Compare commits

...
Author SHA1 Message Date
Andrew Ferlitsch 45a65f1db8 fix: install issue 2023-03-01 20:05:54 +00:00
Andrew Ferlitsch 240c291728 feat: add eval on versioned model 2023-03-01 19:21:21 +00:00
Andrew Ferlitsch 3c64a8aa58 fix: review nits 2023-03-01 18:04:34 +00:00
Andrew Ferlitsch 5f692ea299 fix: missing installs 2023-03-01 16:06:54 +00:00
Andrew Ferlitsch 0f48628782 feat: notebook for custom evaluations 2023-03-01 00:52:20 +00:00
gericdongandGitHub 906aa91fe7 Merge pull request #1549 from GoogleCloudPlatform/mv_pytorch
fix: reorg
2023-02-28 18:41:15 +00:00
gericdongandGitHub 34e282172f Merge pull request #1548 from GoogleCloudPlatform/rm_pytorch_folder
fix: reorg
2023-02-28 18:29:13 +00:00
Andrew Ferlitsch bddf642b58 fix: reorg 2023-02-28 18:24:55 +00:00
Andrew Ferlitsch 25b89c497f fix: reorg 2023-02-28 18:19:55 +00:00
Andrew FerlitschandGitHub 23a51dbcaa Merge pull request #1526 from GoogleCloudPlatform/doc_tag_1
update tag/linkback #1 AutoML Video
2023-02-28 18:15:56 +00:00
Andrew FerlitschandGitHub 26761198d7 Merge pull request #1547 from GoogleCloudPlatform/autoindex_max_1
fix: web index tune
2023-02-28 17:39:34 +00:00
Andrew Ferlitsch 5b4e88e791 fix: web index tune 2023-02-28 17:29:45 +00:00
Andrew Ferlitsch 83cc796687 fix: workaround for running > 24hrs 2023-02-28 17:05:34 +00:00
Andrew FerlitschandGitHub 3e29cc62c7 Merge pull request #1536 from GoogleCloudPlatform/doc_tag_11
update tag/linkback #11 AutoML Video
2023-02-28 17:02:09 +00:00
Andrew FerlitschandGitHub 2b1bb77f4b Merge pull request #1535 from GoogleCloudPlatform/doc_tag_10
update tag/linkback #10 AutoMLVideo
2023-02-28 16:55:55 +00:00
Andrew Ferlitsch 7144b87f01 fix: workaround for running > 24hrs 2023-02-28 16:50:51 +00:00
Andrew Ferlitsch ebf4d6d8be fix: workaround for running > 24hrs 2023-02-28 16:47:14 +00:00
Andrew FerlitschandGitHub 2ccd913e2e Merge pull request #1528 from GoogleCloudPlatform/doc_tag_3
update tag/linkback #3 AutoML Video
2023-02-28 16:45:02 +00:00
Andrew Ferlitsch d777dd8625 fix: workaround for running > 24hrs 2023-02-28 16:32:16 +00:00
Andrew FerlitschandGitHub d912d3d4d8 Merge pull request #1545 from GoogleCloudPlatform/reznitskii-patch-1
Fixed typo
2023-02-28 02:29:29 +00:00
reznitskiiandGitHub bafe623590 Update automl_tabular_regression_model_evaluation.ipynb 2023-02-27 17:34:34 -05:00
gericdongandGitHub 152077a823 Merge pull request #1543 from GoogleCloudPlatform/guidelines
feat: add authoring guidelines
2023-02-27 20:43:43 +00:00
Andrew Ferlitsch 81c94f9711 feat: add authoring guidelines 2023-02-27 19:53:10 +00:00
Andrew FerlitschandGitHub 897e8e3e47 Merge pull request #1541 from GoogleCloudPlatform/template_linkback
fix: add tag/linkback
2023-02-27 18:06:24 +00:00
Andrew Ferlitsch ddc125da69 fix: smaller dataset 2023-02-27 16:16:13 +00:00
Eric SchmidtandGitHub abb66ade41 Merge pull request #1542 from gericdong/b270683209
fix: bad links in notebook
2023-02-24 17:06:07 +00:00
gericdong 2915824641 Lint 2023-02-24 09:03:19 -05:00
gericdong 44447637bd fixed bad links 2023-02-24 08:59:24 -05:00
Andrew FerlitschandGitHub 6a8345446d Merge pull request #1504 from reznitskii/b267661933-2
Replaced Vertex AI Training linkbacks with Custom training
2023-02-24 02:13:42 +00:00
Andrew Ferlitsch 27a7b8b1da fix: add tag/linkback 2023-02-24 00:22:18 +00:00
Andrew FerlitschandGitHub 04471e7104 Merge pull request #1540 from GoogleCloudPlatform/issue_1522
fix: link
2023-02-23 17:43:06 +00:00
Andrew FerlitschandGitHub e78ea4e805 Merge pull request #1538 from wintwoo/dataproc
Specify Dataproc Serverless Runtime version to use for batch workloads.
2023-02-23 04:14:06 +00:00
Andrew Ferlitsch 7487a14783 fix: link 2023-02-22 23:07:37 +00:00
Andrew FerlitschandGitHub 757d0c5087 change copyright back to 2022. Policy is year is the year first authored 2023-02-22 15:00:49 -08:00
Andrew FerlitschandGitHub 60cb93a9a9 Delete pytorch-text-sentiment-classification-custom-train-deploy.ipynb 2023-02-22 14:43:12 -08:00
Andrew FerlitschandGitHub 0fb72773f4 Merge pull request #1533 from GoogleCloudPlatform/doc_tag_8
update tag/linkback #8
2023-02-22 22:21:16 +00:00
Andrew FerlitschandGitHub 5d2fc74f78 Merge pull request #1532 from GoogleCloudPlatform/doc_tag_7
update tag/linkback #7
2023-02-22 22:20:58 +00:00
Andrew FerlitschandGitHub 1cc26481b2 Merge pull request #1531 from GoogleCloudPlatform/doc_tag_6
update tag/linkback #6
2023-02-22 22:20:50 +00:00
Andrew FerlitschandGitHub 85b0346eab Merge pull request #1530 from GoogleCloudPlatform/doc_tag_5
update tag/linkback #5
2023-02-22 22:20:24 +00:00
Andrew FerlitschandGitHub 910cbacb54 Merge pull request #1529 from GoogleCloudPlatform/doc_tag_4
update tag/linkback #4
2023-02-22 22:19:44 +00:00
Andrew FerlitschandGitHub c7206e3cc8 Merge pull request #1527 from GoogleCloudPlatform/doc_tag_2
update tag/linkback #2
2023-02-22 21:46:40 +00:00
Win Woo 12d721d33e Specify Dataproc runtime versions to use for batch jobs 2023-02-22 02:46:00 +00:00
Andrew FerlitschandGitHub 16df2216d1 Merge pull request #1525 from GoogleCloudPlatform/triton_ensenble_2
feat: triton ensemble
2023-02-21 21:32:16 +00:00
Andrew Ferlitsch fb26b7213c fix: links 2023-02-21 21:30:12 +00:00
Andrew Ferlitsch d90f52e122 fix: links 2023-02-21 19:47:23 +00:00
Andrew Ferlitsch 652f34b814 update tag/linkback 2023-02-21 18:37:09 +00:00
Andrew Ferlitsch c84f8738f8 update tag/linkback 2023-02-21 18:16:13 +00:00
Andrew Ferlitsch 9d72bbf237 update tag/linkback 2023-02-21 18:02:11 +00:00
Andrew Ferlitsch 85df50d06f update tag/linkback 2023-02-21 17:39:51 +00:00
Andrew Ferlitsch 9ab6cddade update tag/linkback 2023-02-21 17:33:56 +00:00
Andrew Ferlitsch 02ffffaeb7 update tag/linkback 2023-02-21 17:30:03 +00:00
Andrew Ferlitsch dfc5635d34 update tag/linkback 2023-02-21 17:25:40 +00:00
Andrew Ferlitsch 90b686cf69 update tag/linkback 2023-02-21 17:02:33 +00:00
Andrew Ferlitsch ad63153778 update tag/linkback 2023-02-21 16:58:56 +00:00
Andrew Ferlitsch 7858f8644e update tag/linkback 2023-02-21 16:55:10 +00:00
Andrew Ferlitsch 3df4098364 update tag/linkback 2023-02-21 16:46:03 +00:00
Andrew Ferlitsch ae3c1877e4 update tag/linkback 2023-02-21 16:41:17 +00:00
Andrew FerlitschandGitHub 2798ab1e02 set timeout 2023-02-21 08:33:40 -08:00
Andrew FerlitschandGitHub 7bd2c567a1 Merge pull request #1490 from GoogleCloudPlatform/imkc--matching-engine-embedding-tweak
Fixed typo in sdk_matching_engine_for_indexing.ipynb
2023-02-18 00:48:17 +00:00
Andrew Ferlitsch 56503cce6a feat: triton ensemble 2023-02-17 22:10:20 +00:00
Andrew FerlitschandGitHub a5a9f53a32 Merge pull request #1515 from junyanxu/add_experimental_info_to_automl_image_montioring_notebook
Add experimental information to automl image classifcation monitoring…
2023-02-17 15:52:07 +00:00
Junyan Xu 469e8436e9 Format the automl image online for model monitoring 2023-02-16 22:37:50 +00:00
Junyan Xu e9f9a5c29b Merge branch 'add_experimental_info_to_automl_image_montioring_notebook' of https://github.com/junyanxu/vertex-ai-samples into add_experimental_info_to_automl_image_montioring_notebook 2023-02-16 17:19:49 +00:00
Junyan Xu 102400a67f update online pip install package 2023-02-16 17:19:00 +00:00
Andrew FerlitschandGitHub 0ae53f47d4 Update get_started_with_model_monitoring_automl_image_batch.ipynb 2023-02-16 08:27:40 -08:00
Eric SchmidtandGitHub ff3e0b4784 Merge pull request #1519 from GoogleCloudPlatform/autoindex_official_6
tune: web index
2023-02-15 21:03:10 +00:00
Andrew Ferlitsch 0f4546c18c tune: web index 2023-02-15 20:56:47 +00:00
Andrew FerlitschandGitHub ee381e15f5 Update get_started_with_model_monitoring_automl_image_batch.ipynb 2023-02-15 12:13:36 -08:00
Andrew FerlitschandGitHub 4ddeeb290d Merge pull request #1295 from Ark-kun/Train_tabular_models
Train tabular models with many frameworks and import to Vertex AI using Pipelines
2023-02-15 19:54:11 +00:00
Andrew FerlitschandGitHub 7e99b07440 Merge branch 'main' into Train_tabular_models 2023-02-15 11:53:17 -08:00
Andrew FerlitschandGitHub bc1ecd6260 Merge pull request #1334 from renovate-bot/renovate/isort-5.x
chore(deps): update dependency isort to v5.12.0
2023-02-15 19:49:21 +00:00
Andrew FerlitschandGitHub f662ffb3e1 Merge pull request #1328 from renovate-bot/renovate/black-22.x
chore(deps): update dependency black to v22.12.0
2023-02-15 19:48:46 +00:00
Andrew FerlitschandGitHub c40c0e5247 Merge pull request #1227 from sudarshan-SpringML/auto_tab_on_vertex_pipeline
Update the file automl_tabular_on_vertex_pipelines
2023-02-15 19:32:19 +00:00
Eric SchmidtandGitHub bd3a0f5af0 Merge pull request #1518 from GoogleCloudPlatform/issue_265061259
fix: issue
2023-02-15 17:46:37 +00:00
Andrew Ferlitsch 99f318333b fix: issue 2023-02-15 17:10:28 +00:00
Andrew FerlitschandGitHub 11e7ba47a4 Merge pull request #1470 from GoogleCloudPlatform/dependabot/pip/community-content/pytorch_image_classification_single_gpu_with_vertex_sdk_and_torchserve/trainer/torch-1.13.1
Build(deps): Bump torch from 1.8.1 to 1.13.1 in /community-content/pytorch_image_classification_single_gpu_with_vertex_sdk_and_torchserve/trainer
2023-02-15 16:54:21 +00:00
Andrew FerlitschandGitHub 399427c2de Merge pull request #1495 from TheMichaelHu/mh-prophet
Reduce cost of running prophet notebook
2023-02-15 15:43:17 +00:00
gericdongandGitHub 0f116ab253 Merge pull request #1517 from GoogleCloudPlatform/stable-diffusion-fixes
chore: revisions to Stable Diffusion and TorchServe nb
2023-02-15 13:33:54 +00:00
Michael Hu 304000d719 use n1-standard-2s 2023-02-14 22:18:13 -05:00
Eric Schmidt 36df462615 chore: revisions to Stable Diffusion and TorchServe nb 2023-02-15 02:32:02 +00:00
Andrew FerlitschandGitHub 909f771bfd Merge pull request #1514 from abcdefgs0324/pytorch_ga
Update wording for pre-built pytorch images on Vertex Prediction.
2023-02-14 17:07:03 +00:00
Eric SchmidtandGitHub e9bef3542d Merge pull request #1510 from GoogleCloudPlatform/stable-diffusion-try2
feat: adds stable diffusion notebook with PyTorch serving
2023-02-13 20:33:24 +00:00
Eric Schmidt 037a3b041a linting 2023-02-13 20:31:22 +00:00
Eric Schmidt f55e6c60cd per reviewer 2023-02-13 18:11:09 +00:00
Eric Schmidt 2cd815c640 light edit 2023-02-10 23:00:11 +00:00
Eric Schmidt b1c0cc9a9d linter 2023-02-10 22:44:33 +00:00
Eric Schmidt af63d1c0e2 Revised notebook to use existing model 2023-02-10 22:40:38 +00:00
Junyan Xu c51a82febb format notebook 2023-02-10 19:15:15 +00:00
Junyan Xu 8e6e4729aa Add experimental information to automl image classifcation monitoring notebook 2023-02-10 18:48:09 +00:00
Eric Schmidt 0e97a83836 revisions 2023-02-10 17:06:33 +00:00
Eric Schmidt f6645e0125 moved notebook 2023-02-10 17:01:13 +00:00
Chun-Hsiang Wang 52385a6071 samples: Updated wording and removed preview email. 2023-02-10 00:43:43 +00:00
Chun-Hsiang WangandGitHub 2365d733c4 Merge branch 'GoogleCloudPlatform:main' into pytorch_ga 2023-02-09 12:55:15 -08:00
Eric Schmidt 71e6423066 iter 2023-02-09 18:06:24 +00:00
Eric Schmidt 139ed95ffc iter 2023-02-09 18:02:07 +00:00
Eric Schmidt cafd192417 deleted notebooks from old location 2023-02-09 17:32:32 +00:00
Eric Schmidt 0d7ec7cd60 iter 2023-02-09 17:21:45 +00:00
Eric Schmidt 7fd934045a moved location of notebook 2023-02-09 17:21:02 +00:00
Max Reznitskii 4e82877269 Reverting changes to failing notebooks 2023-02-09 16:42:44 +00:00
Andrew FerlitschandGitHub 6ef111144d fix: lost updates (#1513) 2023-02-08 18:35:05 -05:00
Eric Schmidt e9b8aa02e9 linter 2023-02-07 14:27:36 -08:00
Eric Schmidt ac1af33c8c feat: adds stable diffusion notebook with PyTorch serving 2023-02-07 21:50:22 +00:00
Andrew FerlitschandGitHub 5bc18b01e4 feat: MM for automl image (#1483)
* feat: MM for automl image

* feat: MM for automl image

* fix: missing import for testing

* fix: testing

* fix: test timing issues

* debug: timing

* test: fix timing issue

* tune: updates from TW for web index

* fix: code review
2023-02-07 14:21:16 -05:00
Andrew FerlitschandGitHub e98b9d6eb4 fix: bad link (#1507) 2023-02-07 10:52:10 -08:00
Andrew FerlitschandGitHub 5e6b8bf597 fix: bad link (#1508) 2023-02-07 10:51:28 -08:00
Andrew FerlitschandGitHub 5b39e7d995 fix: bad link (#1506) 2023-02-07 10:51:08 -08:00
Max Reznitskii c4f08589a9 Reverting changes to notebooks that fail tests 2023-02-03 19:21:48 +00:00
04c6ff4ec7 Fixed AutoML Tabular linkbacks. Linkbacks now refer to specific tabular data tasks. (#1503)
Co-authored-by: Max Reznitskii <reznitskii@google.com>
2023-02-03 10:35:37 -08:00
Max Reznitskii bcb0b19dc4 Replaced Vertex AI Training linkbacks with Custom training 2023-02-02 23:04:08 +00:00
Ivan CheungGitHubivanmkc@google.com <ivanmkc@google.com>
5586fd7c4d Fixed comment about GCS (#1500)
Co-authored-by: ivanmkc@google.com <ivanmkc@google.com>
2023-02-02 14:30:02 -08:00
gericdongandGitHub 30e747b966 correct/remove invalid github usernames (#1502) 2023-02-02 13:24:44 -08:00
Andrew FerlitschandGitHub 080e2b5bb5 fix: missed updates (#1499) 2023-02-01 15:52:22 -05:00
Michael Hu e908774b5b doc updates 2023-01-31 13:23:25 -05:00
Ivan CheungGitHubivanmkc@google.com <ivanmkc@google.com>
60e4416a7e cleanup: remove 3 deprecated notebooks (#1497)
Co-authored-by: ivanmkc@google.com <ivanmkc@google.com>
2023-01-31 00:07:38 -08:00
gericdongandGitHub b207b270b4 feat: enable TensorBoard profiler for custom training with prebuilt container (#1494)
* feat: enable TensorBoard profiler for custom training with prebuilt container

* Fixed package install error

* addressed review comments
2023-01-30 09:31:51 -08:00
Andrew FerlitschandGitHub 497e93aba1 fix: issue 263246858 (#1496) 2023-01-30 12:30:04 -05:00
Ivan CheungGitHubivanmkc@google.com <ivanmkc@google.com>
9a6c36d016 Fixed cleanup code for matching engine index endpoint (#1493)
Co-authored-by: ivanmkc@google.com <ivanmkc@google.com>
2023-01-30 09:52:32 -05:00
Renovate Bot c889a57c9e chore(deps): update dependency isort to v5.12.0 2023-01-28 18:21:42 +00:00
Michael Hu dbe5d61929 Reduce cost of running prophet notebook 2023-01-27 20:18:33 -05:00
Andrew FerlitschandGitHub c180408f41 feat: model monitoring (#1488)
* fix: review

* fix: testing
2023-01-26 00:26:53 -08:00
gericdongandGitHub 869b19d342 Add new notebook to support the XAI zero metadata config feature (#1484)
* feat: add new notebook to support the XAI zero metadata config feature

* add missing packages

* Attempt to fix issue of -- user install not performed in the env

* Fixed package issues

* Addressed review comments

* Addressed review comments

* Addressed review comments
2023-01-25 10:02:50 -05:00
Michael HuandGitHub 3d967b180d add prophet on vertex pipelines notebook (#1320)
* add prophet on vertex pipelines notebook

* update notebook

* add explicit bq dependency

* add more explanations for what the pipeline is doing

* oops

* oops

* Update overview and add parameter descriptions

* foo

* foo

* add more parameters and types

* remove future tense and fix links

* fix formatting

* fix docs
2023-01-24 22:41:19 -08:00
halio-gandGitHub 49710a9225 Improve the training code to support the non-distributed job and add … (#1489)
* Improve the training code to support the non-distributed job and add the dashboard access.

* format the notebook.

* Use the 8888 instead of getting the env since DASHBOARD_PORT is not populated in the pipeline.

* Resolved the pull request comments.
2023-01-24 15:59:02 -08:00
Andrew FerlitschandGitHub 9d31463585 fix: TW updates (#1492) 2023-01-24 18:58:40 -05:00
ivanmkc@google.com bcd86a3707 Fixed typo for index display name 2023-01-25 08:56:59 +09:00
Mend RenovateandGitHub 9335ea3591 chore(deps): update dependency flake8 to v6 (#1298) 2023-01-24 11:31:29 -08:00
Andrew FerlitschandGitHub 0f7343feee migration: experiments (#1487)
* migration: experiments

* fix: review
2023-01-24 12:27:19 -05:00
Ivan NardiniandGitHub 12cd965ce6 new demand forecasting pipeline notebook (#1439)
* new demand forecasting pipeline notebook

* linter passed

* review notebook

* linter passed

* review notebook

* linter passed

* andy review

* linter passed
2023-01-24 08:44:14 -08:00
ivanmkc@google.com 4e5c75fed6 Renamed json to jsonl 2023-01-24 21:08:08 +09:00
Ivan Cheung b28f941abe Fixed typo 2023-01-24 21:02:46 +09:00
32632711ff Add co-hosting model notebook (#713)
* Add notebook for co-hosting model

* Add notebook for co-hosting model

* Change co-hosting model notebook inline link to officical

Co-authored-by: Andrew Ferlitsch <aferlitsch@google.com>
Co-authored-by: Eric Schmidt <em.schmidt78@gmail.com>
2023-01-23 10:29:13 -08:00
junkourataandGitHub d676d87cce feat: Add E2E notebook featuring Vertex Feature Store, Training and Prediction (#1398)
* Add E2E tutorial for Feature Store

* Add Codeowner and fix the formatting and lining.

* Fixed lint
2023-01-23 10:24:51 -08:00
Douglass ChenandGitHub 153a8044b8 Update Colab notebooks' default fields and URLs (#1485)
* Add Cloud natural language pipeline colab notebook

* Add ready-to-go text classification pipeline colab notebook

* Ran reformatting scripts on text classification pipeline colab notebooks

* Update CODEOWNERS files

* Fix order of cells in cloud_natural_language_pipeline.ipynb

* Remove unused variables via linter for text classification colabs; fix classification variable for preprocessing component

* Minor fix: remove GCPC version requirement

* Minor fix: remove outputs

* fix formatting with nbfmt

* move ready-to-go pipeline to notebooks/community

* fix link

* update CODEOWNERS

* move text classification colabs to notebooks/community/pipelines

* Address initial comments on NL notebook

* Remove commented lines in NL notebook

* minor cell formatting

* clear outputs

* minor changes to NL notebook

* address comments for ready-to-go pipeline

* run linter locally

* add pipeline description to NL pipeline

* run linter locally (PR check could not lint)

* Add cell to examine metrics, update kernel restart cell from official template

* lint

* Update default fields and URLs in NL notebook

* Fix URLs in ready to go notebook

* run linter
2023-01-23 10:00:25 -08:00
Andrew FerlitschandGitHub c0d9250416 migration: labeling (#1479) 2023-01-23 08:58:28 -08:00
fd16e39f91 Added notebook demonstrating hyperparameter tuning using tensorboard (#1451)
* Added notebook demonstrating hyperparameter tuning using tensorboard

* Added notebook demonstrating hyperparameter tuning using tensorboard - linter finished

* Added notebook demonstrating hyperparameter tuning using tensorboard - first round of comments resolved

* Added notebook demonstrating hyperparameter tuning using tensorboard - fixing uncomment error

* fixing comment and lint error

* Jack's comments resolved

* fixing the cell that caused CI/CDtest error

* attempt to fix CI/CD issue with loading tensorboard

* attempt to fix TF import error

* fix CI/CD issues

* fix CI/CD issue

Co-authored-by: Andrew Ferlitsch <aferlitsch@gmail.com>
2023-01-19 18:13:02 -08:00
Andrew FerlitschandGitHub 1551ca9435 migration: experiments (#1472)
* migration: experiments

* migration: experiments

* debug: experiments
2023-01-19 12:03:28 -08:00
Andrew FerlitschandGitHub 80fcd2904f migration: bqml (#1475)
* fix: require code review

* migration: BQML
2023-01-18 07:26:53 -08:00
Daniel Elias BecerraandGitHub e974c034ba Matching engine tutorial - add networking troubleshooting and updates to notebook (#1465)
* matching engine tutorial add networking troubleshooting

* format check changes

* Change year 2021 to 2023, replace colab, github and workbench links with new template style

* Replace all occurences of ANN and ANN service with matching_engine or Vertex AI Matching Engine to reflect updated product name

* Update Before you Begin section to follow notebook template and add more organization to it

* Update installation of Vertex AI SDK python library from preview to GA version

* Remove outdated set project id section

* Add Authentication section from notebook template

* Update create bucket section to incorporate notebook template guidelines

* Fix format issues

* Fix format issues

* Fix issues when trying the notebook changes, ordered sections and updated some outdated commands

* Add troubleshooting comment for service networking role for worbench instance to create vpc peering

* Add troubleshooting comment for service networking role for worbench instance to create vpc peering

* Revert "Add troubleshooting comment for service networking role for worbench instance to create vpc peering"

This reverts commit ed418a392a.

