (#1322)

* [Develop] Move drop_last into MPA (#1357)

* Apply changes from releases/v0.3.1-geti1.0.0 (#1337)

* anomaly save_model bugfix (#1300)

* upgrade networkx module version (#1303)

* Forward CVS-94422 size bug fix PR to release branch (#1326)

* Valid POT configs for small HRNet models (#1317)

* [Release branch] Disable NNCF optimization for FP16 models  (#1319)

* [RELEASE] CVS-95549 - Hierarchical classification training failed without obvious reason (#1329)

* Fix h-label: per-group softmax (#1332)

* Fix dataset length bug in mpa task (#1338)

* Fix drop_last key issue for det/set (#1340)

* Hot-fix for OV inference for iseg output (#1345)

* Fix nncf model export bug (#1346)

* Fixed merge error (#1359)

* Update evaluation iou_thr of ins-seg (#1354)

* fix pre-commit test (#1366)

* Fix dataset item tests (#1360)

* Fix OV Inference issues (tiling tests & detection tests) (#1361)

* fix black & add xfail test cases (#1367)

* Update check_nncf_graph. (#1330)

* [Develop] Hot-fix OV inference issue in rotated detection (#1375)

* [Develop] updated documents (#1383)

* [CVS-94911] Fix difference between train and validation normalization pipeline (#1310)

* Update configs for padim model (#1378)

* updated QUICK_START_GUIDE.md (#1397)

* Change ote threshold of openvino test for cls (#1401)

* Normalize top-1 metrics to [0, 1] (#1394)

* Tiling deployment (#1387)

* Replace current saliency map generation with Recipro-CAM for cls (#1363)

* Class-wise saliency map generation for the detection task (#1402)

* Change submodule to develop (#1410)

* Send full dataset to POT optimization function (#1379) & Convert NaN to num to make visible in geti UI (#1413)

* Add active score evaluation to the classification task

* [release/0.4.0][OTX] Enabling GPU execution for exported code (#1416)

* [OTE][Release][XAI] Detection fix two stage bbox_head error (#1414)

* Update SDK commit for exportable code (#1423)

* HRNet-x and HRNe-18--mod2 configs update (#1419)

* [Release] Enable tiling oriented detection for v0.4.0/geti1.1.0 (#1427)

* [OTE][Releases v0.4.0][XAI] Hot-fix for Detection fix two stage error (#1433)

* Temporary MPA branch while dev->otx merge process

* Update doc & install for dev->otx changes

* Update ote_sdk -> otx.api

* Update ote_cli -> otx.cli

* Update external/mmsegmentation -> otx/algorithms/segmentation

* Align saliency map media instantiation over tasks (#1447)

* Update external/d-o-r -> otx/algorithms/classification

* Update external/mmdetection -> otx/algorithms/detection

* Update external/mpa -> otx/algorithms/*

* Fix CLI test run for better error message

* Numpy constraint for deprecated np.bool error

* Capture stderr only

* Align numpy requirement

* [OTX/Anomaly] Add changes from external to otx (#1452)

* Add changes from external to otx

* Address PR comments

* Update config files + remove backbone from base

* Fix pre-merge checks

* Fix pre-commit issues

* Update exportable code commit

* Fix indent error

* Fix flake8 issue

* Resolve softmax issue w/ FIXME for future work

* Add tiling tests

* Revert MPA branch to otx

Signed-off-by: Songki Choi <songki.choi@intel.com>
Co-authored-by: Eugene Liu <eugene.liu@intel.com>
Co-authored-by: Ashwin Vaidya <ashwin.vaidya@intel.com>
Co-authored-by: Jaeguk Hyun <jaeguk.hyun@intel.com>
Co-authored-by: Nikita Savelyev <nikita.savelyev@intel.com>
Co-authored-by: Jihwan Eom <jihwan.eom@intel.com>
Co-authored-by: Harim Kang <harim.kang@intel.com>
Co-authored-by: Soobee Lee <soobee.lee@intel.com>
Co-authored-by: Lee, Soobee <soobeele@intel.com>
Co-authored-by: Emily Chun <emily.chun@intel.com>
Co-authored-by: ljcornel <ludo.cornelissen@intel.com>
Co-authored-by: Eunwoo Shin <eunwoo.shin@intel.com>
Co-authored-by: dlyakhov <daniil.lyakhov@intel.com>
Co-authored-by: kprokofi <kirill.prokofiev@intel.com>
Co-authored-by: Sungman Cho <sungman.cho@intel.com>
Co-authored-by: Yunchu Lee <yunchu.lee@intel.com>
Co-authored-by: Ashwin Vaidya <ashwinitinvaidya@gmail.com>
Co-authored-by: Alexander Dokuchaev <alexander.dokuchaev@intel.com>
Co-authored-by: Vladislav Sovrasov <vladislav.sovrasov@intel.com>
Co-authored-by: Evgeny Tsykunov <e.tsykunov@gmail.com>
Co-authored-by: Galina Zalesskaya <galina.zalesskaya@intel.com>
Co-authored-by: dongkwan-kim <dongkwan.kim@intel.com>

* Apply latest MPA openvinotoolkit/model_preparation_algorithm#105

Signed-off-by: Songki Choi <songki.choi@intel.com>

Signed-off-by: Songki Choi <songki.choi@intel.com>
Co-authored-by: eunwoosh <eunwoo.shin@intel.com>
Co-authored-by: Eugene Liu <eugene.liu@intel.com>
Co-authored-by: Ashwin Vaidya <ashwin.vaidya@intel.com>
Co-authored-by: Jaeguk Hyun <jaeguk.hyun@intel.com>
Co-authored-by: Nikita Savelyev <nikita.savelyev@intel.com>
Co-authored-by: Jihwan Eom <jihwan.eom@intel.com>
Co-authored-by: Harim Kang <harim.kang@intel.com>
Co-authored-by: Soobee Lee <soobee.lee@intel.com>
Co-authored-by: Lee, Soobee <soobeele@intel.com>
Co-authored-by: Emily Chun <emily.chun@intel.com>
Co-authored-by: ljcornel <ludo.cornelissen@intel.com>
Co-authored-by: dlyakhov <daniil.lyakhov@intel.com>
Co-authored-by: kprokofi <kirill.prokofiev@intel.com>
Co-authored-by: Sungman Cho <sungman.cho@intel.com>
Co-authored-by: Yunchu Lee <yunchu.lee@intel.com>
Co-authored-by: Ashwin Vaidya <ashwinitinvaidya@gmail.com>
Co-authored-by: Alexander Dokuchaev <alexander.dokuchaev@intel.com>
Co-authored-by: Vladislav Sovrasov <vladislav.sovrasov@intel.com>
Co-authored-by: Evgeny Tsykunov <e.tsykunov@gmail.com>
Co-authored-by: Galina Zalesskaya <galina.zalesskaya@intel.com>
Co-authored-by: dongkwan-kim <dongkwan.kim@intel.com>
Co-authored-by: Lee, Soobee <soobeele@intel.com>
@sungmanc sungmanc force-pushed the feature/otx-mpa-integ branch from b137066 to 920d657 Compare December 29, 2022 08:36
Copy link
Contributor

@harimkang harimkang left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM!
Thank you for your effort.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
ALGO Any changes in OTX Algo Tasks implementation CLI Any changes in OTE CLI DEPENDENCY Any changes in any dependencies (new dep or its version) should be produced via Change Request on PM
Projects
None yet
Development

Successfully merging this pull request may close these issues.

6 participants