* Add wait to deploying index

* Add wait to deploying index

* remove redundant import

* Format file
2023-01-17 09:03:18 -08:00
Aleksey VlasenkoandGitHub d2d4493397 fixed T5x sample links (#1473) 2023-01-13 17:58:33 -08:00
dependabot[bot]andGitHub 6628866130 Build(deps): Bump torch
Bumps [torch](https://github.com/pytorch/pytorch) from 1.8.1 to 1.13.1.
- [Release notes](https://github.com/pytorch/pytorch/releases)
- [Changelog](https://github.com/pytorch/pytorch/blob/master/RELEASE.md)
- [Commits](https://github.com/pytorch/pytorch/compare/v1.8.1...v1.13.1)

---
updated-dependencies:
- dependency-name: torch
  dependency-type: direct:production
...

Signed-off-by: dependabot[bot] <support@github.com>
2023-01-13 17:52:59 +00:00
Andrew FerlitschandGitHub 0cd6146a6c fix: restore requirements.txt (#1468) 2023-01-13 09:52:24 -08:00
Andrew FerlitschandGitHub b61395f465 migration: distributed training (#1466)
* migration: distributed training

* migrate: code review
2023-01-13 09:50:43 -08:00
Andrew FerlitschandGitHub 649b209577 upgrade: revised index (#1463)
* upgrade: revised index

* upgrade: revised index

* upgrade: revised index
2023-01-12 16:29:20 -08:00
Andrew FerlitschandGitHub f42a184171 migration: automl (#1455) 2023-01-12 09:29:25 -08:00
Andrew FerlitschandGitHub 03f0647b76 migration: MM notebook (#1445)
* migration: MM notebook

* migration: fix USER_EMAIL
2023-01-12 09:28:43 -08:00
Andrew FerlitschandGitHub 1f39732ae9 migration: distributed training (#1460) 2023-01-11 22:38:26 -08:00
Andrew FerlitschandGitHub 7dd0b31b58 migration: experiments (#1461) 2023-01-11 18:16:34 -08:00
Andrew FerlitschandGitHub ff843173cf Autoindex official 2 (#1459)
* fix: update linkbacks to vertex pages

* fix: update linkbacks to vertex pages

* fix: update linkbacks to vertex pages
2023-01-11 16:53:28 -08:00
Andrew FerlitschandGitHub da19b116e9 fix: update linkbacks to vertex pages (#1458) 2023-01-11 16:30:10 -08:00
Andrew FerlitschandGitHub d8b365dfd4 fix: update the linkback (#1457) 2023-01-11 15:02:14 -08:00
Andrew FerlitschandGitHub 1247c80fed migration: bqml (#1456) 2023-01-11 14:53:13 -08:00
2cf2bf1080 Adding sample T5x sample for optimized TensorFlow runtime (#1453)
* adding T5x sample

* update for benchmark params

* update for benchmark params

* updated model GCS buckets for optimized TF runtime T5x sample

* added GPU accelerators for deployment pool in Vertex shared VM sample

* final updates for T5x sample

* addressed PR feedback

Co-authored-by: Aleksey Vlasenko <alekseyv@google.com>
2023-01-11 13:32:53 -08:00
Andrew FerlitschandGitHub 5e9e8139c1 Update get_started_with_model_monitoring_xgboost.ipynb 2023-01-11 12:10:46 -08:00
Andrew FerlitschandGitHub 0728a0036f Update get_started_with_model_monitoring_setup.ipynb 2023-01-11 12:10:11 -08:00
Andrew FerlitschandGitHub aa52d21643 Update get_started_with_model_monitoring_custom_tf_serving.ipynb 2023-01-11 12:09:18 -08:00
Andrew FerlitschandGitHub 103888d75e Update get_started_with_model_monitoring_custom.ipynb 2023-01-11 12:08:33 -08:00
Andrew FerlitschandGitHub 13d0d5d9b0 Update get_started_bq_datasets.ipynb 2023-01-11 12:06:47 -08:00
Andrew FerlitschandGitHub 79c6669686 Update get_started_with_data_labeling.ipynb 2023-01-11 12:06:20 -08:00
Andrew FerlitschandGitHub dceb0c4c1c Update get_started_bq_datasets.ipynb 2023-01-11 12:04:31 -08:00
Andrew FerlitschandGitHub 0c9cdca713 migration: MM notebook (#1449)
* migration: MM notebook

* migration: MM notebook
2023-01-10 20:58:19 -08:00
Andrew FerlitschandGitHub 6124092681 migration: MM notebook (#1447)
* migration: MM notebook

* migration: MM notebook
2023-01-10 18:39:45 -08:00
Andrew FerlitschandGitHub 629e739327 migration: MM notebook (#1446)
* migration: MM notebook

* migration: MM notebook
2023-01-10 17:51:00 -08:00
Andrew FerlitschandGitHub b68cbd8255 migration: move to pipelines folder (#1452) 2023-01-10 16:44:17 -08:00
Andrew FerlitschandGitHub 7fec30c12f migration: MM notebook (#1448) 2023-01-10 16:31:14 -08:00
Andrew FerlitschandGitHub a754843c39 migration: MM notebook (#1444)
* migration: MM notebook

* migration: MM notebook

* migration: MM notebook
2023-01-10 15:33:26 -08:00
436db4a35b Fixed documentation links (#1450)
Co-authored-by: Max Reznitskii <reznitskii@google.com>
2023-01-10 15:22:58 -08:00
Ivan NardiniandGitHub 8dba2d040b anomaly detection notebook review (#1440)
* fix some minor issues

* linter passed
2023-01-10 11:50:28 -08:00
Kelsi LakeyandGitHub 560dd9da15 [Community] Added image classification pipeline sample for Ready-to-Go Vertex project (#1404)
* Add image classification pipeline components

* Update CODEOWNERS file with image_ml_model_training

* [Community] Added image classification pipeline sample for Ready-to-Go Vertex project

* Remove unnecessary component download
2023-01-10 11:47:53 -08:00
Nicolas WipfliandGitHub 8ec17d6aca Workaround for shapely (#1397)
Without this workaround, the command "from google.cloud import aiplatform as vertex_ai" fails due to the following issue:

https://github.com/googleapis/python-aiplatform/issues/1852
2023-01-10 11:46:32 -08:00
f545282c36 Cohere demo (#1391)
* Adding Cohere Embedding Demo

* Update cohere_embedding_with_matching_engine.ipynb

* Update cohere_embedding_with_matching_engine.ipynb

* Update CODEOWNERS

* Update CODEOWNERS

* Update CODEOWNERS

* Update cohere_embedding_with_matching_engine.ipynb

* Update cohere_embedding_with_matching_engine.ipynb

Co-authored-by: Andrew Ferlitsch <aferlitsch@google.com>
2023-01-10 11:43:16 -08:00
7e4e14e084 Fixed links to documentation (#1438)
Co-authored-by: Max Reznitskii <reznitskii@google.com>
2023-01-09 15:32:56 -08:00
Douglass ChenandGitHub be5933115d Add Colab notebooks to run text classification model pipelines (#1360)
* Add Cloud natural language pipeline colab notebook

* Add ready-to-go text classification pipeline colab notebook

* Ran reformatting scripts on text classification pipeline colab notebooks

* Update CODEOWNERS files

* Fix order of cells in cloud_natural_language_pipeline.ipynb

* Remove unused variables via linter for text classification colabs; fix classification variable for preprocessing component

* Minor fix: remove GCPC version requirement

* Minor fix: remove outputs

* fix formatting with nbfmt

* move ready-to-go pipeline to notebooks/community

* fix link

* update CODEOWNERS

* move text classification colabs to notebooks/community/pipelines

* Address initial comments on NL notebook

* Remove commented lines in NL notebook

* minor cell formatting

* clear outputs

* minor changes to NL notebook

* address comments for ready-to-go pipeline

* run linter locally

* add pipeline description to NL pipeline

* run linter locally (PR check could not lint)

* Add cell to examine metrics, update kernel restart cell from official template

* lint
2023-01-09 13:12:19 -08:00
reznitskiiandGitHub be46140138 Update README.md (#1433) 2023-01-09 08:59:47 -08:00
Andrew FerlitschandGitHub 79f4dbaafe Autoindex official 2 (#1432)
* fix: alpha sort

* fix: alpha sort

* fix: alpha sort
2023-01-08 12:22:32 -08:00
Andrew FerlitschandGitHub 9577f8324c Autoindex official 2 (#1431)
* fix: alpha sort

* fix: alpha sort
2023-01-08 12:17:06 -08:00
Andrew FerlitschandGitHub 01274fe767 fix: alpha sort (#1430) 2023-01-08 12:13:44 -08:00
Andrew FerlitschandGitHub 1690b07e4a Autoindex official (#1429)
* upgrade: prep for auto docs index

* upgrade: prep for auto docs index

* upgrade: prep work of web index

* upgrade: autoindex, map dirnames to tags

* upgrade: autogen index, folder to tag

* upgrade: autogen index, folder to tag

* upgrade: fine-tune layout for webdoc

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* feat: CL var replacements

* fix: tuning index

* fix: tuning index

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: index tuning

* tuning: linkbak for repo index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* fix: notebook objective

* fix: notebook objective

* fix: alpha sort
2023-01-08 12:04:41 -08:00
Andrew FerlitschandGitHub 6b0de60a5c Autoindex official (#1428)
* upgrade: prep for auto docs index

* upgrade: prep for auto docs index

* upgrade: prep work of web index

* upgrade: autoindex, map dirnames to tags

* upgrade: autogen index, folder to tag

* upgrade: autogen index, folder to tag

* upgrade: fine-tune layout for webdoc

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* feat: CL var replacements

* fix: tuning index

* fix: tuning index

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: index tuning

* tuning: linkbak for repo index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* fix: notebook objective

* fix: notebook objective
2023-01-07 12:53:35 -08:00
Andrew FerlitschandGitHub 19b541b6cb Autoindex official (#1427)
* upgrade: prep for auto docs index

* upgrade: prep for auto docs index

* upgrade: prep work of web index

* upgrade: autoindex, map dirnames to tags

* upgrade: autogen index, folder to tag

* upgrade: autogen index, folder to tag

* upgrade: fine-tune layout for webdoc

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* feat: CL var replacements

* fix: tuning index

* fix: tuning index

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: index tuning

* tuning: linkbak for repo index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* fix: notebook objective
2023-01-07 12:47:03 -08:00
Andrew FerlitschandGitHub 22f6841079 Autoindex official (#1426)
* upgrade: prep for auto docs index

* upgrade: prep for auto docs index

* upgrade: prep work of web index

* upgrade: autoindex, map dirnames to tags

* upgrade: autogen index, folder to tag

* upgrade: autogen index, folder to tag

* upgrade: fine-tune layout for webdoc

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* feat: CL var replacements

* fix: tuning index

* fix: tuning index

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: index tuning

* tuning: linkbak for repo index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index
2023-01-07 12:07:01 -08:00
Andrew FerlitschandGitHub 7c47c95e3a Autoindex official (#1425)
* upgrade: prep for auto docs index

* upgrade: prep for auto docs index

* upgrade: prep work of web index

* upgrade: autoindex, map dirnames to tags

* upgrade: autogen index, folder to tag

* upgrade: autogen index, folder to tag

* upgrade: fine-tune layout for webdoc

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* feat: CL var replacements

* fix: tuning index

* fix: tuning index

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: index tuning

* tuning: linkbak for repo index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index
2023-01-07 11:52:42 -08:00
Andrew FerlitschandGitHub 7735e6ae69 Autoindex official (#1424)
* upgrade: prep for auto docs index

* upgrade: prep for auto docs index

* upgrade: prep work of web index

* upgrade: autoindex, map dirnames to tags

* upgrade: autogen index, folder to tag

* upgrade: autogen index, folder to tag

* upgrade: fine-tune layout for webdoc

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* feat: CL var replacements

* fix: tuning index

* fix: tuning index

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: index tuning

* tuning: linkbak for repo index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index
2023-01-07 11:48:07 -08:00
Andrew FerlitschandGitHub 247a540625 Autoindex official (#1423)
* upgrade: prep for auto docs index

* upgrade: prep for auto docs index

* upgrade: prep work of web index

* upgrade: autoindex, map dirnames to tags

* upgrade: autogen index, folder to tag

* upgrade: autogen index, folder to tag

* upgrade: fine-tune layout for webdoc

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* feat: CL var replacements

* fix: tuning index

* fix: tuning index

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: index tuning

* tuning: linkbak for repo index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index
2023-01-07 11:38:03 -08:00
Andrew FerlitschandGitHub f017606f77 Autoindex official (#1422)
* upgrade: prep for auto docs index

* upgrade: prep for auto docs index

* upgrade: prep work of web index

* upgrade: autoindex, map dirnames to tags

* upgrade: autogen index, folder to tag

* upgrade: autogen index, folder to tag

* upgrade: fine-tune layout for webdoc

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* feat: CL var replacements

* fix: tuning index

* fix: tuning index

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: index tuning

* tuning: linkbak for repo index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index
2023-01-06 18:38:17 -08:00
Andrew FerlitschandGitHub 1d7024341a Autoindex official (#1421)
* upgrade: prep for auto docs index

* upgrade: prep for auto docs index

* upgrade: prep work of web index

* upgrade: autoindex, map dirnames to tags

* upgrade: autogen index, folder to tag

* upgrade: autogen index, folder to tag

* upgrade: fine-tune layout for webdoc

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* feat: CL var replacements

* fix: tuning index

* fix: tuning index

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: index tuning

* tuning: linkbak for repo index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index
2023-01-06 18:33:56 -08:00
Andrew FerlitschandGitHub 9181dba316 Autoindex official (#1420)
* upgrade: prep for auto docs index

* upgrade: prep for auto docs index

* upgrade: prep work of web index

* upgrade: autoindex, map dirnames to tags

* upgrade: autogen index, folder to tag

* upgrade: autogen index, folder to tag

* upgrade: fine-tune layout for webdoc

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* feat: CL var replacements

* fix: tuning index

* fix: tuning index

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: index tuning

* tuning: linkbak for repo index

* tuning: README index

* tuning: README index

* tuning: README index

* tuning: README index
2023-01-06 18:25:13 -08:00
Andrew FerlitschandGitHub a8320f3943 Autoindex official (#1419)
* upgrade: prep for auto docs index

* upgrade: prep for auto docs index

* upgrade: prep work of web index

* upgrade: autoindex, map dirnames to tags

* upgrade: autogen index, folder to tag

* upgrade: autogen index, folder to tag

* upgrade: fine-tune layout for webdoc

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* feat: CL var replacements

* fix: tuning index

* fix: tuning index

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: index tuning

* tuning: linkbak for repo index

* tuning: README index

* tuning: README index
2023-01-06 18:19:45 -08:00
Andrew FerlitschandGitHub 6fc34ae4f1 Autoindex official (#1418)
* upgrade: prep for auto docs index

* upgrade: prep for auto docs index

* upgrade: prep work of web index

* upgrade: autoindex, map dirnames to tags

* upgrade: autogen index, folder to tag

* upgrade: autogen index, folder to tag

* upgrade: fine-tune layout for webdoc

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* feat: CL var replacements

* fix: tuning index

* fix: tuning index

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: index tuning

* tuning: linkbak for repo index

* tuning: README index
2023-01-06 18:13:06 -08:00
Andrew FerlitschandGitHub 5566346fdc Autoindex official (#1417)
* upgrade: prep for auto docs index

* upgrade: prep for auto docs index

* upgrade: prep work of web index

* upgrade: autoindex, map dirnames to tags

* upgrade: autogen index, folder to tag

* upgrade: autogen index, folder to tag

* upgrade: fine-tune layout for webdoc

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* feat: CL var replacements

* fix: tuning index

* fix: tuning index

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: index tuning

* tuning: linkbak for repo index
2023-01-06 17:55:33 -08:00
Andrew FerlitschandGitHub e584acdb48 Autoindex official (#1416)
* upgrade: prep for auto docs index

* upgrade: prep for auto docs index

* upgrade: prep work of web index

* upgrade: autoindex, map dirnames to tags

* upgrade: autogen index, folder to tag

* upgrade: autogen index, folder to tag

* upgrade: fine-tune layout for webdoc

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* feat: CL var replacements

* fix: tuning index

* fix: tuning index

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: index tuning
2023-01-06 15:49:03 -08:00
Andrew FerlitschandGitHub 4022811d3c Autoindex official (#1415)
* upgrade: prep for auto docs index

* upgrade: prep for auto docs index

* upgrade: prep work of web index

* upgrade: autoindex, map dirnames to tags

* upgrade: autogen index, folder to tag

* upgrade: autogen index, folder to tag

* upgrade: fine-tune layout for webdoc

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* feat: CL var replacements

* fix: tuning index

* fix: tuning index

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing
2023-01-06 13:09:11 -08:00
Andrew FerlitschandGitHub 4d29c490f8 Autoindex official (#1414)
* upgrade: prep for auto docs index

* upgrade: prep for auto docs index

* upgrade: prep work of web index

* upgrade: autoindex, map dirnames to tags

* upgrade: autogen index, folder to tag

* upgrade: autogen index, folder to tag

* upgrade: fine-tune layout for webdoc

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* feat: CL var replacements

* fix: tuning index

* fix: tuning index

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing
2023-01-06 12:59:38 -08:00
Andrew FerlitschandGitHub 2315942901 Autoindex official (#1413)
* upgrade: prep for auto docs index

* upgrade: prep for auto docs index

* upgrade: prep work of web index

* upgrade: autoindex, map dirnames to tags

* upgrade: autogen index, folder to tag

* upgrade: autogen index, folder to tag

* upgrade: fine-tune layout for webdoc

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* feat: CL var replacements

* fix: tuning index

* fix: tuning index

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing
2023-01-06 12:22:20 -08:00
Andrew FerlitschandGitHub 505d5049e0 Autoindex official (#1412)
* upgrade: prep for auto docs index

* upgrade: prep for auto docs index

* upgrade: prep work of web index

* upgrade: autoindex, map dirnames to tags

* upgrade: autogen index, folder to tag

* upgrade: autogen index, folder to tag

* upgrade: fine-tune layout for webdoc

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* feat: CL var replacements

* fix: tuning index

* fix: tuning index

* fix: fine tune indexing

* fix: fine tune indexing

* fix: fine tune indexing
2023-01-06 12:18:03 -08:00
Andrew FerlitschandGitHub 20411db737 Delete get_started_bq_datasets.ipynb
duplication
2023-01-06 12:15:56 -08:00
Andrew FerlitschandGitHub 39dbbde22c Autoindex official (#1410)
* upgrade: prep for auto docs index

* upgrade: prep for auto docs index

* upgrade: prep work of web index

* upgrade: autoindex, map dirnames to tags

* upgrade: autogen index, folder to tag

* upgrade: autogen index, folder to tag

* upgrade: fine-tune layout for webdoc

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* feat: CL var replacements

* fix: tuning index

* fix: tuning index
2023-01-06 11:53:14 -08:00
Andrew FerlitschandGitHub 34251594a4 Autoindex official (#1409)
* upgrade: prep for auto docs index

* upgrade: prep for auto docs index

* upgrade: prep work of web index

* upgrade: autoindex, map dirnames to tags

* upgrade: autogen index, folder to tag

* upgrade: autogen index, folder to tag

* upgrade: fine-tune layout for webdoc

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* upgrade: fine-tuning tags and linkbacks

* feat: CL var replacements
2023-01-06 10:57:19 -08:00
7342673331 Add experiments to Dataproc notebook (#1262)
* Show how to use experiments

* Addressed PR feedback

Co-authored-by: Win Woo <wwoo@google.com>
2023-01-05 12:35:42 -08:00
Andrew FerlitschandGitHub b5d19719e0 Update requirements.txt
The <2.11 syntax does not work, since it is interpreted as I/O redirection on the command line.
2023-01-05 12:23:28 -08:00
59536e9e61 Fix google-api-core version to last known working version (#1402)
Co-authored-by: Ivan Cheung <ivanmkc@google.com>
2023-01-05 10:30:48 -08:00
Andrew FerlitschandGitHub 7d74bc3caa workaround: 900 timeout issue (#1400) 2023-01-03 14:56:59 -08:00
wintwooandGitHub f681879a8a Merge branch 'GoogleCloudPlatform:main' into dataproc 2022-12-28 11:23:34 +11:00
Renovate Bot e01174f169 chore(deps): update dependency black to v22.12.0 2022-12-09 16:42:13 +00:00
Win Woo 62c903f47d Addressed PR feedback 2022-12-08 03:01:56 +00:00
Alexey Volkov da77846b88 fix: Fixed the Scikit-learn components 2022-12-06 15:42:27 -08:00
Chun-Hsiang Wang 7c123e68d3 samples: Remove experimental text from Pytorch sample. 2022-11-30 07:46:08 +00:00
Win Woo 9f26e8385d Show how to use experiments 2022-11-18 10:02:02 +00:00
uday kumarandGitHub 0fffaee6dc Merge branch 'main' into auto_tab_on_vertex_pipeline 2022-11-14 15:42:17 +05:30
udaypunna c34b7ab651 linter test 2022-11-14 07:47:02 +00:00
udaypunna d57d9be9d4 downgraded gcpc version 2022-11-14 07:46:18 +00:00
udaypunna 88f3ecf567 downgraded gcpc version 2022-11-14 07:45:12 +00:00
uday kumarandGitHub ecfbeaa8ce Merge branch 'main' into auto_tab_on_vertex_pipeline 2022-11-10 15:46:16 +05:30
udaypunna 3e5a319455 ran linter test 2022-11-10 10:13:43 +00:00
udaypunna f35ce3432a gcpc version downgraded 2022-11-10 10:12:56 +00:00
udaypunna 2e9ff4a941 linter test 2022-11-09 14:40:27 +00:00
udaypunna 61fbd37fa4 service account and version chnages 2022-11-09 14:39:38 +00:00
udaypunna 474e4e602b ran linter test 2022-11-09 14:14:47 +00:00
udaypunna e409cf710b added service account and downgraded gcpc version 2022-11-09 14:14:06 +00:00
udaypunna e9bac09dce ran linter test 2022-11-08 05:40:59 +00:00
udaypunna b799aad08a gcpc version and textual corrections 2022-11-08 05:40:09 +00:00
udaypunna 4d6bd1afff ran linter test 2022-11-07 07:46:14 +00:00
udaypunna b4040ff01d textual corrections 2022-11-07 07:45:25 +00:00
Alexey Volkov c29e4fde26 Train tabular models with many frameworks and import to Vertex AI using Pipelines
These pipelines were previously in community content.
We'd like to move them to the official folder.

These pipelines are:

* Working out of the box (code runs with zero modifications)
* End-to-end (from nothing to a Vertex Model)
* Feature multiple ML frameworks (TensorFlow, PyTorch, XGBoost, Scikit-learn)
* Feature multiple training objectives: tabular classification and tabular regression

The main files are Python-based pipeline code (`pipeline.py`).
2022-11-03 01:53:55 -07:00
udaypunna c6fdad745f linter test 2022-10-18 12:01:03 +00:00
udaypunna 3c60c2cb97 DAG issues 2022-10-18 12:00:18 +00:00
udaypunna 278a014817 ran linter test 2022-10-17 06:36:00 +00:00
udaypunna 89ec43c706 Cloud Storage bucket permission issues resolved 2022-10-17 06:35:26 +00:00
udaypunna ba6be7ee99 Cloud Storage bucket permission issues resolved 2022-10-17 06:30:12 +00:00
149 changed files with 40629 additions and 5930 deletions
@@ -103,7 +103,6 @@ class EndpointResourceCleanupManager(VertexAIResourceCleanupManager):
models.id for models in resource._gca_resource.deployed_models
]:
resource._undeploy(deployed_model_id=deployed_model_id)
resource.delete(force=True)
@@ -117,3 +116,7 @@ class MatchingEngineIndexResourceCleanupManager(VertexAIResourceCleanupManager):
class MatchingEngineIndexEndpointResourceCleanupManager(VertexAIResourceCleanupManager):
vertex_ai_resource = aiplatform.MatchingEngineIndexEndpoint
def delete(self, resource):
resource.undeploy_all()
resource.delete(force=True)
+1 -1
View File
@@ -10,4 +10,4 @@ google-cloud-aiplatform
google-cloud-storage
google-cloud-build
ratemate
GitPython
GitPython
+3 -3
View File
@@ -2,9 +2,9 @@ git+https://github.com/tensorflow/docs
ipython
jupyter
nbconvert
black==22.10.0
black==22.12.0
pyupgrade==2.38.4
isort==5.10.1
flake8==4.0.1
isort==5.12.0
flake8==6.0.0
nbqa==1.5.3
@@ -0,0 +1,62 @@
# python3 -m pip install "kfp<2.0.0" "google-cloud-aiplatform>=1.16.0" --upgrade --quiet
from kfp import components
from kfp.v2 import dsl
# %% Loading components
upload_Tensorflow_model_to_Google_Cloud_Vertex_AI_op = components.load_component_from_url('https://raw.githubusercontent.com/GoogleCloudPlatform/vertex-ai-samples/399405402d95f4a011e2d2e967c96f8508ba5688/community-content/pipeline_components/google-cloud/Vertex_AI/Models/Upload_Tensorflow_model/component.yaml')
deploy_model_to_endpoint_op = components.load_component_from_url('https://raw.githubusercontent.com/GoogleCloudPlatform/vertex-ai-samples/399405402d95f4a011e2d2e967c96f8508ba5688/community-content/pipeline_components/google-cloud/Vertex_AI/Models/Deploy_to_endpoint/component.yaml')
transcode_imagedataset_tfrecord_from_csv_op = components.load_component_from_url('https://raw.githubusercontent.com/GoogleCloudPlatform/vertex-ai-samples/main/community-content/pipeline_components/image_ml_model_training/transcode_tfrecord_image_dataset_from_csv/component.yaml')
load_image_classification_model_from_tfhub_op = components.load_component_from_url('https://raw.githubusercontent.com/GoogleCloudPlatform/vertex-ai-samples/b5b65198a6c2ffe8c0fa2aa70127e3325752df68/community-content/pipeline_components/image_ml_model_training/load_image_classification_model/component.yaml')
preprocess_image_data_op = components.load_component_from_url('https://raw.githubusercontent.com/GoogleCloudPlatform/vertex-ai-samples/main/community-content/pipeline_components/image_ml_model_training/preprocess_image_data/component.yaml')
train_tensorflow_image_classification_model_op = components.load_component_from_url('https://raw.githubusercontent.com/GoogleCloudPlatform/vertex-ai-samples/main/community-content/pipeline_components/image_ml_model_training/train_image_classification_model/component.yaml')
# %% Pipeline definition
def image_classification_pipeline():
class_names = ['daisy', 'dandelion', 'roses', 'sunflowers', 'tulips']
csv_image_data_path = 'gs://cloud-samples-data/ai-platform/flowers/flowers.csv'
deploy_model = False
image_data = dsl.importer(
artifact_uri=csv_image_data_path, artifact_class=dsl.Dataset).output
image_tfrecord_data = transcode_imagedataset_tfrecord_from_csv_op(
csv_image_data_path=image_data,
class_names=class_names
).outputs['tfrecord_image_data_path']
loaded_model_outputs = load_image_classification_model_from_tfhub_op(
class_names=class_names,
).outputs
preprocessed_data = preprocess_image_data_op(
image_tfrecord_data,
height_width_path=loaded_model_outputs['image_size_path'],
).outputs
trained_model = (train_tensorflow_image_classification_model_op(
preprocessed_training_data_path = preprocessed_data['preprocessed_training_data_path'],
preprocessed_validation_data_path = preprocessed_data['preprocessed_validation_data_path'],
model_path=loaded_model_outputs['loaded_model_path']).
set_cpu_limit('96').
set_memory_limit('128G').
add_node_selector_constraint('cloud.google.com/gke-accelerator', 'NVIDIA_TESLA_A100').
set_gpu_limit('8').
outputs['trained_model_path'])
vertex_model_name = upload_Tensorflow_model_to_Google_Cloud_Vertex_AI_op(
model=trained_model,
).outputs['model_name']
# Deploying the model might incur additional costs over time
if deploy_model:
vertex_endpoint_name = deploy_model_to_endpoint_op(
model_name=vertex_model_name,
).outputs['endpoint_name']
pipeline_func = image_classification_pipeline
# %% Pipeline submission
if __name__ == '__main__':
from google.cloud import aiplatform
aiplatform.PipelineJob.from_pipeline_func(pipeline_func=pipeline_func).submit()
@@ -1,3 +1,3 @@
torch==1.8.1
torch==1.13.1
torchvision==0.9.1
tensorboard==2.5.0
@@ -31,17 +31,7 @@
"source": [
"# Deploying a PyTorch Text Classification Model on [Vertex AI](https://cloud.google.com/vertex-ai)\n",
"\n",
"**This is an Experimental release**, covered by the Pre-GA Offerings Terms of your Google Cloud Platform [Terms of Service](https://cloud.google.com/terms).\n",
"\n",
"Experiments are focused on validating a prototype and are not guaranteed to be released. They are not intended for production use or covered by any SLA, support obligation, or deprecation policy and might be subject to backward-incompatible changes.\n",
"\n",
"**Kindly drop us a note before you run any scale tests.**\n",
"\n",
"**Do not hesitate to contact vertexai-prediction-preview-feedback@google.com if you have any questions or run into any issues.**\n",
"\n",
"The usage of the product is free during the Experimental release period: you will still incur charges for other GCP products usage, such as storage.\n",
"\n",
"The projects need to be added to the allowlist in order to deploy PyTorch models using Vertex AI Prediction pre-built PyTorch images. If you are interested in the feature, please send an email to vertexai-prediction-preview-feedback@google.com to provide your project numbers OR project ids."
"**Kindly reach out to Vertex AI before you run any scale tests or you have any questions.**\n"
]
},
{
+4
View File
@@ -30,6 +30,7 @@
/notebooks/community/neo4j/graph_paysim.ipynb @benofben @laeg
/notebooks/community/ml_ops/stage1/get_started_with_visionapi_and_vertex_datasets.ipynb @mansari
/notebooks/community/pipelines/google_cloud_pipeline_components_bqml_pipeline_demand_forecasting.ipynb @inardini
/notebooks/community/cohere/cohere_embedding_with_matching_engine.ipynb @stewart-co
/notebooks/community/ml_ops/stage2/get_started_vertex_hpt_r_kernel.ipynb @fhirschmann
/notebooks/community/ml_ops/stage2/get_started_vertex_training_r_using_r_kernel.ipynb @fhirschmann
/notebooks/community/vertex-ai-samples/notebooks/community/model_registry/vertex_ai_model_registry_bqml_custom_model_versioning.ipynb @inardini
@@ -37,3 +38,6 @@
/notebooks/community/vizier/conversions_vertex_vizier_and_open_source_vizier.ipynb @halio-g
/notebooks/community/experiments/vertex_ai_model_experimentation.ipynb @inardini @asobran
/notebooks/community/pipelines/google_cloud_pipeline_components_bqml_pipeline_anomaly_detection.ipynb @inardini
/notebooks/community/pipelines/google_cloud_pipeline_components_cloud_natural_language_pipeline.ipynb @Narwhalprime
/notebooks/community/pipelines/google_cloud_pipeline_components_ready_to_go_text_classification_pipeline.ipynb @Narwhalprime
/notebooks/community/feature_store/get_started_vertex_feature_store.ipynb @junkourata
+3
View File
@@ -0,0 +1,3 @@
# README
These are notebooks [Cohere](https://cohere.ai/) built in collaboration with Google. They demonstrate how to use Cohere's modeling API along with Vertex AI.
File diff suppressed because it is too large Load Diff
File diff suppressed because it is too large Load Diff
@@ -8,7 +8,7 @@
},
"outputs": [],
"source": [
"# Copyright 2021 Google LLC\n",
"# Copyright 2023 Google LLC\n",
"#\n",
"# Licensed under the Apache License, Version 2.0 (the \"License\");\n",
"# you may not use this file except in compliance with the License.\n",
@@ -24,6 +24,7 @@
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "JAPoU8Sm5E6e"
@@ -32,20 +33,28 @@
"<table align=\"left\">\n",
"\n",
" <td>\n",
" <a href=\"https://console.cloud.google.com/vertex-ai/notebooks/deploy-notebook?download_url=https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/master/notebooks/community/matching_engine/matching_engine_for_indexing.ipynb\">\n",
" Run in Google Cloud Notebooks\n",
" <a href=\"https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/matching_engine/matching_engine_for_indexing.ipynb\">\n",
" <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\" alt=\"Colab logo\">\n",
" Run in Colab\n",
" </a>\n",
" </td>\n",
" <td>\n",
" <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/master/notebooks/community/matching_engine/matching_engine_for_indexing.ipynb\">\n",
" <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/matching_engine/matching_engine_for_indexing.ipynb\">\n",
" <img src=\"https://cloud.google.com/ml-engine/images/github-logo-32px.png\" alt=\"GitHub logo\">\n",
" View on GitHub\n",
" </a>\n",
" </td>\n",
" <td>\n",
" <a href=\"https://console.cloud.google.com/vertex-ai/workbench/deploy-notebook?download_url=https://raw.githubusercontent.com/GoogleCloudPlatform/vertex-ai-samples/main/notebooks/community/matching_engine/matching_engine_for_indexing.ipynb\">\n",
" <img src=\"https://lh3.googleusercontent.com/UiNooY4LUgW_oTvpsNhPpQzsstV5W8F7rYgxgGBD85cWJoLmrOzhVs_ksK_vgx40SHs7jCqkTkCk=e14-rj-sc0xffffff-h130-w32\" alt=\"Vertex AI logo\">\n",
" Open in Vertex AI Workbench\n",
" </a>\n",
" </td> \n",
"</table>"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "tvgnzT1CKxrO"
@@ -53,25 +62,49 @@
"source": [
"## Overview\n",
"\n",
"This example demonstrates how to use the GCP ANN Service. It is a high scale, low latency solution, to find similar vectors (or more specifically \"embeddings\") for a large corpus. Moreover, it is a fully managed offering, further reducing operational overhead. It is built upon [Approximate Nearest Neighbor (ANN) technology](https://ai.googleblog.com/2020/07/announcing-scann-efficient-vector.html) developed by Google Research.\n",
"\n",
"### Dataset\n",
"\n",
"The dataset used for this tutorial is the [GloVe dataset](https://nlp.stanford.edu/projects/glove/).\n",
"\n",
"This example demonstrates how to use Vertex AI Matching Engine. It is a high scale, low latency solution, to find similar vectors (or more specifically \"embeddings\") for a large corpus. Moreover, it is a fully managed offering, further reducing operational overhead. It is built upon [Approximate Nearest Neighbor (ANN) technology](https://ai.googleblog.com/2020/07/announcing-scann-efficient-vector.html) developed by Google Research."
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "56e5f9699c6c"
},
"source": [
"### Objective\n",
"\n",
"In this notebook, you will learn how to create Approximate Nearest Neighbor (ANN) Index, query against indexes, and validate the performance of the index. \n",
"\n",
"The steps performed include:\n",
"\n",
"* Create ANN Index and Brute Force Index\n",
"* Create a Vertex AI Matching Engine Index and Brute Force Index\n",
"* Create an IndexEndpoint with VPC Network\n",
"* Deploy ANN Index and Brute Force Index\n",
"* Perform online query\n",
"* Compute recall\n",
"\n",
"* Deploy a Vertex AI Matching Engine Index and Brute Force Index\n",
"* Perform online queries\n",
"* Submit batch queries\n",
"* Compute recall metric"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "0aaef374550b"
},
"source": [
"### Dataset\n",
"\n",
"The dataset used for this tutorial is the [GloVe dataset](https://nlp.stanford.edu/projects/glove/)."
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "5e2eba58ad71"
},
"source": [
"### Costs \n",
"\n",
"This tutorial uses billable components of Google Cloud:\n",
@@ -87,6 +120,7 @@
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "S5zc4kbEiYCm"
@@ -94,79 +128,47 @@
"source": [
"## Before you begin\n",
"\n",
"* **Prepare a VPC network**. To reduce any network overhead that might lead to unnecessary increase in overhead latency, it is best to call the ANN endpoints from your VPC via a direct [VPC Peering](https://cloud.google.com/vertex-ai/docs/general/vpc-peering) connection. The following section describes how to setup a VPC Peering connection if you don't have one. This is a one-time initial setup task. You can also reuse existing VPC network and skip this section.\n",
"* **WARNING:** The match service gRPC API (to create online queries against your deployed index) has to be executed in a Google Cloud Notebook instance that is created with the following requirements:\n",
" * **In the same region as where your ANN service is deployed** (for example, if you set `REGION = \"us-central1\"` as same as the tutorial, the notebook instance has to be in `us-central1`).\n",
" * **Make sure you select the VPC network you created for ANN service** (instead of using the \"default\" one). That is, you will have to create the VPC network below and then create a new notebook instance that uses that VPC. \n",
" * If you run it in the colab or a Google Cloud Notebook instance in a different VPC network or region, the gRPC API will fail to peer the network (InactiveRPCError)."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "lW2LneA5mmmP"
},
"outputs": [],
"source": [
"PROJECT_ID = \"<your_project_id>\" # @param {type:\"string\"}\n",
"NETWORK_NAME = \"ucaip-haystack-vpc-network\" # @param {type:\"string\"}\n",
"PEERING_RANGE_NAME = \"ucaip-haystack-range\"\n",
"### Set up your Google Cloud project\n",
"\n",
"# Create a VPC network\n",
"! gcloud compute networks create {NETWORK_NAME} --bgp-routing-mode=regional --subnet-mode=auto --project={PROJECT_ID}\n",
"**The following steps are required, regardless of your notebook environment.**\n",
"\n",
"# Add necessary firewall rules\n",
"! gcloud compute firewall-rules create {NETWORK_NAME}-allow-icmp --network {NETWORK_NAME} --priority 65534 --project {PROJECT_ID} --allow icmp\n",
"1. [Select or create a Google Cloud project](https://console.cloud.google.com/cloud-resource-manager). When you first create an account, you get a $300 free credit towards your compute/storage costs.\n",
"\n",
"! gcloud compute firewall-rules create {NETWORK_NAME}-allow-internal --network {NETWORK_NAME} --priority 65534 --project {PROJECT_ID} --allow all --source-ranges 10.128.0.0/9\n",
"2. [Make sure that billing is enabled for your project](https://cloud.google.com/billing/docs/how-to/modify-project).\n",
"\n",
"! gcloud compute firewall-rules create {NETWORK_NAME}-allow-rdp --network {NETWORK_NAME} --priority 65534 --project {PROJECT_ID} --allow tcp:3389\n",
"3. [Enable the Vertex AI API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com).\n",
"\n",
"! gcloud compute firewall-rules create {NETWORK_NAME}-allow-ssh --network {NETWORK_NAME} --priority 65534 --project {PROJECT_ID} --allow tcp:22\n",
"\n",
"# Reserve IP range\n",
"! gcloud compute addresses create {PEERING_RANGE_NAME} --global --prefix-length=16 --network={NETWORK_NAME} --purpose=VPC_PEERING --project={PROJECT_ID} --description=\"peering range for uCAIP Haystack.\"\n",
"\n",
"# Set up peering with service networking\n",
"! gcloud services vpc-peerings connect --service=servicenetworking.googleapis.com --network={NETWORK_NAME} --ranges={PEERING_RANGE_NAME} --project={PROJECT_ID}"
"4. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk)."
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "d3uj8x73nDX_"
},
"source": [
"* Authentication: `$ gcloud auth login` rerun this in Google Cloud Notebook terminal when you are logged out and need the credential again."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "i7EUnXsZhAGF"
"id": "4700b0e39c5d"
},
"source": [
"### Installation\n",
"\n",
"Download and install the latest (preview) version of the Vertex SDK for Python."
"Download and install the latest version of the Vertex AI SDK for Python."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "wyy5Lbnzg5fi"
"id": "014470c6a8de"
},
"outputs": [],
"source": [
"! pip install -U git+https://github.com/googleapis/python-aiplatform.git@main-test --user"
"! pip install -U git+https://github.com/googleapis/python-aiplatform.git@main --user"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "irSMQn6gZ19l"
"id": "cf00462144f7"
},
"source": [
"Install the `h5py` to prepare sample dataset, and the `grpcio-tools` for querying against the index. "
@@ -176,11 +178,15 @@
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "-h5sqwOEZ5Yq"
"id": "3f3e45e5a1d1"
},
"outputs": [],
"source": [
"! pip install -U grpcio-tools --user\n",
"! pip install protobuf==3.20.*\n",
"! pip install -U google-api-python-client==1.8.0 --user\n",
"! pip install -U grpcio-tools==1.47.0 --user\n",
"! pip install -U grpcio==1.47.0 --user\n",
"! pip install -U grpcio-status==1.47.0 --user\n",
"! pip install -U h5py --user"
]
},
@@ -199,7 +205,7 @@
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "EzrelQZ22IZj"
"id": "aa1d87bdc90b"
},
"outputs": [],
"source": [
@@ -215,79 +221,216 @@
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "BF1j6f9HApxa"
"id": "249da91c1011"
},
"source": [
"### Set up your Google Cloud project\n",
"### Set your project ID\n",
"\n",
"**The following steps are required, regardless of your notebook environment.**\n",
"\n",
"1. [Select or create a Google Cloud project](https://console.cloud.google.com/cloud-resource-manager).\n",
"\n",
"1. [Make sure that billing is enabled for your project](https://cloud.google.com/billing/docs/how-to/modify-project).\n",
"\n",
"1. [Enable the Vertex AI API and Compute Engine API, and Service Networking API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com,compute_component,servicenetworking.googleapis.com).\n",
"\n",
"1. Enter your project ID in the cell below. Then run the cell to make sure the\n",
"Cloud SDK uses the right project for all the commands in this notebook.\n",
"\n",
"**Note**: Jupyter runs lines prefixed with `!` as shell commands, and it interpolates Python variables prefixed with `$` into these commands."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "WReHDGG5g0XY"
},
"source": [
"#### Set your project ID\n",
"\n",
"**If you don't know your project ID**, you may be able to get your project ID using `gcloud`."
"**If you don't know your project ID**, try the following:\n",
"* Run `gcloud config list`.\n",
"* Run `gcloud projects list`.\n",
"* See the support page: [Locate the project ID](https://support.google.com/googleapi/answer/7014113)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "oM1iC_MfAts1"
"id": "10e0d2ee8c45"
},
"outputs": [],
"source": [
"import os\n",
"PROJECT_ID = \"[your-project-id]\" # @param {type:\"string\"}\n",
"\n",
"PROJECT_ID = \"\"\n",
"\n",
"# Get your Google Cloud project ID from gcloud\n",
"if not os.getenv(\"IS_TESTING\"):\n",
" shell_output=!gcloud config list --format 'value(core.project)' 2>/dev/null\n",
" PROJECT_ID = shell_output[0]\n",
" print(\"Project ID: \", PROJECT_ID)"
"# Set the project id\n",
"! gcloud config set project {PROJECT_ID}"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "qJYoRfYng0XZ"
"id": "3fbfae3ff12a"
},
"source": [
"Otherwise, set your project ID here."
"### Set the region\n",
"\n",
"You can also change the `REGION` variable used by Vertex AI. Learn more about [Vertex AI regions](https://cloud.google.com/vertex-ai/docs/general/locations).\n",
"* **WARNING:** \n",
" * **Make sure to [choose a region where Vertex AI services are available](https://cloud.google.com/vertex-ai/docs/general/locations#available_regions).**\n",
" * **If you use Vertex Workbench, the Notebook instance needs to be in the same region where your Vertex AI Matching Engine is deployed.** (for example, if you set `REGION = \"us-central1\"` as same as the tutorial, the notebook instance has to be in `us-central1`)."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "riG_qUokg0XZ"
"id": "71c3fd82024e"
},
"outputs": [],
"source": [
"if PROJECT_ID == \"\" or PROJECT_ID is None:\n",
" PROJECT_ID = \"<your_project_id>\" # @param {type:\"string\"}"
"REGION = \"us-central1\" # @param {type: \"string\"}\n",
"\n",
"# Set the regions\n",
"! gcloud config set ai_platform/region {REGION}"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "60c5a0f69ad8"
},
"source": [
"### Authenticate your Google Cloud account\n",
"\n",
"Depending on your Jupyter environment, you may have to manually authenticate. Follow the relevant instructions below."
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "d118c95af93f"
},
"source": [
"**1. Vertex AI Workbench**\n",
"* Do nothing as you are already authenticated."
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "3035286fcdda"
},
"source": [
"**2. Local JupyterLab instance, uncomment and run:**"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "455882ec0f11"
},
"outputs": [],
"source": [
"# ! gcloud auth login"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "5097f3233d53"
},
"source": [
"**3. Colab, uncomment and run:**"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "2b88e46ac2c8"
},
"outputs": [],
"source": [
"# from google.colab import auth\n",
"# auth.authenticate_user()"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "fcdbb8929927"
},
"source": [
"**4. Service account or other**\n",
"* See how to grant Cloud Storage permissions to your service account at https://cloud.google.com/storage/docs/gsutil/commands/iam#ch-examples."
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "7c6eef70dfdb"
},
"source": [
"### Prepare a VPC network\n",
"\n",
"To reduce any network overhead that might lead to unnecessary increase in overhead latency, it is best to call the Vertex AI Matching Engine endpoints from your VPC via a direct [VPC Peering](https://cloud.google.com/vertex-ai/docs/general/vpc-peering) connection. The following section describes how to setup a VPC Peering connection if you don't have one. This is a one-time initial setup task. You can also reuse existing VPC network and skip this section.\n",
"\n",
"* **WARNING:** The match service gRPC API (to create online queries against your deployed index) has to be executed in a Google Cloud Notebook instance that is created with the following requirements:\n",
" * **Make sure you select the VPC network you created for Vertex AI Matching Engine service** (instead of using the \"default\" one). That is, you will have to create the VPC network below and then create a new notebook instance that uses that VPC. \n",
" * If you run it in the colab or a Google Cloud Notebook instance in a different VPC network or region, the gRPC API will fail to peer the network (InactiveRPCError)."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "ab38a8cc634c"
},
"outputs": [],
"source": [
"NETWORK_NAME = \"ucaip-haystack-vpc-network\" # @param {type:\"string\"}\n",
"PEERING_RANGE_NAME = \"ucaip-haystack-range\""
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "ec6bf3199835"
},
"outputs": [],
"source": [
"# Create a VPC network\n",
"! gcloud compute networks create {NETWORK_NAME} --bgp-routing-mode=regional --subnet-mode=auto --project={PROJECT_ID}\n",
"\n",
"# Add necessary firewall rules\n",
"! gcloud compute firewall-rules create {NETWORK_NAME}-allow-icmp --network {NETWORK_NAME} --priority 65534 --project {PROJECT_ID} --allow icmp\n",
"\n",
"! gcloud compute firewall-rules create {NETWORK_NAME}-allow-internal --network {NETWORK_NAME} --priority 65534 --project {PROJECT_ID} --allow all --source-ranges 10.128.0.0/9\n",
"\n",
"! gcloud compute firewall-rules create {NETWORK_NAME}-allow-rdp --network {NETWORK_NAME} --priority 65534 --project {PROJECT_ID} --allow tcp:3389\n",
"\n",
"! gcloud compute firewall-rules create {NETWORK_NAME}-allow-ssh --network {NETWORK_NAME} --priority 65534 --project {PROJECT_ID} --allow tcp:22\n",
"\n",
"# Reserve IP range\n",
"! gcloud compute addresses create {PEERING_RANGE_NAME} --global --prefix-length=16 --network={NETWORK_NAME} --purpose=VPC_PEERING --project={PROJECT_ID} --description=\"peering range for uCAIP Haystack.\""
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "ddbace09fe81"
},
"source": [
"Create the VPC Peering. If you are running this from Vertex AI Workbench it is possible you might need your notebook's instance service or user account to have the Service Networking Admin Role"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "d329aa3c54d3"
},
"outputs": [],
"source": [
"# Set up peering with service networking\n",
"! gcloud services vpc-peerings connect --service=servicenetworking.googleapis.com --network={NETWORK_NAME} --ranges={PEERING_RANGE_NAME} --project={PROJECT_ID}"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "zgPO1eR3CYjk"
@@ -297,13 +440,11 @@
"\n",
"**The following steps are required, regardless of your notebook environment.**\n",
"\n",
"Set the name of your Cloud Storage bucket below. It must be unique across all\n",
"Create a storage bucket to store intermediate artifacts such as datasets. Set the name of your Cloud Storage bucket below. It must be unique across all\n",
"Cloud Storage buckets.\n",
"\n",
"You may also change the `REGION` variable, which is used for operations\n",
"throughout the rest of this notebook. Make sure to [choose a region where Vertex AI services are\n",
"available](https://cloud.google.com/vertex-ai/docs/general/locations#available_regions). You may\n",
"not use a Multi-Regional Storage bucket for training with Vertex AI."
"* **WARNING:** \n",
" * **You may not use a Multi-Regional Storage bucket for training with Vertex AI.**"
]
},
{
@@ -314,8 +455,7 @@
},
"outputs": [],
"source": [
"BUCKET_NAME = \"gs://[your-bucket-name]\" # @param {type:\"string\"}\n",
"REGION = \"us-central1\" # @param {type:\"string\"}"
"BUCKET_NAME = \"gs://[your-bucket-name-unique]\" # @param {type:\"string\"}"
]
},
{
@@ -328,10 +468,14 @@
"source": [
"from datetime import datetime\n",
"\n",
"TIMESTAMP = datetime.now().strftime(\"%Y%m%d%H%M%S\")\n",
"UUID = datetime.now().strftime(\"%Y%m%d%H%M%S\")\n",
"\n",
"if BUCKET_NAME == \"\" or BUCKET_NAME is None or BUCKET_NAME == \"gs://[your-bucket-name]\":\n",
" BUCKET_NAME = \"gs://\" + PROJECT_ID + \"aip-\" + TIMESTAMP"
"if (\n",
" BUCKET_NAME == \"\"\n",
" or BUCKET_NAME is None\n",
" or BUCKET_NAME == \"gs://[your-bucket-name-unique]\"\n",
"):\n",
" BUCKET_NAME = \"gs://\" + PROJECT_ID + \"aip-\" + UUID"
]
},
{
@@ -351,7 +495,7 @@
},
"outputs": [],
"source": [
"! gsutil mb -l $REGION $BUCKET_NAME"
"! gsutil mb -l $REGION -p $PROJECT_ID $BUCKET_NAME"
]
},
{
@@ -416,10 +560,7 @@
},
"outputs": [],
"source": [
"REGION = \"us-central1\"\n",
"ENDPOINT = \"{}-aiplatform.googleapis.com\".format(REGION)\n",
"NETWORK_NAME = \"ucaip-haystack-vpc-network\" # @param {type:\"string\"}\n",
"\n",
"\n",
"AUTH_TOKEN = !gcloud auth print-access-token\n",
"PROJECT_NUMBER = !gcloud projects list --filter=\"PROJECT_ID:'{PROJECT_ID}'\" --format='value(PROJECT_NUMBER)'\n",
@@ -429,10 +570,7 @@
"\n",
"print(\"ENDPOINT: {}\".format(ENDPOINT))\n",
"print(\"PROJECT_ID: {}\".format(PROJECT_ID))\n",
"print(\"REGION: {}\".format(REGION))\n",
"\n",
"!gcloud config set project {PROJECT_ID}\n",
"!gcloud config set ai_platform/region {REGION}"
"print(\"REGION: {}\".format(REGION))"
]
},
{
@@ -523,12 +661,13 @@
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "QuVl8DrWG8NS"
},
"source": [
"Upload the training data to GCS."
"Upload the training data to Google Cloud Storage"
]
},
{
@@ -539,9 +678,9 @@
},
"outputs": [],
"source": [
"# NOTE: Everything in this GCS DIR will be DELETED before uploading the data.\n",
"# NOTE: Everything in this Google Cloud Storage directory will be DELETED before uploading the data\n",
"\n",
"! gsutil rm -rf {BUCKET_NAME}/*"
"! gsutil rm -raf {BUCKET_NAME}/** 2> /dev/null || true"
]
},
{
@@ -567,21 +706,23 @@
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "mglUPwHpJH98"
},
"source": [
"## Create Indexes\n"
"## Create the indexes\n"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "qhIBCQ7dDSbW"
},
"source": [
"### Create ANN Index (for Production Usage)"
"### Create Vertex AI Matching Engine index (for production usage)"
]
},
{
@@ -597,6 +738,16 @@
")"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "14e1ed031d66"
},
"source": [
"Set constants"
]
},
{
"cell_type": "code",
"execution_count": null,
@@ -611,14 +762,15 @@
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "svLYiDf0OD2G"
},
"source": [
"Create the ANN index configuration:\n",
"#### Create the Vertex AI Matching Engine index configuration\n",
"\n",
"Please read the documentation to understand the various configuration parameters that can be used to tune the index\n"
"Please read the [documentation](https://cloud.google.com/vertex-ai/docs/matching-engine/configuring-indexes) to understand the various configuration parameters that can be used to tune the index"
]
},
{
@@ -656,9 +808,9 @@
" }\n",
")\n",
"\n",
"ann_index = {\n",
"matching_engine_index = {\n",
" \"display_name\": DISPLAY_NAME,\n",
" \"description\": \"Glove 100 ANN index\",\n",
" \"description\": \"Glove 100 Vertex AI Matching Engine Index\",\n",
" \"metadata\": struct_pb2.Value(struct_value=metadata),\n",
"}"
]
@@ -671,7 +823,9 @@
},
"outputs": [],
"source": [
"ann_index = index_client.create_index(parent=PARENT, index=ann_index)"
"matching_engine_index = index_client.create_index(\n",
" parent=PARENT, index=matching_engine_index\n",
")"
]
},
{
@@ -686,7 +840,7 @@
"# This will take ~45 min.\n",
"\n",
"while True:\n",
" if ann_index.done():\n",
" if matching_engine_index.done():\n",
" break\n",
" print(\"Poll the operation to create index...\")\n",
" time.sleep(60)"
@@ -700,17 +854,18 @@
},
"outputs": [],
"source": [
"INDEX_RESOURCE_NAME = ann_index.result().name\n",
"INDEX_RESOURCE_NAME = matching_engine_index.result().name\n",
"INDEX_RESOURCE_NAME"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "kSsqZuyoA1SG"
},
"source": [
"### Create Brute Force Index (for Ground Truth)\n",
"### Create brute force index (for ground truth)\n",
"\n",
"The brute force index uses a naive brute force method to find the nearest neighbors. This method is not fast or efficient. Hence brute force indices are not recommended for production usage. They are to be used to find the \"ground truth\" set of neighbors, so that the \"ground truth\" set can be used to measure recall of the indices being tuned for production usage. To ensure an apples to apples comparison, the `distanceMeasureType` and `featureNormType`, `dimensions` of the brute force index should match those of the production indices being tuned.\n",
"\n",
@@ -725,8 +880,6 @@
},
"outputs": [],
"source": [
"from google.protobuf import *\n",
"\n",
"algorithmConfig = struct_pb2.Struct(\n",
" fields={\"bruteForceConfig\": struct_pb2.Value(struct_value=struct_pb2.Struct())}\n",
")\n",
@@ -796,12 +949,13 @@
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "mglUPwHpJH98"
},
"source": [
"## Update Indexes\n",
"## Update the indexes\n",
"\n",
"Create incremental data file.\n"
]
@@ -863,10 +1017,10 @@
" }\n",
")\n",
"\n",
"ann_index = {\n",
"matching_engine_index = {\n",
" \"name\": INDEX_RESOURCE_NAME,\n",
" \"display_name\": DISPLAY_NAME,\n",
" \"description\": \"Glove 100 ANN index\",\n",
" \"description\": \"Glove 100 Vertex AI Matching Engine Index\",\n",
" \"metadata\": struct_pb2.Value(struct_value=metadata),\n",
"}"
]
@@ -879,7 +1033,7 @@
},
"outputs": [],
"source": [
"ann_index = index_client.update_index(index=ann_index)"
"matching_engine_index = index_client.update_index(index=matching_engine_index)"
]
},
{
@@ -894,7 +1048,7 @@
"# This will take ~45 min.\n",
"\n",
"while True:\n",
" if ann_index.done():\n",
" if matching_engine_index.done():\n",
" break\n",
" print(\"Poll the operation to update index...\")\n",
" time.sleep(60)"
@@ -908,17 +1062,18 @@
},
"outputs": [],
"source": [
"INDEX_RESOURCE_NAME = ann_index.result().name\n",
"INDEX_RESOURCE_NAME = matching_engine_index.result().name\n",
"INDEX_RESOURCE_NAME"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "qV2xjAnDDObD"
},
"source": [
"## Create an IndexEndpoint with VPC Network"
"## Create an index endpoint with VPC network"
]
},
{
@@ -997,21 +1152,23 @@
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "np2cgVuuIe9k"
},
"source": [
"## Deploy Indexes"
"## Deploy the indexes"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "8Ew1UgcIIiJG"
},
"source": [
"### Deploy ANN Index"
"### Deploy a Vertex AI Matching Engine index"
]
},
{
@@ -1022,7 +1179,7 @@
},
"outputs": [],
"source": [
"DEPLOYED_INDEX_ID = \"ann_glove_deployed\""
"DEPLOYED_INDEX_ID = \"matching_engine_glove_deployed\""
]
},
{
@@ -1033,13 +1190,23 @@
},
"outputs": [],
"source": [
"deploy_ann_index = {\n",
"deploy_matching_engine_index = {\n",
" \"id\": DEPLOYED_INDEX_ID,\n",
" \"display_name\": DEPLOYED_INDEX_ID,\n",
" \"index\": INDEX_RESOURCE_NAME,\n",
"}"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "cb6d956d7419"
},
"source": [
"If errors occur with the next command wait some minutes for the index endpoint to be created and retry."
]
},
{
"cell_type": "code",
"execution_count": null,
@@ -1049,7 +1216,7 @@
"outputs": [],
"source": [
"r = index_endpoint_client.deploy_index(\n",
" index_endpoint=INDEX_ENDPOINT_NAME, deployed_index=deploy_ann_index\n",
" index_endpoint=INDEX_ENDPOINT_NAME, deployed_index=deploy_matching_engine_index\n",
")"
]
},
@@ -1082,12 +1249,13 @@
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "RNZnXmO5AhDO"
},
"source": [
"### Deploy Brute Force Index"
"### Deploy brute force index"
]
},
{
@@ -1158,12 +1326,13 @@
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "6LCGvBNvBd8D"
},
"source": [
"## Create Online Queries\n",
"## Create online queries\n",
"\n",
"After you built your indexes, you may query against the deployed index through the online querying gRPC API (Match service) within the virtual machine instances from the same region (for example 'us-central1' in this tutorial). \n",
"\n",
@@ -1178,7 +1347,15 @@
"\n",
"* Compile the protocal buffer (see below)\n",
"* Obtain the index endpoint\n",
"* Use a code-generated stub to make the call, passing the parameter values"
"* Use a code-generated stub to make the call, passing the parameter values\n",
"\n",
"### Troubleshooting connectivity issues\n",
"\n",
"In case you have connectivity errors please perform the following:\n",
"\n",
"* Verify that the index endpoint, index, and VPC are all in the same Google Cloud project\n",
"* Verify that the index endpoint, index, and VPC are all in the same region and it is a valid (e.g. us-central1)\n",
"* Verify the Network does not have a firewall rule which denies all egress connections. Else, disable this rule or overwrite it with another rule that allows connection to the index endpoint IP"
]
},
{
@@ -1351,12 +1528,13 @@
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "8wXTSgz1Bl0x"
},
"source": [
"Obtain the Private Endpoint: "
"Obtain the private endpoint: "
]
},
{
@@ -1521,12 +1699,13 @@
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "_mNwdU9_B_Ez"
},
"source": [
"### Batch Query\n",
"## Submit a batch query\n",
"\n",
"You can run multiple queries in a single RPC call using the BatchMatch API:"
]
@@ -1764,18 +1943,20 @@
"]\n",
"\n",
"batch_request = match_service_pb2.BatchMatchRequest()\n",
"batch_request_ann = match_service_pb2.BatchMatchRequest.BatchMatchRequestPerIndex()\n",
"batch_request_matching_engine = (\n",
" match_service_pb2.BatchMatchRequest.BatchMatchRequestPerIndex()\n",
")\n",
"batch_request_brute_force = (\n",
" match_service_pb2.BatchMatchRequest.BatchMatchRequestPerIndex()\n",
")\n",
"batch_request_ann.deployed_index_id = DEPLOYED_INDEX_ID\n",
"batch_request_matching_engine.deployed_index_id = DEPLOYED_INDEX_ID\n",
"batch_request_brute_force.deployed_index_id = DEPLOYED_BRUTE_FORCE_INDEX_ID\n",
"for query in queries:\n",
" batch_request_ann.requests.append(get_request(query, DEPLOYED_INDEX_ID))\n",
" batch_request_matching_engine.requests.append(get_request(query, DEPLOYED_INDEX_ID))\n",
" batch_request_brute_force.requests.append(\n",
" get_request(query, DEPLOYED_BRUTE_FORCE_INDEX_ID)\n",
" )\n",
"batch_request.requests.append(batch_request_ann)\n",
"batch_request.requests.append(batch_request_matching_engine)\n",
"batch_request.requests.append(batch_request_brute_force)\n",
"\n",
"response = stub.BatchMatch(batch_request)\n",
@@ -1783,14 +1964,15 @@
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "_mNwdU9_B_Ez"
},
"source": [
"### Compute Recall\n",
"### Compute the recall metric\n",
"\n",
"Use deployed brute force Index as the ground truth to calculate the recall of ANN Index:"
"Use the deployed brute force index as the ground truth to calculate the recall of the Vertex AI Matching Engine index:"
]
},
{
@@ -1835,6 +2017,7 @@
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "TpV-iwP9qw9c"
@@ -1844,7 +2027,18 @@
"\n",
"To clean up all Google Cloud resources used in this project, you can [delete the Google Cloud\n",
"project](https://cloud.google.com/resource-manager/docs/creating-managing-projects#shutting_down_projects) you used for the tutorial.\n",
"You can also manually delete resources that you created by running the following code."
"\n",
"Otherwise, you can delete the individual resources you created in this tutorial:"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "390c331dc7d9"
},
"source": [
"### Delete the Vertex AI Matching Engine resources"
]
},
{
@@ -1869,6 +2063,31 @@
"source": [
"index_endpoint_client.delete_index_endpoint(name=INDEX_ENDPOINT_NAME)"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "ff14a85c85fb"
},
"source": [
"### Delete the Google Cloud Storage bucket"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "68d4781faac4"
},
"outputs": [],
"source": [
"import os\n",
"\n",
"delete_bucket = False\n",
"if delete_bucket or os.getenv(\"IS_TESTING\"):\n",
" ! gsutil -m rm -r $BUCKET_NAME"
]
}
],
"metadata": {
@@ -1333,7 +1333,6 @@
"Next, you compile the pipeline and then exeute it. The pipeline takes the following parameters, which are passed as the dictionary `parameter_values`:\n",
"\n",
"- `display_name`: A human readable name for the pipeline job.\n",
"- `import_file`: The Cloud Storage location to the dataset.\n",
"- `worker_pool_specs`: The the machine and container, and auto-scaling requirements, as well as command line arguments.\n",
"- `study_spec_metrics`: The metrics to optimize in the study trials.\n",
"- `study_spec_parameters`: The parameters to tune."
@@ -1028,6 +1028,9 @@
"deployment_resource_pool.dedicated_resources.min_replica_count = MIN_NODES\n",
"deployment_resource_pool.dedicated_resources.max_replica_count = MAX_NODES\n",
"deployment_resource_pool.dedicated_resources.machine_spec.machine_type = DEPLOY_COMPUTE\n",
"if DEPLOY_NGPU:\n",
" deployment_resource_pool.dedicated_resources.machine_spec.accelerator_type = DEPLOY_GPU\n",
" deployment_resource_pool.dedicated_resources.machine_spec.accelerator_count = DEPLOY_NGPU\n",
"\n",
"request = aip_beta.CreateDeploymentResourcePoolRequest(\n",
" parent=f\"projects/{PROJECT_ID}/locations/{REGION}\",\n",
@@ -198,7 +198,7 @@
"if IS_WORKBENCH_NOTEBOOK:\n",
" USER_FLAG = \"--user\"\n",
"\n",
"! pip3 install --upgrade tensorflow google-cloud-bigquery google-cloud-aiplatform {USER_FLAG} -q --no-warn-conflicts"
"! pip3 install --upgrade tensorflow google-cloud-bigquery google-cloud-aiplatform \"shapely<2\" {USER_FLAG} -q --no-warn-conflicts"
]
},
{
@@ -72,7 +72,7 @@
"source": [
"## Overview\n",
"\n",
"Anomaly detection is the identification of rare obesrvations which deviate significantly from the data using ML. Anomaly detection can be done in many ways. Supervised, unsupervised, graph-based. It is particularly important for certain industries like telecommunications, manufacturing, and financial services.\n",
"Anomaly detection is the identification of rare observations which deviate significantly from the data using ML. Anomaly detection can be done in many ways. Supervised, unsupervised, graph-based. It is particularly important for certain industries like telecommunications, manufacturing, and financial services.\n",
"\n",
"For instance, in a manufacturing scenario, you may collect some sensor data to predict the number remaining cycles before engine failure (TTF). In this way, you can take actionable decisions about maintenance planning."
]
@@ -397,13 +397,12 @@
"source": [
"import os\n",
"\n",
"SRC_PATH = \"src\"\n",
"KFP_COMPONENTS_PATH = \"components\"\n",
"PIPELINES_PATH = \"pipelines\"\n",
"TRAIN_PIPELINES_PATH = os.path.join(PIPELINES_PATH, \"train_pipelines\")\n",
"TEST_PIPELINES_PATH = os.path.join(PIPELINES_PATH, \"test_pipelines\")\n",
"\n",
"! mkdir -m 777 -p {SRC_PATH} {KFP_COMPONENTS_PATH} {TRAIN_PIPELINES_PATH} {TEST_PIPELINES_PATH}"
"! mkdir -m 777 -p {KFP_COMPONENTS_PATH} {TRAIN_PIPELINES_PATH} {TEST_PIPELINES_PATH}"
]
},
{
@@ -425,14 +424,12 @@
},
"outputs": [],
"source": [
"from urllib.parse import urlparse\n",
"\n",
"PUBLIC_DATA_URI = (\n",
" \" gs://cloud-samples-data/vertex-ai/pipeline-deployment/datasets/turbofan_anomaly\"\n",
" \"gs://cloud-samples-data/vertex-ai/pipeline-deployment/datasets/turbofan_anomaly\"\n",
")\n",
"GCS_TRAIN_URI = urlparse(PUBLIC_DATA_URI)._replace(path=\"train_FD001.csv\").geturl()\n",
"GCS_TEST_URI = urlparse(PUBLIC_DATA_URI)._replace(path=\"test_FD001.csv\").geturl()\n",
"GCS_LABELS_URI = urlparse(PUBLIC_DATA_URI)._replace(path=\"RUL_FD001.csv\").geturl()"
"GCS_TRAIN_URI = f\"{PUBLIC_DATA_URI}/train_FD001.csv\"\n",
"GCS_TEST_URI = f\"{PUBLIC_DATA_URI}/test_FD001.csv\"\n",
"GCS_LABELS_URI = f\"{PUBLIC_DATA_URI}/RUL_FD001.csv\""
]
},
{
@@ -1478,7 +1475,6 @@
"# Remove local resorces\n",
"delete_local_resources = False\n",
"if delete_local_resources:\n",
" ! rm -rf {SRC_PATH}\n",
" ! rm -rf {KFP_COMPONENTS_PATH}\n",
" ! rm -rf {TRAIN_PIPELINES_PATH}\n",
" ! rm -rf {TEST_PIPELINES_PATH}"
@@ -54,18 +54,20 @@
{
"cell_type": "markdown",
"metadata": {
"id": "tvgnzT1CKxrO"
"id": "239ba71252d3"
},
"source": [
"## Overview\n",
"\n",
"This notebook shows how to use `Vertex AI Pipelines` and `BigQuery ML pipeline components` to train and evaluate a demand forecasting model.\n",
"\n",
"### Dataset\n",
"\n",
"The dataset is a modified version of the dataset in [Build and visualize demand forecast predictions using Datastream, Dataflow, BigQuery ML, and Looker\n",
"](https://cloud.google.com/architecture/build-visualize-demand-forecast-prediction-datastream-dataflow-bigqueryml-looker) solution architecture\n",
"\n",
"This notebook shows how to use `Vertex AI Pipelines` and `BigQuery ML pipeline components` to train and evaluate a demand forecasting model."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "25c28706c23e"
},
"source": [
"### Objective\n",
"\n",
"In this tutorial, you learn how to train and evaluate a BigQuery ML model using Vertex AI Pipelines and BigQuery ML pipeline components. \n",
@@ -87,8 +89,27 @@
" - Generate the ARIMA Plus forecasts\n",
" - Generate the ARIMA PLUS forecast explainations\n",
"- Compile the pipeline.\n",
"- Execute the pipeline.\n",
"- Execute the pipeline."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "586acfa9b502"
},
"source": [
"### Dataset\n",
"\n",
"The dataset is a modified version of the dataset in [Build and visualize demand forecast predictions using Datastream, Dataflow, BigQuery ML, and Looker\n",
"](https://cloud.google.com/architecture/build-visualize-demand-forecast-prediction-datastream-dataflow-bigqueryml-looker) solution architecture\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "tvgnzT1CKxrO"
},
"source": [
"### Costs \n",
"\n",
"This tutorial uses billable components of Google Cloud:\n",
@@ -352,9 +373,8 @@
"id": "06571eb4063b"
},
"source": [
"#### Timestamp\n",
"\n",
"If you are in a live tutorial session, you might be using a shared test account or project. To avoid name collisions between users on resources created, you create a timestamp for each instance session, and append it onto the name of resources you create in this tutorial."
"#### UUID\n",
"If you are in a live tutorial session, you might be using a shared test account or project. To avoid name collisions between users on resources created, you create a uuid for each instance session, and append it onto the name of resources you create in this tutorial."
]
},
{
@@ -365,9 +385,16 @@
},
"outputs": [],
"source": [
"from datetime import datetime\n",
"import random\n",
"import string\n",
"\n",
"TIMESTAMP = datetime.now().strftime(\"%Y%m%d%H%M%S\")"
"\n",
"# Generate a uuid of a specifed length(default=8)\n",
"def generate_uuid(length: int = 8) -> str:\n",
" return \"\".join(random.choices(string.ascii_lowercase + string.digits, k=length))\n",
"\n",
"\n",
"UUID = generate_uuid()"
]
},
{
@@ -485,7 +512,7 @@
"outputs": [],
"source": [
"if BUCKET_NAME == \"\" or BUCKET_NAME is None or BUCKET_NAME == \"[your-bucket-name]\":\n",
" BUCKET_NAME = PROJECT_ID + \"-aip-\" + TIMESTAMP\n",
" BUCKET_NAME = PROJECT_ID + \"-aip-\" + UUID\n",
" BUCKET_URI = f\"gs://{BUCKET_NAME}\""
]
},
@@ -706,6 +733,7 @@
"KFP_COMPONENTS_PATH = \"components\"\n",
"PIPELINES_PATH = \"pipelines\"\n",
"\n",
"! mkdir -m 777 -p {DATA_PATH}\n",
"! mkdir -m 777 -p {KFP_COMPONENTS_PATH}\n",
"! mkdir -m 777 -p {PIPELINES_PATH}"
]
@@ -771,7 +799,7 @@
" --location={LOCATION} \\\n",
" --source_format=CSV \\\n",
" --skip_leading_rows=1\\\n",
" fast_fresh.orders_{TIMESTAMP} \\\n",
" fast_fresh.orders_{UUID} \\\n",
" {RAW_DATA_URI} \\\n",
" time_of_sale:DATETIME,order_id:INTEGER,product_name:STRING,price:NUMERIC,quantity:NUMERIC,payment_method:STRING,store_id:INTEGER,user_id:INTEGER"
]
@@ -782,7 +810,7 @@
"id": "ZrgOD30o7HcL"
},
"source": [
"## BQML Training Formalization\n",
"## BigQuery ML Training Formalization\n",
"\n",
"In the next cells, you build the components and pipeline to train and evaluate the BQML demand forecasting model."
]
@@ -820,13 +848,13 @@
"BQ_EVALUATE_MODEL_TABLE_PREFIX = \"orders_arima_model_evaluate\"\n",
"BQ_FORECAST_TABLE_PREFIX = \"orders_arima_forecast\"\n",
"BQ_EXPLAIN_FORECAST_TABLE_PREFIX = \"orders_arima_explain_forecast\"\n",
"BQ_ORDERS_TABLE = f\"{BQ_ORDERS_TABLE_PREFIX}_{TIMESTAMP}\"\n",
"BQ_TRAINING_TABLE = f\"{BQ_TRAINING_TABLE_PREFIX}_{TIMESTAMP}\"\n",
"BQ_MODEL_TABLE = f\"{BQ_MODEL_TABLE_PREFIX}_{TIMESTAMP}\"\n",
"BQ_EVALUATE_TS_TABLE = f\"{BQ_EVALUATE_TS_TABLE_PREFIX}_{TIMESTAMP}\"\n",
"BQ_EVALUATE_MODEL_TABLE = f\"{BQ_EVALUATE_MODEL_TABLE_PREFIX}_{TIMESTAMP}\"\n",
"BQ_FORECAST_TABLE = f\"{BQ_FORECAST_TABLE_PREFIX}_{TIMESTAMP}\"\n",
"BQ_EXPLAIN_FORECAST_TABLE = f\"{BQ_EXPLAIN_FORECAST_TABLE_PREFIX}_{TIMESTAMP}\"\n",
"BQ_ORDERS_TABLE = f\"{BQ_ORDERS_TABLE_PREFIX}_{UUID}\"\n",
"BQ_TRAINING_TABLE = f\"{BQ_TRAINING_TABLE_PREFIX}_{UUID}\"\n",
"BQ_MODEL_TABLE = f\"{BQ_MODEL_TABLE_PREFIX}_{UUID}\"\n",
"BQ_EVALUATE_TS_TABLE = f\"{BQ_EVALUATE_TS_TABLE_PREFIX}_{UUID}\"\n",
"BQ_EVALUATE_MODEL_TABLE = f\"{BQ_EVALUATE_MODEL_TABLE_PREFIX}_{UUID}\"\n",
"BQ_FORECAST_TABLE = f\"{BQ_FORECAST_TABLE_PREFIX}_{UUID}\"\n",
"BQ_EXPLAIN_FORECAST_TABLE = f\"{BQ_EXPLAIN_FORECAST_TABLE_PREFIX}_{UUID}\"\n",
"\n",
"BQ_TRAIN_CONFIGURATION = {\n",
" \"destinationTable\": {\n",
@@ -1022,7 +1050,7 @@
"id": "pcSL1FHk69KT"
},
"source": [
"### Build the BQML training pipeline\n",
"### Build the BigQuery ML training pipeline\n",
"\n",
"Define your workflow using Kubeflow Pipelines DSL package. \n",
"\n",
@@ -1094,8 +1122,8 @@
" location=location,\n",
" ).set_display_name(\"get train data\")\n",
"\n",
" # Train the ARIMA PLUS model\n",
" bq_arima_model_op = (\n",
" # Run an ARIMA PLUS experiment\n",
" bq_arima_model_exp_op = (\n",
" BigqueryCreateModelJobOp(\n",
" query=f\"\"\"\n",
" -- create model table\n",
@@ -1104,10 +1132,7 @@
" MODEL_TYPE = \\'ARIMA_PLUS\\',\n",
" TIME_SERIES_TIMESTAMP_COL = \\'hourly_timestamp\\',\n",
" TIME_SERIES_DATA_COL = \\'total_sold\\',\n",
" TIME_SERIES_ID_COL = [\\'product_name\\'],\n",
" MODEL_REGISTRY = \\'vertex_ai\\',\n",
" VERTEX_AI_MODEL_ID = \\'order_demand_forecasting\\',\n",
" VERTEX_AI_MODEL_VERSION_ALIASES = [\\'staging\\']\n",
" TIME_SERIES_ID_COL = [\\'product_name\\']\n",
" ) AS\n",
" SELECT\n",
" hourly_timestamp,\n",
@@ -1119,7 +1144,7 @@
" project=project,\n",
" location=location,\n",
" )\n",
" .set_display_name(\"train arima plus model\")\n",
" .set_display_name(\"run arima+ model experiment\")\n",
" .after(create_training_dataset_op)\n",
" )\n",
"\n",
@@ -1128,12 +1153,12 @@
" BigqueryMLArimaEvaluateJobOp(\n",
" project=project,\n",
" location=location,\n",
" model=bq_arima_model_op.outputs[\"model\"],\n",
" model=bq_arima_model_exp_op.outputs[\"model\"],\n",
" show_all_candidate_models=False,\n",
" job_configuration_query=bq_evaluate_time_series_configuration,\n",
" )\n",
" .set_display_name(\"evaluate arima plus time series\")\n",
" .after(bq_arima_model_op)\n",
" .after(bq_arima_model_exp_op)\n",
" )\n",
"\n",
" # Evaluate ARIMA Plus model\n",
@@ -1141,12 +1166,12 @@
" BigqueryEvaluateModelJobOp(\n",
" project=project,\n",
" location=location,\n",
" model=bq_arima_model_op.outputs[\"model\"],\n",
" model=bq_arima_model_exp_op.outputs[\"model\"],\n",
" query_statement=f\"\"\"SELECT * FROM `{project}.{bq_dataset}.{bq_training_table}` WHERE split='TEST'\"\"\",\n",
" job_configuration_query=bq_evaluate_model_configuration,\n",
" )\n",
" .set_display_name(\"evaluate arima plus model\")\n",
" .after(bq_arima_model_op)\n",
" .after(bq_arima_model_exp_op)\n",
" )\n",
"\n",
" # Plot model metrics\n",
@@ -1164,6 +1189,34 @@
" < PERF_THRESHOLD,\n",
" name=\"avg. mae good\",\n",
" ):\n",
" # Train the ARIMA PLUS model\n",
" bq_arima_model_op = (\n",
" BigqueryCreateModelJobOp(\n",
" query=f\"\"\"\n",
" -- create model table\n",
" CREATE OR REPLACE MODEL `{project}.{bq_dataset}.{bq_model_table}`\n",
" OPTIONS(\n",
" MODEL_TYPE = \\'ARIMA_PLUS\\',\n",
" TIME_SERIES_TIMESTAMP_COL = \\'hourly_timestamp\\',\n",
" TIME_SERIES_DATA_COL = \\'total_sold\\',\n",
" TIME_SERIES_ID_COL = [\\'product_name\\'],\n",
" MODEL_REGISTRY = \\'vertex_ai\\',\n",
" VERTEX_AI_MODEL_ID = \\'order_demand_forecasting\\',\n",
" VERTEX_AI_MODEL_VERSION_ALIASES = [\\'staging\\']\n",
" ) AS\n",
" SELECT\n",
" DATETIME_TRUNC(time_of_sale, HOUR) as hourly_timestamp,\n",
" product_name,\n",
" SUM(quantity) AS total_sold,\n",
" FROM `{project}.{bq_dataset}.{bq_orders_table}`\n",
" GROUP BY hourly_timestamp, product_name;\n",
" \"\"\",\n",
" project=project,\n",
" location=location,\n",
" )\n",
" .set_display_name(\"train arima+ model\")\n",
" .after(get_evaluation_model_metrics_op)\n",
" )\n",
"\n",
" # Generate the ARIMA PLUS forecasts\n",
" bq_arima_forecast_op = (\n",
@@ -1224,7 +1277,7 @@
"source": [
"### Execute your pipeline\n",
"\n",
"Next, you execute the pipeline. It takes the following parameters which we set as default:\n",
"Next, we execute the pipeline. It takes the following parameters which we set as default:\n",
"\n",
"- `bq_dataset`: The BigQuery dataset to train on.\n",
"- `bq_orders_table` : The BigQuery table of raw data.\n",
@@ -1266,7 +1319,7 @@
"source": [
"### View BigQuery ML training pipeline results\n",
"\n",
"Finally, you will view the artifact outputs of each task in the pipeline."
"Finally, you view the artifact outputs of each task in the pipeline."
]
},
{
@@ -1342,8 +1395,8 @@
"print(\"bigquery-ml-arima-evaluate-job\")\n",
"artifacts = print_pipeline_output(bqml_pipeline, \"bigquery-ml-arima-evaluate-job\")\n",
"print(\"\\n\\n\")\n",
"print(\"get-model-evaluation-metrics\")\n",
"artifacts = print_pipeline_output(bqml_pipeline, \"get-model-evaluation-metrics\")\n",
"print(\"bigquery-evaluate-model-job\")\n",
"artifacts = print_pipeline_output(bqml_pipeline, \"bigquery-evaluate-model-job\")\n",
"print(\"\\n\\n\")\n",
"print(\"bigquery-forecast-model-job\")\n",
"artifacts = print_pipeline_output(bqml_pipeline, \"bigquery-forecast-model-job\")\n",
@@ -1407,7 +1460,8 @@
"\n",
"# Remove local resorces\n",
"! rm -rf {KFP_COMPONENTS_PATH}\n",
"! rm -rf {PIPELINES_PATH}"
"! rm -rf {PIPELINES_PATH}\n",
"! rm -rf {DATA_PATH}"
]
}
],
@@ -0,0 +1,870 @@
{
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "view-in-github"
},
"source": [
"<a href=\"https://colab.research.google.com/github/Narwhalprime/vertex-ai-samples/blob/main/notebooks/community/pipelines/google_cloud_pipeline_components_cloud_natural_language_pipeline.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "1142fd18"
},
"outputs": [],
"source": [
"# Copyright 2022 Google LLC\n",
"#\n",
"# Licensed under the Apache License, Version 2.0 (the \"License\");\n",
"# you may not use this file except in compliance with the License.\n",
"# You may obtain a copy of the License at\n",
"#\n",
"# https://www.apache.org/licenses/LICENSE-2.0\n",
"#\n",
"# Unless required by applicable law or agreed to in writing, software\n",
"# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
"# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
"# See the License for the specific language governing permissions and\n",
"# limitations under the License."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "BwO30Ag12YcB"
},
"source": [
"# Vertex Pipelines: Cloud Natural Language model training pipeline\n",
"<table align=\"left\">\n",
"\n",
" <td>\n",
" <a href=\"https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/pipelines/google_cloud_pipeline_components_cloud_natural_language_pipeline.ipynb\">\n",
" <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\" alt=\"Colab logo\"> Run in Colab\n",
" </a>\n",
" </td>\n",
" <td>\n",
" <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/pipelines/google_cloud_pipeline_components_cloud_natural_language_pipeline.ipynb\">\n",
" <img src=\"https://cloud.google.com/ml-engine/images/github-logo-32px.png\" alt=\"GitHub logo\">\n",
" View on GitHub\n",
" </a>\n",
" </td>\n",
" <td>\n",
" <a href=\"https://console.cloud.google.com/ai/platform/notebooks/deploy-notebook?download_url=https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/natural_language/cloud_natural_language_pipeline.ipynb\">\n",
" <img src=\"https://lh3.googleusercontent.com/UiNooY4LUgW_oTvpsNhPpQzsstV5W8F7rYgxgGBD85cWJoLmrOzhVs_ksK_vgx40SHs7jCqkTkCk=e14-rj-sc0xffffff-h130-w32\" alt=\"Vertex AI logo\">\n",
" Open in Vertex AI Workbench\n",
" </a>\n",
" </td>\n",
"</table>"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "tvgnzT1CKxrO"
},
"source": [
"## Overview\n",
"This notebook shows how to use [Google Cloud Pipeline Components SDK](https://cloud.google.com/vertex-ai/docs/pipelines/components-introduction) and additional components in this directory to run a machine learning pipeline in [Vertex AI Pipelines](https://cloud.google.com/vertex-ai/docs/pipelines/introduction) to train a TensorFlow text classification model.\n",
"\n",
"In this pipeline, the model training Docker image utilizes [TFHub](https://tfhub.dev/) models to perform state-of-the-art text classification training. The image is pre-built and ready to use, so no additional Docker setup is required."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "d975e698c9a4"
},
"source": [
"### Objective\n",
"\n",
"In this tutorial, you learn how to construct an end-to-end training pipeine within Vertex AI pipelines that ingests a dataset, trains a text classification model on it, and outputs evaluation metrics.\n",
"\n",
"This tutorial uses the following Google Cloud ML services and resources:\n",
"\n",
"- Vertex AI Pipelines\n",
"- Vertex AI Datasets\n",
"\n",
"The steps performed include:\n",
"\n",
"- Define Kubeflow pipeline components\n",
"- Setup Kubeflow pipeline\n",
"- Run pipeline on Vertex AI"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "08d289fa873f"
},
"source": [
"## Dataset\n",
"\n",
"This notebook requires that the user has two datasets exported from Vertex AI [managed datasets](https://cloud.google.com/vertex-ai/docs/training/using-managed-datasets): one with train and validation data splits, and the other with test data used for evaluation. Please ensure no data is shared between the two datasets (in particular, no evaluation data should be part of the train or validation splits). To export a Vertex AI dataset, please follow the following public docs:\n",
"* [Preparing data](https://cloud.google.com/vertex-ai/docs/text-data/classification/prepare-data)\n",
"* [Creating a Vertex AI dataset](https://cloud.google.com/vertex-ai/docs/text-data/classification/create-dataset) from the above data\n",
"* [Exporting dataset and its annotations](https://cloud.google.com/vertex-ai/docs/datasets/export-metadata-annotations); ensure the resulting export is located in a Google Cloud Storage (GCS) bucket you own. You may need to manually separate the test split data into its own file."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "aed92deeb4a0"
},
"source": [
"## Costs\n",
"\n",
"This tutorial uses billable components of Google Cloud:\n",
"\n",
"* Vertex AI\n",
"* Cloud Storage\n",
"\n",
"Learn about [Vertex AI\n",
"pricing](https://cloud.google.com/vertex-ai/pricing) and [Cloud Storage\n",
"pricing](https://cloud.google.com/storage/pricing), and use the [Pricing\n",
"Calculator](https://cloud.google.com/products/calculator/)\n",
"to generate a cost estimate based on your projected usage."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "setup_local"
},
"source": [
"## Setup\n",
"\n",
"If you are using Colab or Google Vertex AI Workbench Notebooks, your environment already meets all the requirements to run this notebook. You can skip this step.\n",
"\n",
"***NOTE***: This notebook has been tested in the following environment:\n",
"\n",
"* Python version = 3.8\n",
"\n",
"Otherwise, make sure your environment meets this notebook's requirements. You need the following:\n",
"\n",
"- The Cloud Storage SDK\n",
"- Python 3\n",
"- virtualenv\n",
"- Jupyter notebook running in a virtual environment with Python 3\n",
"\n",
"The Cloud Storage guide to [Setting up a Python development environment](https://cloud.google.com/python/setup) and the [Jupyter installation guide](https://jupyter.org/install) provide detailed instructions for meeting these requirements. The following steps provide a condensed set of instructions:\n",
"\n",
"1. [Install and initialize the SDK](https://cloud.google.com/sdk/docs/).\n",
"\n",
"2. [Install Python 3](https://cloud.google.com/python/setup#installing_python).\n",
"\n",
"3. [Install virtualenv](https://cloud.google.com/python/setup#installing_and_using_virtualenv) and create a virtual environment that uses Python 3. Activate the virtual environment.\n",
"\n",
"4. Activate that environment and run `pip3 install Jupyter` in a terminal shell to install Jupyter.\n",
"\n",
"5. Run `jupyter notebook` on the command line in a terminal shell to launch Jupyter.\n",
"\n",
"6. Open this notebook in the Jupyter Notebook Dashboard.\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "568d5c16"
},
"source": [
"### Install additional packages\n",
"\n",
"Run the following commands to setup the packages for this notebook. Note that the last code snippet in this section restarts your kernel in order to load the installs properly, so when initalizing this notebook from scratch, it is recommended to run up to that cell, then afterwards you may start running the cell after that."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "dac98aac"
},
"outputs": [],
"source": [
"# Install using pip3\n",
"!pip3 install -U tensorflow google-cloud-pipeline-components google-cloud-aiplatform kfp==1.8.16 \"shapely<2\" -q"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "alRWYgYTdz7P"
},
"outputs": [],
"source": [
"# Version check\n",
"# This has been tested with KFP 1.8.16\n",
"! python3 -c \"import kfp; print('KFP SDK version: {}'.format(kfp.__version__))\"\n",
"! python3 -c \"import google_cloud_pipeline_components; print('google_cloud_pipeline_components version: {}'.format(google_cloud_pipeline_components.__version__))\""
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "d0a15440"
},
"outputs": [],
"source": [
"import os\n",
"\n",
"if not os.getenv(\"IS_TESTING\"):\n",
" # Automatically restart kernel after installs\n",
" import IPython\n",
"\n",
" app = IPython.Application.instance()\n",
" app.kernel.do_shutdown(True)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "B9IYalYObAbY"
},
"source": [
"## Before you begin\n",
"\n",
"### Set up your Google Cloud project\n",
"\n",
"**The following steps are required, regardless of your notebook environment.**\n",
"\n",
"1. [Select or create a Google Cloud project](https://console.cloud.google.com/cloud-resource-manager). When you first create an account, you get a $300 free credit towards your compute/storage costs.\n",
"\n",
"2. [Make sure that billing is enabled for your project](https://cloud.google.com/billing/docs/how-to/modify-project).\n",
"\n",
"3. [Enable the Vertex AI API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com,storage.googleapis.com).\n",
"\n",
"4. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk)."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "VA_kzAIIj2G_"
},
"source": [
"### Authenticate your Google Cloud account\n",
"\n",
"**If you are using Vertex AI Workbench Notebooks**, your environment is already\n",
"authenticated. Skip this step.\n",
"\n",
"**If you are using Colab**, run the cell below and follow the instructions\n",
"when prompted to authenticate your account via oAuth.\n",
"\n",
"**Otherwise**, follow these steps:\n",
"\n",
"1. In the Cloud Console, go to the [**Create service account key**\n",
" page](https://console.cloud.google.com/apis/credentials/serviceaccountkey).\n",
"\n",
"2. Click **Create service account**.\n",
"\n",
"3. In the **Service account name** field, enter a name, and\n",
" click **Create**.\n",
"\n",
"4. In the **Grant this service account access to project** section, click the **Role** drop-down list. Type \"Vertex AI\"\n",
"into the filter box, and select\n",
" **Vertex AI Administrator**. Type \"Storage Object Admin\" into the filter box, and select **Storage Object Admin**.\n",
"\n",
"5. Click *Create*. A JSON file that contains your key downloads to your\n",
"local environment.\n",
"\n",
"6. Enter the path to your service account key as the\n",
"`GOOGLE_APPLICATION_CREDENTIALS` variable in the cell below and run the cell."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "PyQmSRbKA8r-"
},
"outputs": [],
"source": [
"# If you are running this notebook in Colab, run this cell and follow the\n",
"# instructions to authenticate your GCP account. This provides access to your\n",
"# Cloud Storage bucket and lets you submit training jobs and prediction\n",
"# requests.\n",
"\n",
"import os\n",
"import sys\n",
"\n",
"# If on Vertex AI Workbench, then don't execute this code\n",
"IS_COLAB = \"google.colab\" in sys.modules\n",
"if not os.path.exists(\"/opt/deeplearning/metadata/env_version\") and not os.getenv(\n",
" \"DL_ANACONDA_HOME\"\n",
"):\n",
" if \"google.colab\" in sys.modules:\n",
" from google.colab import auth as google_auth\n",
"\n",
" google_auth.authenticate_user()\n",
"\n",
" # If you are running this notebook locally, replace the string below with the\n",
" # path to your service account key and run this cell to authenticate your GCP\n",
" # account.\n",
" elif not os.getenv(\"IS_TESTING\"):\n",
" %env GOOGLE_APPLICATION_CREDENTIALS ''"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "set_service_account"
},
"source": [
"### Set project ID\n",
"\n",
"Set your project ID here. If you don't know this, the following snippet attempts to deterine this from your gcloud config. Please continue only if the notebook can see your desired project."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "AkqEd5Gin9mn"
},
"outputs": [],
"source": [
"PROJECT_ID = \"your-project-id\" # @param {type:\"string\"}\n",
"if PROJECT_ID == \"\" or PROJECT_ID is None or PROJECT_ID == \"[your-project-id]\":\n",
" # Get your GCP project id from gcloud\n",
" shell_output = !gcloud config list --format 'value(core.project)' 2>/dev/null\n",
" PROJECT_ID = shell_output[0]\n",
"print(\"Project ID:\", PROJECT_ID)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "OVO_gUqpFEP2"
},
"outputs": [],
"source": [
"!gcloud config set project $PROJECT_ID"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "a27d4cee"
},
"source": [
"### Setup project information\n",
"\n",
"Enter information about your project and datasets here."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "7e9477a2"
},
"outputs": [],
"source": [
"REGION = \"us\" # @param {type:\"string\"}\n",
"LOCATION = \"us-central1\" # @param {type:\"string\"}\n",
"TRAINING_DATA_LOCATION = \"gs://your-training-data-location\" # @param {type:\"string\"}\n",
"TASK_TYPE = \"CLASSIFICATION\" # @param [\"CLASSIFICATION\", \"MULTILABEL_CLASSIFICATION\"]"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "o-MZnHsimbOH"
},
"outputs": [],
"source": [
"# Since we are training a custom model, we need to specify the list of possible\n",
"# classes/labels.\n",
"# e.g, [\"FirstClass\", \"SecondClass\"]\n",
"# An additional class \"[UNK]\" will be added to the list indicating that none of\n",
"# the specified labels are a match.\n",
"CLASS_NAMES = [\"\"]\n",
"\n",
"# This is a list of GCS URIs; e.g., [\"gs://your-bucket-name-here/your-input-file.jsonl\"].\n",
"TEST_DATA_URIS = [\"gs://your-bucket-name-here/your-input-file.jsonl\"]"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "timestamp"
},
"source": [
"#### UUID\n",
"\n",
"To avoid name collisions with other resources in your project, you can create a UUID with the code below and append it onto the name of the bucket(s) created in this notebook."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "wh9sgzemwLXE"
},
"outputs": [],
"source": [
"import random\n",
"import string\n",
"\n",
"\n",
"# Generate a uuid of a specifed length(default=8)\n",
"def generate_uuid(length: int = 8) -> str:\n",
" return \"\".join(random.choices(string.ascii_lowercase + string.digits, k=length))\n",
"\n",
"\n",
"UUID = generate_uuid()"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "bucket:mbsdk"
},
"source": [
"### Create a Cloud Storage bucket\n",
"\n",
"**The following steps are required, regardless of your notebook environment.**\n",
"\n",
"When you initialize the Vertex AI SDK for Python, you specify a Cloud Storage staging bucket. The staging bucket is where all the data associated with your dataset and model resources are retained across sessions.\n",
"\n",
"Set the name of your Cloud Storage bucket below. Bucket names must be globally unique across all Google Cloud projects, including those outside of your organization."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "bucket"
},
"outputs": [],
"source": [
"BUCKET_NAME = \"[your-bucket-name]\" # @param {type:\"string\"}\n",
"BUCKET_URI = f\"gs://{BUCKET_NAME}\""
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "autoset_bucket"
},
"outputs": [],
"source": [
"if BUCKET_NAME == \"\" or BUCKET_NAME is None or BUCKET_NAME == \"[your-bucket-name]\":\n",
" BUCKET_NAME = PROJECT_ID + \"aip-\" + UUID\n",
" BUCKET_URI = \"gs://\" + BUCKET_NAME"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "create_bucket"
},
"source": [
"**Only if your bucket doesn't already exist**: Run the following cell to create your Cloud Storage bucket."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "dO0NV93IwLXF"
},
"outputs": [],
"source": [
"!gsutil mb -l $REGION $BUCKET_URI"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "validate_bucket"
},
"source": [
"Finally, validate access to your Cloud Storage bucket by examining its contents:"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "Hg5f2oKBwLXG"
},
"outputs": [],
"source": [
"!gsutil ls -al $BUCKET_URI"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "EuFETRptyKXc"
},
"outputs": [],
"source": [
"from google.cloud import aiplatform\n",
"\n",
"aiplatform.init(project=PROJECT_ID, staging_bucket=BUCKET_URI)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "f3a09765"
},
"source": [
"## Create training pipeline"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "89bb4a50"
},
"source": [
"### Import libraries"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "0f361e65"
},
"outputs": [],
"source": [
"from google_cloud_pipeline_components.aiplatform import ModelBatchPredictOp\n",
"from google_cloud_pipeline_components.experimental import natural_language\n",
"from google_cloud_pipeline_components.experimental.evaluation import (\n",
" GetVertexModelOp, ModelEvaluationClassificationOp,\n",
" TargetFieldDataRemoverOp)\n",
"from kfp import components\n",
"from kfp.v2 import compiler, dsl"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "d33c87e4-2ada-4b87-bf75-064247f3162d"
},
"source": [
"### Define constants"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "36ceb9f8"
},
"outputs": [],
"source": [
"# Worker pool specs\n",
"TRAINING_MACHINE_TYPE = \"n1-highmem-8\"\n",
"ACCELERATOR_TYPE = \"NVIDIA_TESLA_T4\"\n",
"ACCELERATOR_COUNT = 1\n",
"EVAL_MACHINE_TYPE = \"n1-highmem-8\""
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "zAaMJKrhAe5L"
},
"source": [
"## Define components\n",
"\n",
"This pipeline is composed from the following components:\n",
"\n",
"- **train-tfhub-model** - Trains a new Tensorflow model using TFHub layers from pre-built Docker image\n",
"- **upload-tensorflow-model-to-google-cloud-vertex-ai** - Uploads resulting model to Vertex AI model registry\n",
"- **get-vertex-model** - Gets model that has just been uploaded as an artifact in pipeline\n",
"- **convert-dataset-export-for-batch-predict** - Preprocessing component that takes the test dataset exported from Vertex datasets and converts it to a simpler compatible one that is readable from the batch predict component\n",
"- **target-field-data-remover** - Removes the target field (i.e., label) in the test dataset for the downstream batch predict component\n",
"- **model-batch-predict** - Performs a batch prediction job\n",
"- **model-evaluation-classification** - Calculates the evaluation metrics from the above batch predict job and exports the metrics artifact\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "DKe2iQNKgpKG"
},
"outputs": [],
"source": [
"# Load upload TF model component\n",
"upload_tensorflow_model_to_vertex_op = components.load_component_from_url(\n",
" \"https://raw.githubusercontent.com/Ark-kun/pipeline_components/c6a8b67d1ada2cc17665c99ff6b410df588bee28/components/google-cloud/Vertex_AI/Models/Upload_Tensorflow_model/workaround_for_buggy_KFPv2_compiler/component.yaml\"\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "TEnh9Pcx6Xfi"
},
"source": [
"### Define the pipeline\n",
"\n",
"The pipeline performs the following steps:\n",
"- Trains new text classification model\n",
"- Uploads model to Vertex AI Model Registry\n",
"- Performs preprocessing steps on test dataset export: formats data for batch predcition, removes target field\n",
"- Performs batch prediction on preprocessed test data\n",
"- Evaluates performance of model based on batch prediction output"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "2a67cde8"
},
"outputs": [],
"source": [
"@dsl.pipeline(name=\"text-classification-model\")\n",
"def pipeline():\n",
" train_task = natural_language.TrainTextClassificationOp()(\n",
" project=PROJECT_ID,\n",
" location=LOCATION,\n",
" machine_type=TRAINING_MACHINE_TYPE,\n",
" accelerator_type=ACCELERATOR_TYPE,\n",
" accelerator_count=ACCELERATOR_COUNT,\n",
" input_data_path=TRAINING_DATA_LOCATION,\n",
" input_format=\"jsonl\",\n",
" natural_language_task_type=TASK_TYPE,\n",
" )\n",
"\n",
" upload_task = upload_tensorflow_model_to_vertex_op(\n",
" model=train_task.outputs[\"model_output\"]\n",
" )\n",
"\n",
" get_model_task = GetVertexModelOp(\n",
" model_resource_name=upload_task.outputs[\"model_name\"]\n",
" )\n",
"\n",
" classification_type = (\n",
" \"multilabel\" if TASK_TYPE == \"MULTILABEL_CLASSIFICATION\" else \"multiclass\"\n",
" )\n",
"\n",
" convert_dataset_task = natural_language.ConvertDatasetExportForBatchPredictOp(\n",
" file_paths=TEST_DATA_URIS, classification_type=classification_type\n",
" )\n",
"\n",
" target_field_remover_task = TargetFieldDataRemoverOp(\n",
" project=PROJECT_ID,\n",
" location=LOCATION,\n",
" root_dir=BUCKET_URI,\n",
" gcs_source_uris=convert_dataset_task.outputs[\"output_files\"],\n",
" target_field_name=\"labels\",\n",
" instances_format=\"jsonl\",\n",
" )\n",
"\n",
" # Note: ModelBatchPredictOp doesn't support accelerators currently.\n",
" batch_predict_task = ModelBatchPredictOp(\n",
" project=PROJECT_ID,\n",
" location=LOCATION,\n",
" model=get_model_task.outputs[\"model\"],\n",
" job_display_name=\"nl-batch-predict-evaluation\",\n",
" gcs_source_uris=target_field_remover_task.outputs[\"gcs_output_directory\"],\n",
" instances_format=\"jsonl\",\n",
" predictions_format=\"jsonl\",\n",
" gcs_destination_output_uri_prefix=BUCKET_URI,\n",
" machine_type=EVAL_MACHINE_TYPE,\n",
" )\n",
"\n",
" # Note: Because we're running a custom training pipeline, the model source\n",
" # is detected as Custom and thus it doesn't use AutoML NL's default settings\n",
" # and fails if class_labels is excluded.\n",
" ModelEvaluationClassificationOp(\n",
" project=PROJECT_ID,\n",
" location=LOCATION,\n",
" root_dir=BUCKET_URI,\n",
" class_labels=CLASS_NAMES + [\"[UNK]\"],\n",
" predictions_gcs_source=batch_predict_task.outputs[\"gcs_output_directory\"],\n",
" predictions_format=\"jsonl\",\n",
" prediction_label_column=\"prediction.displayNames\",\n",
" prediction_score_column=\"prediction.confidences\",\n",
" ground_truth_gcs_source=convert_dataset_task.outputs[\"output_files\"],\n",
" ground_truth_format=\"jsonl\",\n",
" target_field_name=\"labels\",\n",
" classification_type=TASK_TYPE,\n",
" )"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "3211ba19"
},
"source": [
"### Compile the pipeline"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "c368c73f"
},
"outputs": [],
"source": [
"compiler.Compiler().compile(pipeline, \"nl_pipeline.json\")"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "l_Vxwz5cdF5f"
},
"source": [
"Running the above line will generate a file locally or in Colab's directory."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "ax0jOxIaholy"
},
"source": [
"### Run the pipeline\n",
"\n",
"This sends a create pipeline job request to Vertex Pipelines. Note that this task run synchronously and may take a while to complete.\n",
"\n",
"You may view the progress of the job at any time by clicking on the generated links (after \"View Pipeline Job\" in the console output of the cell below). Once the pipeline finishes, you may examine the artifacts produced from this pipeline."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "Wfs7QOSxhp_n"
},
"outputs": [],
"source": [
"job = aiplatform.PipelineJob(\n",
" display_name=\"nl_pipeline\",\n",
" template_path=\"nl_pipeline.json\",\n",
" location=LOCATION,\n",
" enable_caching=True,\n",
" parameter_values={},\n",
")\n",
"\n",
"job.run()"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "UIyGPaihWJWn"
},
"source": [
"Once the pipeline successfully finishes, go to the pipeline and examine the resulting metrics artifacts for the results. Otherwise, refer to the failing step(s) in the pipeline to determine the cause of any errors."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "OoexTJTy9jnH"
},
"source": [
"## View model evaluation results\n",
"\n",
"To check the results of evaluation after pipeline execution, find the \"model-evaluation-classification\" subdirectory in the Cloud Storage bucket created by this pipeline. You may also run the following to directly output the contents of the metrics file:"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "h9EqPCQF9lN9"
},
"outputs": [],
"source": [
"import tensorflow as tf\n",
"\n",
"EVAL_TASK_NAME = \"model-evaluation-classification\"\n",
"PROJECT_NUMBER = job.gca_resource.name.split(\"/\")[1]\n",
"for _ in range(len(job.gca_resource.job_detail.task_details)):\n",
" TASK_ID = job.gca_resource.job_detail.task_details[_].task_id\n",
" EVAL_METRICS = (\n",
" BUCKET_URI\n",
" + \"/\"\n",
" + PROJECT_NUMBER\n",
" + \"/\"\n",
" + job.name\n",
" + \"/\"\n",
" + EVAL_TASK_NAME\n",
" + \"_\"\n",
" + str(TASK_ID)\n",
" + \"/executor_output.json\"\n",
" )\n",
" if tf.io.gfile.exists(EVAL_METRICS):\n",
" ! gsutil cat $EVAL_METRICS"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "TpV-iwP9qw9c"
},
"source": [
"## Cleaning up\n",
"\n",
"To clean up the resources used by this pipeline, run the command below:"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "sx_vKniMq9ZX"
},
"outputs": [],
"source": [
"# Delete GCS bucket.\n",
"!gsutil -m rm -r {BUCKET_URI}"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "UMuyzrnZLoUa"
},
"source": [
"# Next steps\n",
"\n",
"For an alternate approach, please check out the [\"ready-to-go\" text classification pipeline](https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/master/notebooks/community/pipelines/google_cloud_pipeline_components_ready_to_go_text_classification_pipeline.ipynb). This pipeline exposes the model logic for further customization if needed, and adds an additional pipeline step to deploy the model to enable online predictions."
]
}
],
"metadata": {
"colab": {
"collapsed_sections": [
"d975e698c9a4",
"08d289fa873f",
"d33c87e4-2ada-4b87-bf75-064247f3162d",
"3211ba19",
"TpV-iwP9qw9c",
"UMuyzrnZLoUa"
],
"name": "google_cloud_pipeline_components_cloud_natural_language_pipeline.ipynb",
"toc_visible": true
},
"kernelspec": {
"display_name": "Python 3",
"name": "python3"
}
},
"nbformat": 4,
"nbformat_minor": 0
}
File diff suppressed because one or more lines are too long
@@ -112,7 +112,7 @@ def benchmark(
results = []
for qps in qps_list:
num_requests = max(qps * duration_sec, 10)
num_requests = int(max(qps * duration_sec, 10))
requests_for_qps = list(
itertools.islice(itertools.cycle(requests), num_requests)
)
@@ -0,0 +1,987 @@
{
"cells": [
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "18ebbd838e32"
},
"outputs": [],
"source": [
"# Copyright 2023 Google LLC\n",
"#\n",
"# Licensed under the Apache License, Version 2.0 (the \"License\");\n",
"# you may not use this file except in compliance with the License.\n",
"# You may obtain a copy of the License at\n",
"#\n",
"# https://www.apache.org/licenses/LICENSE-2.0\n",
"#\n",
"# Unless required by applicable law or agreed to in writing, software\n",
"# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
"# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
"# See the License for the specific language governing permissions and\n",
"# limitations under the License."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "219f1b1fe8fe"
},
"source": [
"# Deploy and host a Stable Diffusion model on Vertex AI\n"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "JAPoU8Sm5E6e"
},
"source": [
"<table align=\"left\">\n",
"\n",
" <td>\n",
" <a href=\"https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/vertex_endpoints/torchserve/dreambooth_stablediffusion.ipynb\">\n",
" <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\" alt=\"Colab logo\"> Run in Colab\n",
" </a>\n",
" </td>\n",
" <td>\n",
" <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/vertex_endpoints/torchserve/dreambooth_stablediffusion.ipynb\">\n",
" <img src=\"https://cloud.google.com/ml-engine/images/github-logo-32px.png\" alt=\"GitHub logo\">\n",
" View on GitHub\n",
" </a>\n",
" </td>\n",
" <td>\n",
" <a href=\"https://console.cloud.google.com/vertex-ai/workbench/deploy-notebook?download_url=https://raw.githubusercontent.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/community/vertex_endpoints/torchserve/dreambooth_stablediffusion.ipynb\">\n",
" <img src=\"https://lh3.googleusercontent.com/UiNooY4LUgW_oTvpsNhPpQzsstV5W8F7rYgxgGBD85cWJoLmrOzhVs_ksK_vgx40SHs7jCqkTkCk=e14-rj-sc0xffffff-h130-w32\" alt=\"Vertex AI logo\">\n",
" Open in Vertex AI Workbench\n",
" </a>\n",
" </td>\n",
"</table>"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "fce05a8186d6"
},
"source": [
"## Overview\n",
"\n",
"This notebook demonstrates how to deploy and host a fine-tuned [Stable Diffusion 1.5](https://huggingface.co/runwayml/stable-diffusion-v1-5) model on Vertex AI. For hosting, you use the PyTorch 3 container built for Vertex AI with [TorchServe](https://pytorch.org/serve/index.html)."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "c76216b03fec"
},
"source": [
"### Objective\n",
"\n",
"In this tutorial, you learn how to host and deploy a Stable Diffusion 1.5 model on Vertex AI.\n",
"\n",
"This tutorial uses the following Google Cloud ML services:\n",
"\n",
"+ Vertex AI `Model` resource\n",
"+ Vertex AI `Endpoint` resource\n",
"\n",
"The steps performed include:\n",
"\n",
"+ Create a `torchserve` handler for responding to prediction requests.\n",
"+ Upload a Stable Diffusion 1.5 model on a prebuilt PyTorch container in Vertex AI.\n",
"+ Deploy a model to a Vertex AI Endpoint.\n",
"+ Send requests to the endpoint and parse the responses using Vertex AI Prediction service."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "c6deba5a8557"
},
"source": [
"### Model\n",
"\n",
"This notebook uses a collection of model artifacts fine-tuned to generate images of a small dog. These are the same images used in the original [DreamBooth paper](https://dreambooth.github.io/)."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "911dc651ea9c"
},
"source": [
"### Costs\n",
"\n",
"This tutorial uses billable components of Google Cloud:\n",
"\n",
"* Vertex AI models\n",
"* Vertex AI endpoints\n",
"* Vertex AI prediction\n",
"* Cloud Storage\n",
"* (Optionally) Vertex AI Workbench\n",
"\n",
"Learn about [Vertex AI pricing](https://cloud.google.com/vertex-ai/pricing) and [Cloud Storage pricing](https://cloud.google.com/storage/pricing), and use the [Pricing Calculator](https://cloud.google.com/products/calculator/) to generate a cost estimate based on your projected usage."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "0d36bd3d53fa"
},
"source": [
"## Hardware requirements\n",
"\n",
"This notebook requires that you use a GPU with a sufficient amount of VRAM available. It was tested on a `NVIDIA Tesla A100 GPU` with 85 GB of VRAM. Run the following cell to ensure that you have the correct hardware configuration."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "3dd4022552e5"
},
"outputs": [],
"source": [
"!nvidia-smi --query-gpu=name,memory.total,memory.free --format=\"csv,noheader\""
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "a782627e5f73"
},
"source": [
"### Create a user-managed notebook on Vertex AI\n",
"\n",
"If you are using Vertex AI Workbench, you can create a notebook with the correct configuration by doing the following:\n",
"\n",
"+ Go to [Vertex AI Workbench](https://console.cloud.google.com/vertex-ai/workbench/user-managed) in the Google Cloud Console.\n",
"+ Click **New Notebook** and then click **PyTorch 1.13** > **With 1 NVIDIA T4**.\n",
"+ In the **New notebook** dialog box, click **Advanced Options**. The **Create a user-managed notebook** page opens up.\n",
"+ In the **Create a user-managed notebook** page, do the following:\n",
" * In the **Notebook name** box, type a name for your notebook, for example \"my-stablediffusion-nb\".\n",
" * In the **Machine type** drop-down, select **A2 highgpu** > **a2-highgpu-1g**.\n",
" * In the **GPU type** drop-down, select **NVIDIA Tesla A100**.\n",
" * Check the box next to **Install NVIDIA GPU driver automatically for me**\n",
" * Expand **Disk(s)** and do the following:\n",
" - Under **Boot disk type**, select **SSD Persistent Disk**.\n",
" - Under **Data disk type**, select **SSD Persistent Disk**.\n",
" * Click **Create**."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "FAPoU8Sm5E6e"
},
"source": [
"<div style=\"background:#feefe3; padding:5px; color:#aa0000\">\n",
"<strong>Caution:</strong> Using a Vertex AI Workbench notebook with the above configuration can increase your costs significantly. You can estimate your costs using the <a href=\"https://cloud.google.com/products/calculator\"><u>costs calculator</u></a>.</div>"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "0bb4201cc99a"
},
"source": [
"## Installation\n",
"\n",
"Install the following packages required to execute this notebook.\n",
"\n",
"**Note**: You might need to change the version of PyTorch (`torch`) installed by `pip`."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "9c769df171a6"
},
"outputs": [],
"source": [
"%%writefile requirements.txt\n",
"diffusers\n",
"ftfy\n",
"google-cloud-aiplatform\n",
"gradio\n",
"ninja\n",
"tensorboard==1.15.0\n",
"torch\n",
"torchaudio\n",
"torchvision\n",
"torchserve\n",
"torch-model-archiver\n",
"torch-workflow-archiver\n",
"transformers"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "e46804ac90d8"
},
"outputs": [],
"source": [
"%pip install -r requirements.txt"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "58707a750154"
},
"source": [
"### Colab only: Uncomment the following cell to restart the kernel."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "77c11549298a"
},
"outputs": [],
"source": [
"# Automatically restart kernel after installs so that your environment can access the new packages\n",
"# import IPython\n",
"\n",
"# app = IPython.Application.instance()\n",
"# app.kernel.do_shutdown(True)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "294df346a918"
},
"source": [
"## Before you begin"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "BF1j6f9HApxa"
},
"source": [
"### Set up your Google Cloud project\n",
"\n",
"**The following steps are required, regardless of your notebook environment.**\n",
"\n",
"1. [Select or create a Google Cloud project](https://console.cloud.google.com/cloud-resource-manager). When you first create an account, you get a $300 free credit towards your compute/storage costs.\n",
"\n",
"2. [Make sure that billing is enabled for your project](https://cloud.google.com/billing/docs/how-to/modify-project).\n",
"\n",
"3. [Enable the Vertex AI API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com).\n",
"\n",
"4. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk)."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "WReHDGG5g0XY"
},
"source": [
"#### Set your project ID\n",
"\n",
"**If you don't know your project ID**, try the following:\n",
"* Run `gcloud config list`.\n",
"* Run `gcloud projects list`.\n",
"* See the support page: [Locate the project ID](https://support.google.com/googleapi/answer/7014113)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "oM1iC_MfAts1"
},
"outputs": [],
"source": [
"PROJECT_ID = \"[your-project-id]\" # @param {type:\"string\"}\n",
"\n",
"# Set the project id\n",
"! gcloud config set project {PROJECT_ID}"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "region"
},
"source": [
"#### Region\n",
"\n",
"You can also change the `REGION` variable used by Vertex AI. Learn more about [Vertex AI regions](https://cloud.google.com/vertex-ai/docs/general/locations)."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "region"
},
"outputs": [],
"source": [
"REGION = \"us-central1\""
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "sBCra4QMA2wR"
},
"source": [
"### Authenticate your Google Cloud account\n",
"\n",
"Depending on your Jupyter environment, you may have to manually authenticate. Follow the relevant instructions below."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "74ccc9e52986"
},
"source": [
"**1. Vertex AI Workbench**\n",
"* Do nothing as you are already authenticated."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "de775a3773ba"
},
"source": [
"**2. Local JupyterLab instance, uncomment and run:**"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "254614fa0c46"
},
"outputs": [],
"source": [
"# ! gcloud auth login"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "ef21552ccea8"
},
"source": [
"**3. Colab, uncomment and run:**"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "603adbbf0532"
},
"outputs": [],
"source": [
"# from google.colab import auth\n",
"# auth.authenticate_user()"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "f6b2ccc891ed"
},
"source": [
"**4. Service account or other**\n",
"* See how to grant Cloud Storage permissions to your service account at https://cloud.google.com/storage/docs/gsutil/commands/iam#ch-examples."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "cb5c4ca3e851"
},
"source": [
"### Import libraries"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "7348591eda51"
},
"outputs": [],
"source": [
"import base64\n",
"import math\n",
"\n",
"import torch\n",
"from diffusers import StableDiffusionPipeline\n",
"from google.cloud import aiplatform\n",
"from IPython import display\n",
"from PIL import Image\n",
"from torch import autocast"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "697566b5f660"
},
"source": [
"## Optional: View model inferences\n",
"\n",
"Before uploading the model to Vertex AI, you can review the expected output from the model. The model used in this notebook is available for your use and can be downloaded from Cloud Storage. This download may take a few minutes to complete."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "d63df8d91215"
},
"outputs": [],
"source": [
"!gsutil -m cp gs://cloud-samples-data/vertex-ai/model-deployment/models/stable-diffusion/model_artifacts.zip \\\n",
" ."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "89d613ae9573"
},
"outputs": [],
"source": [
"!unzip model_artifacts.zip"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "9192ea4f3b57"
},
"source": [
"### Create new images\n",
"\n",
"With everything in place, you can now generate new images from the Stable Diffusion model. First you must load your model into a `StableDiffusionPipeline`."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "cd1f30223b79"
},
"outputs": [],
"source": [
"model_path = \"model_artifacts\"\n",
"\n",
"pipe = StableDiffusionPipeline.from_pretrained(\n",
" model_path, torch_dtype=torch.float16\n",
").to(\"cuda\")\n",
"\n",
"g_cuda = None"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "dd91520f58cf"
},
"outputs": [],
"source": [
"g_cuda = torch.Generator(device=\"cuda\")\n",
"seed = 52362\n",
"g_cuda.manual_seed(seed)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "1e9c5d734b9f"
},
"source": [
"With the model loaded into a `StableDiffusionPipeline`, you can now generate results (inferences) from the model. Each set of inference requires an input (called a [prompt](https://learnprompting.org/)) that specifies what the model should create.\n",
"\n",
"You can also vary other inputs into the model, as shown in the following cell."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "a28b73de55ce"
},
"outputs": [],
"source": [
"prompt = \"photo of examplePup dog in a Monet style\"\n",
"\n",
"num_samples = 4\n",
"num_batches = 1\n",
"num_columns = 2\n",
"guidance_scale = 10\n",
"num_inference_steps = 50\n",
"height = 512\n",
"width = 512"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "11a7bb79de59"
},
"outputs": [],
"source": [
"def image_grid(imgs, cols):\n",
" total = len(imgs)\n",
" rows = math.ceil(total / cols)\n",
"\n",
" w, h = imgs[0].size\n",
" grid = Image.new(\"RGB\", size=(cols * w, rows * h))\n",
" grid_w, grid_h = grid.size\n",
"\n",
" for i, img in enumerate(imgs):\n",
" grid.paste(img, box=(i % cols * w, i // cols * h))\n",
" return grid\n",
"\n",
"\n",
"all_images = []\n",
"for _ in range(num_batches):\n",
" with autocast(\"cuda\"):\n",
" images = pipe(\n",
" [prompt] * num_samples,\n",
" height=height,\n",
" width=width,\n",
" num_inference_steps=num_inference_steps,\n",
" guidance_scale=guidance_scale,\n",
" ).images\n",
" all_images.extend(images)\n",
"\n",
"\n",
"grid = image_grid(all_images, num_columns)\n",
"grid"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "bbc72963ff88"
},
"source": [
"## Deploy the model to Vertex AI\n",
"\n",
"You can host your Stable Diffusion 1.5 model on a Vertex AI endpoint where you can get inferences from it online. Uploading your model is a four step process: \n",
"\n",
"1. Create a custom TorchServe handler.\n",
"1. Upload the model artifacts onto Cloud Storage.\n",
"2. Create a Vertex AI model with the model artifacts and a prebuilt PyTorch container image.\n",
"3. Deploy the Vertex AI model onto an endpoint."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "eafbb0e0-40e6-43a0-a38e-edc54323da51"
},
"source": [
"### Create the custom TorchServe handler\n",
"\n",
"The model deployed to Vertex AI uses [TorchServe](https://pytorch.org/serve/) to handle requests and return responses from the model. You must create a custom TorchServe handler to include in with the model artifacts uploaded to Vertex AI.\n",
"\n",
"The handler file should be included in the directory with the other model artifacts."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "94567a87-9d74-4c87-a749-306ddaf01b61"
},
"outputs": [],
"source": [
"%%writefile model_artifacts/handler.py\n",
"\n",
"\"\"\"Customized handler for Stable Diffusion 1.5.\"\"\"\n",
"import base64\n",
"import logging\n",
"from io import BytesIO\n",
"\n",
"import torch\n",
"from diffusers import EulerDiscreteScheduler\n",
"from diffusers import StableDiffusionPipeline\n",
"from ts.torch_handler.base_handler import BaseHandler\n",
"\n",
"logger = logging.getLogger(__name__)\n",
"model_id = 'runwayml/stable-diffusion-v1-5'\n",
"\n",
"\n",
"class ModelHandler(BaseHandler):\n",
"\n",
" def __init__(self):\n",
" self.initialized = False\n",
" self.map_location = None\n",
" self.device = None\n",
" self.use_gpu = True\n",
" self.store_avg = True\n",
" self.pipe = None\n",
"\n",
" def initialize(self, context):\n",
" \"\"\"Initializes the pipe.\"\"\"\n",
" properties = context.system_properties\n",
" gpu_id = properties.get('gpu_id')\n",
"\n",
" self.map_location, self.device, self.use_gpu = \\\n",
" ('cuda', torch.device('cuda:' + str(gpu_id)),\n",
" True) if torch.cuda.is_available() else \\\n",
" ('cpu', torch.device('cpu'), False)\n",
"\n",
" # Use the Euler scheduler here instead\n",
" scheduler = EulerDiscreteScheduler.from_pretrained(model_id,\n",
" subfolder='scheduler')\n",
" pipe = StableDiffusionPipeline.from_pretrained(model_id,\n",
" scheduler=scheduler,\n",
" torch_dtype=torch.float16)\n",
" pipe = pipe.to('cuda')\n",
" # Uncomment the following line to reduce the GPU memory usage.\n",
" # pipe.enable_attention_slicing()\n",
" self.pipe = pipe\n",
"\n",
" self.initialized = True\n",
"\n",
" def preprocess(self, requests):\n",
" \"\"\"Noting to do here.\"\"\"\n",
" logger.info('requests: %s', requests)\n",
" return requests\n",
"\n",
" def inference(self, preprocessed_data, *args, **kwargs):\n",
" \"\"\"Run the inference.\"\"\"\n",
" images = []\n",
" for pd in preprocessed_data:\n",
" prompt = pd['prompt']\n",
" images.extend(self.pipe(prompt).images)\n",
" return images\n",
"\n",
" def postprocess(self, output_batch):\n",
" \"\"\"Converts the images to base64 string.\"\"\"\n",
" postprocessed_data = []\n",
" for op in output_batch:\n",
" fp = BytesIO()\n",
" op.save(fp, format='JPEG')\n",
" postprocessed_data.append(base64.b64encode(fp.getvalue()).decode('utf-8'))\n",
" fp.close()\n",
" return postprocessed_data\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "6ace1dac0af0"
},
"source": [
"After creating the handler file, you must package the handler as a model archiver (MAR) file. The output file must be named 'model.mar'."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "67707f95d440"
},
"outputs": [],
"source": [
"!torch-model-archiver \\\n",
" -f \\\n",
" --model-name model \\\n",
" --version 1.0 \\\n",
" --handler model_artifacts/handler.py \\\n",
" --export-path model_artifacts"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "ffab030f4bc8"
},
"source": [
"### Upload the model artifacts to Cloud Storage\n",
"\n",
"Create a new folder in your Cloud Storage bucket to hold the model artifacts"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "MzGDU7TWdts_"
},
"outputs": [],
"source": [
"BUCKET_NAME = \"your-bucket-name-unique\" # @param {type:\"string\"}\n",
"BUCKET_URI = f\"gs://{BUCKET_NAME}/\"\n",
"FULL_GCS_PATH = f\"{BUCKET_URI}model_artifacts\""
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "-EcIXiGsCePi"
},
"source": [
"**Only if your bucket doesn't already exist**: Run the following cell to create your Cloud Storage bucket."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "NIq7R4HZCfIc"
},
"outputs": [],
"source": [
"! gsutil mb -l $REGION -p $PROJECT_ID $BUCKET_URI"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "971232e28657"
},
"source": [
"Next, upload the model archive file and your trained Stable Diffusion 1.5 model to the folder on Cloud Storage."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "ef6baf44c808"
},
"outputs": [],
"source": [
"!gsutil cp -r model_artifacts $BUCKET_URI"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "402370ca9396"
},
"source": [
"### Create the Vertex AI model\n",
"\n",
"Once you've uploaded the model artifacts into a Cloud Storage bucket, you can create a new Vertex AI model. This notebook uses the [Vertex AI SDK](https://cloud.google.com/vertex-ai/docs/start/use-vertex-ai-python-sdk) to create the model."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "b6c58a74a0fd"
},
"outputs": [],
"source": [
"PYTORCH_PREDICTION_IMAGE_URI = (\n",
" \"us-docker.pkg.dev/vertex-ai/prediction/pytorch-gpu.1-12:latest\"\n",
")\n",
"APP_NAME = \"my-stable-diffusion\"\n",
"VERSION = 1\n",
"MODEL_DISPLAY_NAME = \"stable_diffusion_1_5-unique\"\n",
"MODEL_DESCRIPTION = \"stable_diffusion_1_5 container\"\n",
"ENDPOINT_DISPLAY_NAME = f\"{APP_NAME}-endpoint\""
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "07c3503a1a2e"
},
"outputs": [],
"source": [
"aiplatform.init(project=PROJECT_ID, location=REGION, staging_bucket=BUCKET_NAME)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "FAPoU8Sm5E6e"
},
"source": [
"<div style=\"background:#e3effe; padding:5px; color:#0000aa\">\n",
"<strong>Note:</strong> The next cell fails if you haven't <a href=\"https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com\"><u>enabled the Vertex API</u></a>.</div>"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "a776324dd16f"
},
"outputs": [],
"source": [
"model = aiplatform.Model.upload(\n",
" display_name=MODEL_DISPLAY_NAME,\n",
" description=MODEL_DESCRIPTION,\n",
" serving_container_image_uri=PYTORCH_PREDICTION_IMAGE_URI,\n",
" artifact_uri=FULL_GCS_PATH,\n",
")\n",
"\n",
"model.wait()\n",
"\n",
"print(model.display_name)\n",
"print(model.resource_name)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "0fbc3d371574"
},
"source": [
"### Deploy the model to an endpoint\n",
"\n",
"To get online preductions from your Stable Diffusion 2.0 model, you must [deploy it to a Vertex AI endpoint](https://cloud.google.com/vertex-ai/docs/predictions/overview). You can again use the Vertex AI SDK to create the endpoint and deploy your model."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "ab29f0a770cb"
},
"outputs": [],
"source": [
"endpoint = aiplatform.Endpoint.create(display_name=ENDPOINT_DISPLAY_NAME)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "25f703df88c7"
},
"outputs": [],
"source": [
"model.deploy(\n",
" endpoint=endpoint,\n",
" deployed_model_display_name=MODEL_DISPLAY_NAME,\n",
" machine_type=\"n1-standard-8\",\n",
" accelerator_type=\"NVIDIA_TESLA_P100\",\n",
" accelerator_count=1,\n",
" traffic_percentage=100,\n",
" deploy_request_timeout=1200,\n",
" sync=True,\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "c9fc560df0a9"
},
"source": [
"The previous cell, which deploys your model to the endpoint, can take a while to complete. If the previous cell times out before returning, your endpoint might still be successfully deployed to an endpoint. Check the [Cloud Console](https://console.cloud.google.com/vertex-ai/endpoints) to verify the results.\n",
"\n",
"You can also extend the time to wait for deployment by changing the `deploy_request_timeout` argument passed to `model.deploy()`."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "88a5304dfdc9"
},
"source": [
"## Get online predictions\n",
"\n",
"Finally, with your Stable Diffusion 1.5 model deployed to a Vertex AI endpoint, you can now get online predictions from it. Using the Vertex AI SDK, you only need a few lines of code to get an inference."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "0d6bc4aa34d6"
},
"outputs": [],
"source": [
"instances = [{\"prompt\": \"An examplePup dog with a baseball jersey.\"}]\n",
"response = endpoint.predict(instances=instances)\n",
"\n",
"with open(\"img5.jpg\", \"wb\") as g:\n",
" g.write(base64.b64decode(response.predictions[0]))"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "65bafefda60c"
},
"outputs": [],
"source": [
"display.Image(\"img5.jpg\")"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "TpV-iwP9qw9c"
},
"source": [
"## Cleaning up\n",
"\n",
"To clean up all Google Cloud resources used in this project, you can [delete the Google Cloud\n",
"project](https://cloud.google.com/resource-manager/docs/creating-managing-projects#shutting_down_projects) you used for the tutorial.\n",
"\n",
"Otherwise, you can delete the individual resources you created in this tutorial:"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "sx_vKniMq9ZX"
},
"outputs": [],
"source": [
"import os\n",
"\n",
"# Delete endpoint resource\n",
"endpoint.undeploy_all()\n",
"endpoint.delete()\n",
"\n",
"# Delete model resource\n",
"model.delete()\n",
"\n",
"# Delete Cloud Storage objects that were created\n",
"delete_bucket = False\n",
"if delete_bucket or os.getenv(\"IS_TESTING\"):\n",
" ! gsutil -m rm -r $BUCKET_URI"
]
}
],
"metadata": {
"colab": {
"name": "dreambooth_stablediffusion.ipynb",
"toc_visible": true
},
"kernelspec": {
"display_name": "Python 3",
"name": "python3"
}
},
"nbformat": 4,
"nbformat_minor": 0
}
+3 -1
View File
@@ -74,7 +74,9 @@
"source": [
"## Overview\n",
"\n",
"{TODO: Include a paragraph or two explaining what this example demonstrates, who should be interested in it, and what you need to know before you get started.}"
"{TODO: Include a paragraph or two explaining what this example demonstrates, who should be interested in it, and what you need to know before you get started.}\n",
"\n",
"Learn more about [web-doc-title](linkback-to-webdoc-page). {TODO: if more than one primary feature, add tag/linkback for each one}"
]
},
{
+192 -45
View File
@@ -12,6 +12,10 @@
--errors-codes: A list of error codes to report errors. Otherwise, all errors are reported.
--errors-csv: Report errors in CSV format
# options for automatic fixing
--fix: Automatic fix
--fix-codes: A list of fix codes to fix. Otherwise, all fix codes are enabled.
# index generatation
--repo: Generate index in markdown format
--web: Generate index in HTML format
@@ -19,6 +23,7 @@
--desc: Add description to index
--steps: Add steps to index
--uses: Add "resources" used to index
--linkback: Add linkback to index
Format of CSV file for notebooks to review:
@@ -61,6 +66,8 @@ parser.add_argument('--uses', dest='uses', action='store_true',
default=False, help='Output uses (resources)')
parser.add_argument('--steps', dest='steps', action='store_true',
default=False, help='Ouput steps')
parser.add_argument('--linkback', dest='linkback', action='store_true',
default=False, help='Ouput linkback')
parser.add_argument('--web', dest='web', action='store_true',
default=False, help='Output format in HTML')
parser.add_argument('--repo', dest='repo', action='store_true',
@@ -109,13 +116,14 @@ class ErrorCode(Enum):
# Costs cell required
# Check for required Vertex and optional BQ and Dataflow
ERROR_OVERVIEW_NOTFOUND = 10,
ERROR_OBJECTIVE_NOTFOUND = 11,
ERROR_OBJECTIVE_MISSING_DESC = 12,
ERROR_OBJECTIVE_MISSING_USES = 13,
ERROR_OBJECTIVE_MISSING_STEPS = 14,
ERROR_DATASET_NOTFOUND = 15,
ERROR_COSTS_NOTFOUND = 16,
ERROR_COSTS_MISSING = 17,
ERROR_LINKBACK_NOTFOUND = 11,
ERROR_OBJECTIVE_NOTFOUND = 12,
ERROR_OBJECTIVE_MISSING_DESC = 13,
ERROR_OBJECTIVE_MISSING_USES = 14,
ERROR_OBJECTIVE_MISSING_STEPS = 15,
ERROR_DATASET_NOTFOUND = 16,
ERROR_COSTS_NOTFOUND = 17,
ERROR_COSTS_MISSING = 18,
# Installation cell
# Installation cell required
@@ -126,34 +134,34 @@ class ErrorCode(Enum):
# option {USER_FLAG} required
# installation code cell not match template
# all packages must be installed as a single pip3
ERROR_INSTALLATION_NOTFOUND = 18,
ERROR_INSTALLATION_HEADING = 19,
ERROR_INSTALLATION_CODE_NOTFOUND = 20,
ERROR_INSTALLATION_PIP3 = 21,
ERROR_INSTALLATION_QUIET = 22,
ERROR_INSTALLATION_USER_FLAG = 23,
ERROR_INSTALLATION_CODE_TEMPLATE = 24,
ERROR_INSTALLATION_SINGLE_PIP3 = 25,
ERROR_INSTALLATION_NOTFOUND = 19,
ERROR_INSTALLATION_HEADING = 20,
ERROR_INSTALLATION_CODE_NOTFOUND = 21,
ERROR_INSTALLATION_PIP3 = 22,
ERROR_INSTALLATION_QUIET = 23,
ERROR_INSTALLATION_USER_FLAG = 24,
ERROR_INSTALLATION_CODE_TEMPLATE = 25,
ERROR_INSTALLATION_SINGLE_PIP3 = 26,
# Restart kernel cell
# Restart code cell required
# Restart code cell not found
ERROR_RESTART_NOTFOUND = 23,
ERROR_RESTART_CODE_NOTFOUND = 24,
ERROR_RESTART_NOTFOUND = 27,
ERROR_RESTART_CODE_NOTFOUND = 28,
# Before you begin cell
# Before you begin cell required
# Before you begin cell incomplete
ERROR_BEFOREBEGIN_NOTFOUND = 25,
ERROR_BEFOREBEGIN_INCOMPLETE = 26,
ERROR_BEFOREBEGIN_NOTFOUND = 29,
ERROR_BEFOREBEGIN_INCOMPLETE = 30,
# Set Project ID
# Set project ID cell required
# Set project ID code cell not found
# Set project ID not match template
ERROR_PROJECTID_NOTFOUND = 27,
ERROR_PROJECTID_CODE_NOTFOUND = 28,
ERROR_PROJECTID_TEMPLATE = 29,
ERROR_PROJECTID_NOTFOUND = 31,
ERROR_PROJECTID_CODE_NOTFOUND = 32,
ERROR_PROJECTID_TEMPLATE = 33,
# Technical Writer Rules
ERROR_TWRULE_TODO = 51,
@@ -182,7 +190,21 @@ def parse_dir(directory: str) -> int:
"""
exit_code = 0
sorted_entries = []
entries = os.scandir(directory)
for entry in entries:
inserted = False
for ix in range(len(sorted_entries)):
if entry.name < sorted_entries[ix].name:
sorted_entries.insert(ix, entry)
inserted = True
break
if not inserted:
sorted_entries.append(entry)
entries = sorted_entries
for entry in entries:
if entry.is_dir():
if entry.name[0] == '.':
@@ -590,6 +612,8 @@ class OverviewRule(NotebookRule):
linkback = more.split('(')[1].split(')')[0]
self.tags.append(tag)
self.linkbacks.append(linkback)
else:
return notebook.report_error(ErrorCode.ERROR_LINKBACK_NOTFOUND, "Linkback missing in overview section")
return True
@@ -618,6 +642,10 @@ class ObjectiveRule(NotebookRule):
in_steps = False
for line in cell['source'][1:]:
# TOC anchor
if line.startswith('<a name='):
continue
if line.startswith('This tutorial uses'):
in_desc = False
in_steps = False
@@ -654,16 +682,39 @@ class ObjectiveRule(NotebookRule):
# check for italic font setting
if ch == '*' and sline[1] != ' ':
in_steps = False
# special case
elif sline.startswith('* Prediction Service'):
in_steps = False
else:
self.steps += line
elif ch == '#':
in_steps = False
if self.desc == '':
ret = notebook.report_error(ErrorCode.ERROR_OBJECTIVE_MISSING_DESC, "Objective section missing desc")
else:
self.desc = self.desc.lstrip()
sentences = self.desc.split('.')
bracket = False
paren = False
sentences = ""
for _ in range(len(self.desc)):
if self.desc[_] == '[':
bracket = True
continue
elif self.desc[_] == ']':
bracket = False
continue
elif self.desc[_] == '(':
paren = True
elif self.desc[_] == ')':
paren = False
continue
if not paren:
sentences += self.desc[_]
sentences = sentences.split('.')
if len(sentences) > 1:
self.desc = sentences[0] + '.\n'
if self.desc.startswith('In this tutorial, you learn') or self.desc.startswith('In this notebook, you learn'):
@@ -683,7 +734,7 @@ class ObjectiveRule(NotebookRule):
ret = notebook.report_error(ErrorCode.ERROR_OBJECTIVE_MISSING_STEPS, "Objective section missing steps list")
notebook.costs = self.costs
ret = True
return ret
class RecommendationsRule(NotebookRule):
@@ -1080,48 +1131,65 @@ def add_index(path: str,
title = title.split(':')[-1].strip()
title = title[0].upper() + title[1:]
if args.web:
title = title.replace('`', '')
title = replace_cl(title.replace('`', ''))
print(' <tr>')
print(' <td>')
for tag in tags:
tag = replace_cl(tag)
print(f' {tag.strip()}<br/>\n')
print(' </td>')
print(' <td>')
print(f' <b>{title}</b><br/>\n')
print(f' <b>{title}</b>. ')
if args.desc:
desc = desc.replace('`', '')
desc = replace_cl(desc.replace('`', ''))
print('<br/>')
print(f' {desc}<br/>\n')
print(f' {desc}\n')
if args.steps:
print('<br/>' + steps.replace('\n', '<br/>').replace('-', '&nbsp;&nbsp;-').replace('*', '&nbsp;&nbsp;-') + '<br/>')
if linkbacks:
if args.linkback and linkbacks:
num = len(tags)
for _ in range(num):
if linkbacks[_].startswith("vertex-ai"):
print(f'<br/> Learn more about <a href="https://cloud.google.com/{linkbacks[_]}">{tags[_]}</a>\n')
print(f' Learn more about <a href="https://cloud.google.com/{linkbacks[_]}" target="_blank">{replace_cl(tags[_])}</a>.\n')
else:
print(f'<br/> Learn more about <a href="{linkbacks[_]}">{tags[_]}</a>\n')
print(f' Learn more about <a href="{linkbacks[_]}." target="_blank">{replace_cl(tags[_])}</a>.\n')
if args.steps:
print("<devsite-expandable>\n")
print(' <p class="showalways">Tutorial steps</p>\n')
print(' <ul>\n')
if ":" in steps:
steps = steps.split(':')[1].replace('*', '').replace('-', '').strip().split('\n')
else:
steps = []
for step in steps:
print(f' <li>{replace_cl(step)}</li>\n')
print(' </ul>\n')
print("</devsite-expandable>\n")
print(' </td>')
print(' <td>')
if colab_link:
print(f' <a href="{colab_link}" target="_blank">Colab</a><br/>\n')
print(f' <a href="{colab_link}" target="_blank" class="external" track-type="notebookTutorial" track-name="colabLink">Colab</a><br/>\n')
if git_link:
print(f' <a href="{git_link}" target="_blank">GitHub</a><br/>\n')
print(f' <a href="{git_link}" target="_blank" class="external" track-type="notebookTutorial" track-name="gitHubLink">GitHub</a><br/>\n')
if workbench_link:
print(f' <a href="{workbench_link}" target="_blank">Vertex AI Workbench</a><br/>\n')
print(f' <a href="{workbench_link}" target="_blank" class="external" track-type="notebookTutorial" track-name="workbenchLink">Vertex AI Workbench</a><br/>\n')
print(' </td>')
print(' </tr>\n')
elif args.repo:
if tags != last_tag and tag != '':
last_tag = tags
flat_list = ''
for item in tags:
flat_list += item.replace("'", '') + ' '
print(f"\n### {flat_list}\n")
try:
if tags != last_tag and tag != '':
last_tag = tags
flat_list = ''
for item in tags:
flat_list += item.replace("'", '') + ' '
print(f"\n### {flat_list}\n")
except:
pass
print(f"\n[{title}]({git_link})\n")
print("```")
@@ -1133,7 +1201,80 @@ def add_index(path: str,
if args.steps:
print(steps.rstrip() + '\n')
print("```\n")
if args.linkback and linkbacks:
num = len(tags)
for _ in range(num):
if linkbacks[_].startswith("vertex-ai"):
print(f'&nbsp;&nbsp;&nbsp;Learn more about [{tags[_]}]({linkbacks[_]}).\n')
else:
print(f'&nbsp;&nbsp;&nbsp;Learn more about [{tags[_]}]({linkbacks[_]}).\n')
def replace_cl(text : str ) -> str:
'''
Replace product names with CL substitution variables
'''
substitutions = {
#'AutoML Tabular Workflow': '{{automl_name}} Tabular Workflow',
#'AutoML Tables': '{{automl_tables_name}}',
#'AutoML Tabular': '{{automl_tables_name}}',
#'AutoML Vision': '{automl_vision_name}}',
#'AutoML Image': '{automl_vision_name}}',
'AutoML': '{{automl_name}}',
'BigQuery ML': '{{bigqueryml_name}}',
'BQML': '{{bigqueryml_name}}',
'BigQuery': '{{bigquery_name}}',
'BQ': '{{bigquery_name}}',
'Vertex Dataset': '{{vertex_ai_name}} Dataset',
'Vertex Model': '{{vertex_ai_name}} Model',
'Vertex Endpoint': '{{vertex_ai_name}} Endpoint',
'Vertex Model Registry': '{{vertex_model_registry_name}}',
'Vertex AI Model Registry': '{{vertex_model_registry_name}}',
'Vertex Training': '{{vertex_training_name}}',
'Vertex AI Training': '{{vertex_training_name}}',
'Vertex Prediction': '{{vertex_prediction_name}}',
'Vertex AI Prediction': '{{vertex_prediction_name}}',
'Vertex TensorBoard': '{{vertex_tensorboard_name}}',
'Vertex AI TensorBoard': '{{vertex_tensorboard_name}}',
'Vertex ML Metadata': '{{vertex_metadata_name}}',
'Vertex Pipelines': '{{vertex_pipelines_name}}',
'Vertex AI Pipelines': '{{vertex_pipelines_name}}',
'Vertex AI Data Labeling': '{{vertex_data_labeling_name}}',
'Vertex AI Experiments': '{{vertex_experiments_name}}',
'Vertex Experiments': '{{vertex_experiments_name}}',
'Vertex AI Matching Engine': '{{vertex_matching_engine_name}}',
'Vertex Matching Engine': '{{vertex_matching_engine_name}}',
'Vertex Model Monitoring': '{{vertex_model_monitoring_name}}',
'Vertex AI Model Monitoring': '{{vertex_model_monitoring_name}}',
'Vertex Feature Store': '{{vertex_featurestore_name}}',
'Vertex AI Feature Store': '{{vertex_featurestore_name}}',
'Vertex Vizier': '{{vertex_vizier_name}}',
'Vertex AI Vizier': '{{vertex_vizier_name}}',
'Vertex Explainable AI': '{{vertex_xai_name}}',
'NAS': '{{vertex_nas_name}',
'Vertex AI Neural Architectural Search': '{{vertex_nas_name}}',
'Vertex Workbench': '{{vertex_workbench_name}}',
'Vertex AI Workbench': '{{vertex_workbench_name}}',
'Vertex AI Edge Manager': '{{vertex_edge_manager_name}}',
'Vertex SDK': '{{vertex_sdk_name}}',
'Vertex AI SDK': '{{vertex_sdk_name}}',
'Vertex AI': '{{vertex_ai_name}}',
'Cloud Storage': '{{storage_name}}',
'TensorFlow Enterprise': '{{tf4gcp_name}}',
'TensorFlow': '{{tensorflow_name}}',
}
for key, value in substitutions.items():
if key in text:
text = text.replace(key, value)
return text
# Instantiate the rules
@@ -1172,9 +1313,14 @@ if args.web:
print('}')
print('</style>')
print('<table>')
print(' <th width="180px">Services</th>')
print(' <th>Description</th>')
print(' <th width="80px">Open in</th>')
print(' <thead>')
print(' <tr>')
print(' <th width="180px">Services</th>')
print(' <th>Description</th>')
print(' <th width="80px">Open in</th>')
print(' </tr>')
print(' </thead>')
print(' <tbody class="list">')
if args.notebook_dir:
if not os.path.isdir(args.notebook_dir):
@@ -1210,6 +1356,7 @@ else:
exit(1)
if args.web:
print(' </tbody>\n')
print('</table>\n')
exit(exit_code)
@@ -23,7 +23,62 @@
- Incorrect examples: "Let's update the field", "We'll update the field", "The user should update the field"
- **Googlers**: Please follow our [branding guidelines](http://goto/cloud-branding).
### Code
## Authoring guidelines
### Focus
Notebooks for official are expected to be narrow focused, which highlight a subset of features of a Vertex AI product/service.
The product/feature is to be highlighted in the Overview section. For example:
```
This tutorial demonstrates using Vertex AI Training to train an XGBoost model using a XGBoost pre-built training container.
```
In the above example, the Vertex AI product/service is `Vertex AI Training` and the feature is `XGBoost pre-built training container`.
### Scope
Notebooks for official are expected to be narrow in scope, without extra extraneous steps. For example, if the notebook is about training, we discourage ending the notebook with deploying the model and doing an online/batch prediction. On the later, we recommend a separate notebook about prediction that uses a pretrained model.
#### Training
Notebooks for training should be constructed as follows:
1. If the training script(s) are small, embed them in the notebook and use %writefile to store them locally.
2. If the training script(s) are large, store them in our public bucket: gs://cloud-samples-data/vertex-ai/dataset-management/script, and use !wgets to retrieve and store the script locally.
3. Train the model using the Vertex AI SDK methods for custom training.
4. Preferrable have the service upload the trained model to the Vertex AI Model Registry.
4. Have the script do an evaluation.
5. Retrieve the evaluation metrics and attach them as an artifact to the corresponding entry in the Model Registry.
6. Optionally, download the model artifacts and test locally -- i.e., make a local prediction request.
#### Evaluation
Notebooks for evaluation should be constructed as follows:
1. Use a pretrained model from a public repository.
2. Upload the pretrained model to the Vertex AI Model Registry.
3. Perform a model evaluation.
4. Review the model evaluation.
4. Attach the model evaluation to the corresponding entry in the Model Registry.
#### Prediction
Notebooks for prediction should be constructed as follows:
1. Use a pretrained model from a public repository.
2. If relevant, attach a serving function to the model artifacts.
3. Upload the pretrained model to the Vertex AI Model Registry.
4. For online:<br/>
A. Deploy the model.<br/>
B. Perform an online prediction.</br>
C. Review the result.
5. For batch:<br/>
A. Perform a batch prediction.</br/>
B. Review the result.
## Code
- Put all your installs and imports in a setup section.
- Save the notebook with the Table of Contents open.
@@ -31,7 +86,7 @@
- Follow the [Google Python Style guide](https://github.com/google/styleguide/blob/gh-pages/pyguide.md) and write readable code.
- Keep cells small (max ~20 lines).
## TensorFlow code style
### TensorFlow code style
Use the highest level API that gets the job done (unless the goal is to demonstrate the low level API). For example, when using Tensorflow:
+4 -7
View File
@@ -29,17 +29,14 @@
/pipelines/google_cloud_pipelines_dataproc_tabular @inardini
/automl/automl_forecasting_bqml_arima_plus_comparison.ipynb @TheMichaelHu
/automl/automl_tabular_on_vertex_pipelines.ipynb @helinwang
/custom/custom_training_tensorboard_profiler.ipynb @itseric
/custom/custom_training_tensorboard_profiler.ipynb @gericdong
/custom/get_started_with_vertex_endpoint_and_shared_vm.ipynb @andrewferlitsch
/workbench/spark/spark_sample_notebook.ipynb @bradmiro
/workbench/spark/spark_ml.ipynb @bradmiro
/model_registry/bqml_vertexai_model_registry.ipynb @soheilazangeneh
/workbench/exploratory_data_analysis/explore_data_in_bigquery_with_workbench.ipynb @alokpattani
/model_evaluation/automl_tabular_classification_model_evaluation.ipynb @soheilazangeneh
/model_evaluation/automl_tabular_regression_model_evaluation.ipynb @soheilazangeneh
/tabular_workflows/tabnet_on_vertex_pipelines.ipynb @sakagarwal
/tabular_workflows/wide_and_deep_on_vertex_pipelines.ipynb @sakagarwal
/tabular_workflows/prophet_on_vertex_pipelines.ipynb @TheMichaelHu
/model_evaluation/custom_tabular_classification_model_evaluation.ipynb @soheilazangeneh
/sdk/SDK_FBProphet_Forecasting_Online.ipynb @brianchunkang
/automl/sdk_automl_forecasting_hierarchical_batch.ipynb @ivanmkc
/prediction/custom_batch_prediction_feature_filter.ipynb @soheilazangeneh
/feature_store/feature_store_streaming_ingestion_sdk.ipynb @soheilazangeneh
/pipelines/Train_tabular_models_with_many_frameworks_and_import_to_Vertex_AI_using_Pipelines @Ark-kun
+189 -106
View File
@@ -1,4 +1,5 @@
[AutoML Tabular Training and Prediction](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/automl-tabular-classification.ipynb)
[AutoML Tabular training and prediction](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/automl-tabular-classification.ipynb)
```
Learn how to train and make predictions on an AutoML model based on a tabular dataset.
@@ -13,6 +14,8 @@ The steps performed include the following:
```
&nbsp;&nbsp;&nbsp;Learn more about [Classification for tabular data](https://cloud.google.com/vertex-ai/docs/tabular-data/classification-regression/overview).
[Create, train, and deploy an AutoML text classification model](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/automl-text-classification.ipynb)
@@ -31,97 +34,7 @@ The steps performed include:
```
[AutoML training video classification model for batch prediction](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/sdk_automl_video_classification_batch.ipynb)
```
Learn how to create an AutoML video classification model from a Python script, and then do a batch prediction using the Vertex AI SDK.
The steps performed include:
- Create a Vertex `Dataset` resource.
- Train the model.
- View the model evaluation.
- Make a batch prediction.
```
[AutoML training text entity extraction model for online prediction](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/sdk_automl_text_entity_extraction_online.ipynb)
```
Learn how to create an AutoML text entity extraction model and deploy for online prediction from a Python script using the Vertex SDK.
The steps performed include:
- Create a Vertex `Dataset` resource.
- Train the model.
- View the model evaluation.
- Deploy the `Model` resource to a serving `Endpoint` resource.
- Make a prediction.
- Undeploy the `Model`.
```
[AutoML tabular forecasting model for batch prediction](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/sdk_automl_tabular_forecasting_batch.ipynb)
```
Learn how to create an `AutoML` tabular forecasting model from a Python script, and then do a batch prediction using the Vertex AI SDK.
The steps performed include:
- Create a `Vertex AI Dataset` resource.
- Train an `AutoML` tabular forecasting `Model` resource.
- Obtain the evaluation metrics for the `Model` resource.
- Make a batch prediction.
```
[AutoML training video action recognition model for batch prediction](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/sdk_automl_video_action_recognition_batch.ipynb)
```
Learn how to create an AutoML video action recognition model from a Python script, and then do a batch prediction using the Vertex AI SDK.
The steps performed include:
- Create a `Vertex AI Dataset` resource.
- Train the model.
- View the model evaluation.
- Make a batch prediction.
```
[AutoML Tabular Pipeline](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/automl_tabular_on_vertex_pipelines.ipynb)
```
Learn how to create two regression models using [Vertex Pipelines](https://cloud.
The steps performed are:
- Create a training pipeline that reduces the search space from the default to save time.
- Create a training pipeline that reuses the architecture search results from the previous pipeline to save time.
```
[Training an AutoML text sentiment analysis model for online predictions](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/sdk_automl_text_sentiment_analysis_online.ipynb)
```
Learn how to create an AutoML text sentiment analysis model and deploy it for online predictions from a Python script using the Vertex AI SDK.
The steps performed include:
- Create a `Vertex AI Dataset` resource.
- Create a training job for the AutoML model on the dataset.
- View the model evaluation metrics.
- Deploy the `Vertex AI Model` resource to a serving `Vertex AI Endpoint`.
- Make a prediction request to the deployed model.
- Undeploy the model from endpoint.
- Perform clean up process.
```
&nbsp;&nbsp;&nbsp;Learn more about [Classification for text data](https://cloud.google.com/vertex-ai/docs/training-overview#classification_for_text).
[Compare Vertex AI Forecasting and BigQuery ML ARIMA_PLUS](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/automl_forecasting_bqml_arima_plus_comparison.ipynb)
@@ -141,6 +54,114 @@ The steps performed are:
```
&nbsp;&nbsp;&nbsp;Learn more about [BQML ARIMA+ forecasting for tabular data](https://cloud.google.com/vertex-ai/docs/tabular-data/forecasting-arima/overview).
[AutoML Tabular Workflow pipelines](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/automl_tabular_on_vertex_pipelines.ipynb)
```
Learn how to create two regression models using [Vertex AI Pipelines](https://cloud.
The steps performed are:
- Create a training pipeline that reduces the search space from the default to save time.
- Create a training pipeline that reuses the architecture search results from the previous pipeline to save time.
```
&nbsp;&nbsp;&nbsp;Learn more about [Tabular Workflow for E2E AutoML](https://cloud.google.com/vertex-ai/docs/tabular-data/tabular-workflows/e2e-automl).
[Get started with AutoML Training](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/get_started_automl_training.ipynb)
```
Learn how to use `AutoML` for training with `Vertex AI`.
The steps performed include:
- Train an image model
- Export the image model as an edge model
- Train a tabular model
- Export the tabular model as a cloud model
- Train a text model
- Train a video model
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI for AutoML](https://cloud.google.com/vertex-ai/docs/start/automl-users).
[AutoML training hierarchical forecasting for batch prediction](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/sdk_automl_forecasting_hierarchical_batch.ipynb)
```
In this tutorial, you create an AutoML hierarchical forecasting model and deploy it for batch prediction using the Vertex AI SDK for Python.
The steps performed include:
- Create a Vertex AI `TimeSeriesDataset` resource.
- Train the model.
- View the model evaluation.
- Deploy the `Model` resource to a serving `Endpoint` resource.
- Make a prediction.
- Undeploy the `Model`.
```
&nbsp;&nbsp;&nbsp;Learn more about [Hierarchical forecasting for tabular data](https://cloud.google.com/vertex-ai/docs/tabular-data/forecasting/hierarchical).
[AutoML training image object detection model for batch prediction](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/sdk_automl_image_object_detection_batch.ipynb)
```
In this tutorial, you create an AutoML image object detection model from a Python script, and then do a batch prediction using the Vertex AI SDK.
The steps performed include:
- Create a Vertex `Dataset` resource.
- Train the model.
- View the model evaluation.
- Make a batch prediction.
```
&nbsp;&nbsp;&nbsp;Learn more about [Object detection for image data](https://cloud.google.com/vertex-ai/docs/training-overview#object_detection_for_images).
[AutoML tabular forecasting model for batch prediction](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/sdk_automl_tabular_forecasting_batch.ipynb)
```
Learn how to create an `AutoML` tabular forecasting model from a Python script, and then do a batch prediction using the Vertex AI SDK.
The steps performed include:
- Create a `Vertex AI Dataset` resource.
- Train an `AutoML` tabular forecasting `Model` resource.
- Obtain the evaluation metrics for the `Model` resource.
- Make a batch prediction.
```
&nbsp;&nbsp;&nbsp;Learn more about [Forecasting for tabular data](https://cloud.google.com/vertex-ai/docs/tabular-data/forecasting/overview).
[AutoML training tabular regression model for batch prediction using BigQuery](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/sdk_automl_tabular_regression_batch_bq.ipynb)
```
Learn how to create an AutoML tabular regression model and deploy it for batch prediction using the Vertex AI SDK for Python.
The steps performed include:
- Create a Vertex AI `Dataset` resource.
- Train the model.
- View the model evaluation.
- Deploy the `Model` resource to a serving `Endpoint` resource.
- Make a prediction.
- Undeploy the `Model`.
```
&nbsp;&nbsp;&nbsp;Learn more about [Regression for tabular data](https://cloud.google.com/vertex-ai/docs/tabular-data/classification-regression/overview).
[AutoML training tabular regression model for online prediction using BigQuery](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/sdk_automl_tabular_regression_online_bq.ipynb)
@@ -158,6 +179,81 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Regression for tabular data](https://cloud.google.com/vertex-ai/docs/tabular-data/classification-regression/overview).
[AutoML training text entity extraction model for online prediction](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/sdk_automl_text_entity_extraction_online.ipynb)
```
Learn how to create an AutoML text entity extraction model and deploy for online prediction from a Python script using the Vertex AI SDK.
The steps performed include:
- Create a Vertex `Dataset` resource.
- Train the model.
- View the model evaluation.
- Deploy the `Model` resource to a serving `Endpoint` resource.
- Make a prediction.
- Undeploy the `Model`.
```
&nbsp;&nbsp;&nbsp;Learn more about [Entity extraction for text data](https://cloud.google.com/vertex-ai/docs/training-overview#entity_extraction_for_text).
[Training an AutoML text sentiment analysis model for online predictions](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/sdk_automl_text_sentiment_analysis_online.ipynb)
```
Learn how to create an AutoML text sentiment analysis model and deploy it for online predictions from a Python script using the Vertex AI SDK.
The steps performed include:
- Create a `Vertex AI Dataset` resource.
- Create a training job for the AutoML model on the dataset.
- View the model evaluation metrics.
- Deploy the `Vertex AI Model` resource to a serving `Vertex AI Endpoint`.
- Make a prediction request to the deployed model.
- Undeploy the model from endpoint.
- Perform clean up process.
```
&nbsp;&nbsp;&nbsp;Learn more about [Sentiment analysis for text data](https://cloud.google.com/vertex-ai/docs/training-overview#sentiment_analysis_for_text).
[AutoML training video action recognition model for batch prediction](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/sdk_automl_video_action_recognition_batch.ipynb)
```
Learn how to create an AutoML video action recognition model from a Python script, and then do a batch prediction using the Vertex AI SDK.
The steps performed include:
- Create a `Vertex AI Dataset` resource.
- Train the model.
- View the model evaluation.
- Make a batch prediction.
```
&nbsp;&nbsp;&nbsp;Learn more about [Action recognition for video data](https://cloud.google.com/vertex-ai/docs/training-overview#action_recognition_for_videos).
[AutoML training video classification model for batch prediction](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/sdk_automl_video_classification_batch.ipynb)
```
Learn how to create an AutoML video classification model from a Python script, and then do a batch prediction using the Vertex AI SDK.
The steps performed include:
- Create a Vertex `Dataset` resource.
- Train the model.
- View the model evaluation.
- Make a batch prediction.
```
&nbsp;&nbsp;&nbsp;Learn more about [Classification for video data](https://cloud.google.com/vertex-ai/docs/training-overview#classification_for_videos).
[AutoML training video object tracking model for batch prediction](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/sdk_automl_video_object_tracking_batch.ipynb)
@@ -170,21 +266,8 @@ The steps performed include:
- Train the model.
- View the model evaluation.
- Make a batch prediction.
```
[AutoML training tabular regression model for batch prediction using BigQuery](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/automl/sdk_automl_tabular_regression_batch_bq.ipynb)
```
Learn how to create an AutoML tabular regression model and deploy it for batch prediction using the Vertex AI SDK for Python.
The steps performed include:
&nbsp;&nbsp;&nbsp;Learn more about [Object tracking for video data](https://cloud.google.com/vertex-ai/docs/training-overview#object_tracking_for_videos).
- Create a Vertex AI `Dataset` resource.
- Train the model.
- View the model evaluation.
- Deploy the `Model` resource to a serving `Endpoint` resource.
- Make a prediction.
- Undeploy the `Model`.
```
@@ -65,7 +65,7 @@
"\n",
"**Note**: you may incur charges for training, prediction, storage, or usage of other Google Cloud products in connection with testing this SDK.\n",
"\n",
"Learn more about [AutoML Tabular](https://cloud.google.com/vertex-ai/docs/training-overview#tabular_data)."
"Learn more about [Classification for tabular data](https://cloud.google.com/vertex-ai/docs/tabular-data/classification-regression/overview)."
]
},
{
@@ -70,7 +70,7 @@
"\n",
"This notebook walks you through the major phases of building and using an AutoML text classification model on [Vertex AI](https://cloud.google.com/vertex-ai/docs/). \n",
"\n",
"Learn more about [AutoML Text classification](https://cloud.google.com/vertex-ai/docs/text-data/classification/train-model)."
"Learn more about [Classification for text data](https://cloud.google.com/vertex-ai/docs/training-overview#classification_for_text)."
]
},
{
@@ -63,7 +63,7 @@
"\n",
"In this tutorial, you take on the role of a store planner who must determine how much inventory they will need to order for each of their products and stores for November 2019. You accomplish this by training forecasting models using historical sales data. You start with a baseline model using BigQuery ML (BQML) [ARIMA_PLUS](https://cloud.google.com/bigquery-ml/docs/reference/standard-sql/bigqueryml-syntax-create-time-series) and then compare it against a [Vertex AI Forecasting](https://cloud.google.com/vertex-ai/docs/tabular-data/forecasting/overview) model.\n",
"\n",
"Learn more about [AutoML Forecasting](https://cloud.google.com/vertex-ai/docs/tabular-data/forecasting/overview) and [BQML Forecasting](https://cloud.google.com/vertex-ai/docs/tabular-data/forecasting-arima/overview)."
"Learn more about [BQML ARIMA+ forecasting for tabular data](https://cloud.google.com/vertex-ai/docs/tabular-data/forecasting-arima/overview)."
]
},
{
@@ -63,7 +63,7 @@
"\n",
"In this tutorial, you will use two Vertex AI Tabular Workflows pipelines to train AutoML models using different configurations. You will see how `get_automl_tabular_pipeline_and_parameters` gives you the ability to customize the default AutoML Tabular pipeline, and how `get_skip_architecture_search_pipeline_and_parameters` allows you to reduce the training time and cost for an AutoML model by using the tuning results from a previous pipeline run.\n",
"\n",
"Learn more about [AutoML Tabular Workflows](https://cloud.google.com/vertex-ai/docs/tabular-data/tabular-workflows/e2e-automl)."
"Learn more about [Tabular Workflow for E2E AutoML](https://cloud.google.com/vertex-ai/docs/tabular-data/tabular-workflows/e2e-automl)."
]
},
{
@@ -170,7 +170,7 @@
},
"outputs": [],
"source": [
"!pip install -U google-cloud-pipeline-components -q"
"!pip install -U google-cloud-pipeline-components==1.0.25 -q"
]
},
{
@@ -333,7 +333,10 @@
},
"outputs": [],
"source": [
"REGION = \"[your-region]\" # @param {type: \"string\"}"
"REGION = \"us-central1\" # @param {type: \"string\"}\n",
"\n",
"if REGION == \"[your-region]\":\n",
" REGION = \"us-central1\""
]
},
{
@@ -379,8 +382,11 @@
"import os\n",
"import sys\n",
"\n",
"# If on Google Cloud Notebook, then don't execute this code\n",
"if not os.path.exists(\"/opt/deeplearning/metadata/env_version\"):\n",
"# If on Vertex AI Workbench, then don't execute this code\n",
"IS_COLAB = \"google.colab\" in sys.modules\n",
"if not os.path.exists(\"/opt/deeplearning/metadata/env_version\") and not os.getenv(\n",
" \"DL_ANACONDA_HOME\"\n",
"):\n",
" if \"google.colab\" in sys.modules:\n",
" from google.colab import auth as google_auth\n",
"\n",
@@ -388,10 +394,9 @@
"\n",
" # If you are running this notebook locally, replace the string below with the\n",
" # path to your service account key and run this cell to authenticate your GCP\n",
" # account. Alternatively, you may edit this notebook to authenticate using\n",
" # gcloud.\n",
" # account.\n",
" elif not os.getenv(\"IS_TESTING\"):\n",
" %env GOOGLE_APPLICATION_CREDENTIALS ''"
" %env GOOGLE_APPLICATION_CREDENTIALS '[your-service-account-key-path]'"
]
},
{
@@ -478,6 +483,77 @@
"! gsutil ls -al $BUCKET_URI"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "44accda192d5"
},
"source": [
"#### Service Account\n",
"\n",
"You use a service account to create Vertex AI Pipeline jobs. If you do not want to use your project's Compute Engine service account, set `SERVICE_ACCOUNT` to another service account ID."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "e0c9c4f84849"
},
"outputs": [],
"source": [
"SERVICE_ACCOUNT = \"[your-service-account]\""
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "604ae09ab6d3"
},
"outputs": [],
"source": [
"if (\n",
" SERVICE_ACCOUNT == \"\"\n",
" or SERVICE_ACCOUNT is None\n",
" or SERVICE_ACCOUNT == \"[your-service-account]\"\n",
"):\n",
" # Get your service account from gcloud\n",
" if not IS_COLAB:\n",
" shell_output = !gcloud auth list 2>/dev/null\n",
" SERVICE_ACCOUNT = shell_output[2].replace(\"*\", \"\").strip()\n",
"\n",
" else: # IS_COLAB:\n",
" shell_output = ! gcloud projects describe $PROJECT_ID\n",
" project_number = shell_output[-1].split(\":\")[1].strip().replace(\"'\", \"\")\n",
" SERVICE_ACCOUNT = f\"{project_number}-compute@developer.gserviceaccount.com\"\n",
"\n",
" print(\"Service Account:\", SERVICE_ACCOUNT)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "d1ecb60964d5"
},
"source": [
"#### Set service account access for Vertex AI Pipelines\n",
"Run the following commands to grant your service account access to read and write pipeline artifacts in the bucket that you created in the previous step. You only need to run this step once per service account."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "a592f0a380c2"
},
"outputs": [],
"source": [
"! gsutil iam ch serviceAccount:{SERVICE_ACCOUNT}:roles/storage.objectCreator $BUCKET_URI\n",
"\n",
"! gsutil iam ch serviceAccount:{SERVICE_ACCOUNT}:roles/storage.objectViewer $BUCKET_URI"
]
},
{
"cell_type": "markdown",
"metadata": {
@@ -836,6 +912,7 @@
"\n",
"job.run()\n",
"\n",
"\n",
"pipeline_task_details = job.gca_resource.job_detail.task_details\n",
"\n",
"if export_additional_model_without_custom_ops:\n",
@@ -877,6 +954,7 @@
"stage_1_tuner_task = get_task_detail(\n",
" pipeline_task_details, \"automl-tabular-stage-1-tuner\"\n",
")\n",
"\n",
"stage_1_tuning_result_artifact_uri = (\n",
" stage_1_tuner_task.outputs[\"tuning_result_output\"].artifacts[0].uri\n",
")"
File diff suppressed because it is too large Load Diff
@@ -63,7 +63,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK for Python to create hierarchical forecasting models using a Google Cloud [AutoML](https://cloud.google.com/vertex-ai/docs/start/automl-users)and do batch prediction. Specifically, you predict a fictional store's sales based on historical sales data.\n",
"\n",
"Learn more about [AutoML Forecasting](https://cloud.google.com/vertex-ai/docs/tabular-data/forecasting/overview)."
"Learn more about [Hierarchical forecasting for tabular data](https://cloud.google.com/vertex-ai/docs/tabular-data/forecasting/hierarchical)."
]
},
{
@@ -65,7 +65,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK to create image object detection models and do batch prediction using a Google Cloud [AutoML](https://cloud.google.com/vertex-ai/docs/start/automl-users) model.\n",
"\n",
"Learn more about [AutoML Image](https://cloud.google.com/vertex-ai/docs/image-data/object-detection/train-model)."
"Learn more about [Object detection for image data](https://cloud.google.com/vertex-ai/docs/training-overview#object_detection_for_images)."
]
},
{
@@ -64,7 +64,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK to create tabular forecasting models and do batch prediction using a Google Cloud [AutoML](https://cloud.google.com/vertex-ai/docs/start/automl-users) model.\n",
"\n",
"Learn more about [AutoML Forecasting](https://cloud.google.com/vertex-ai/docs/tabular-data/forecasting/tutorials-samples)."
"Learn more about [Forecasting for tabular data](https://cloud.google.com/vertex-ai/docs/tabular-data/forecasting/overview)."
]
},
{
@@ -833,6 +833,7 @@
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {
"id": "99b7a9287ba6"
@@ -840,7 +841,7 @@
"source": [
"For AutoML models, manual scaling can be adjusted by setting both min and max nodes i.e., `starting_replica_count` and `max_replica_count` as the same value(in this example, set to 1). The node count can be increased or decreased as required by load.\n",
" \n",
"`batch_predict` can export predictions either to BigQuery or GCS. The BigQuery options are commented out below and the predictions will be exported to the BUCKET_URI."
"`batch_predict` can export predictions either to BigQuery or GCS. This example exports to BigQuery."
]
},
{
@@ -63,7 +63,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK for Python to create tabular regression models and do batch prediction using a Google Cloud [AutoML](https://cloud.google.com/vertex-ai/docs/start/automl-users) model.\n",
"\n",
"Learn more about [AutoML Tabular](https://cloud.google.com/vertex-ai/docs/training-overview#tabular_data)."
"Learn more about [Regression for tabular data](https://cloud.google.com/vertex-ai/docs/tabular-data/classification-regression/overview)."
]
},
{
@@ -64,7 +64,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK for Python to create tabular regression models and do online prediction using a Google Cloud [AutoML](https://cloud.google.com/vertex-ai/docs/start/automl-users) model.\n",
"\n",
"Learn more about [AutoML Tabular](https://cloud.google.com/vertex-ai/docs/training-overview#tabular_data)."
"Learn more about [Regression for tabular data](https://cloud.google.com/vertex-ai/docs/tabular-data/classification-regression/overview)."
]
},
{
@@ -64,7 +64,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK for Python to create text entity extraction models and do online prediction using a Google Cloud [AutoML](https://cloud.google.com/vertex-ai/docs/start/automl-users) model.\n",
"\n",
"Learn more about [AutoML Text](https://cloud.google.com/vertex-ai/docs/text-data/entity-extraction/train-model)."
"Learn more about [Entity extraction for text data](https://cloud.google.com/vertex-ai/docs/training-overview#entity_extraction_for_text)."
]
},
{
@@ -63,7 +63,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK to train and deploy an [AutoML](https://cloud.google.com/vertex-ai/docs/start/automl-users) text sentiment analysis model and get online predictions from it.\n",
"\n",
"Learn more about [AutoML Text](https://cloud.google.com/vertex-ai/docs/text-data/sentiment-analysis/train-model)."
"Learn more about [Sentiment analysis for text data](https://cloud.google.com/vertex-ai/docs/training-overview#sentiment_analysis_for_text)."
]
},
{
@@ -65,7 +65,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK to create video action recognition models and do batch prediction using a Google Cloud [AutoML](https://cloud.google.com/vertex-ai/docs/start/automl-users) model.\n",
"\n",
"Learn more about [AutoML Video](https://cloud.google.com/vertex-ai/docs/video-data/action-recognition/train-model)."
"Learn more about [Action recognition for video data](https://cloud.google.com/vertex-ai/docs/training-overview#action_recognition_for_videos)."
]
},
{
@@ -64,7 +64,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK to create video classification models and do batch prediction using a Google Cloud [AutoML](https://cloud.google.com/vertex-ai/docs/start/automl-users) model.\n",
"\n",
"Learn more about [AutoML Video](https://cloud.google.com/vertex-ai/docs/video-data/classification/train-model)."
"Learn more about [Classification for video data](https://cloud.google.com/vertex-ai/docs/training-overview#classification_for_videos)."
]
},
{
@@ -64,7 +64,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK for Python to create video object tracking models and do batch prediction using a Google Cloud [AutoML](https://cloud.google.com/vertex-ai/docs/start/automl-users) model.\n",
"\n",
"Learn more about [AutoML Video](https://cloud.google.com/vertex-ai/docs/video-data/object-tracking/train-model)."
"Learn more about [Object tracking for video data](https://cloud.google.com/vertex-ai/docs/training-overview#object_tracking_for_videos)."
]
},
{
+23
View File
@@ -1,3 +1,4 @@
[Online prediction with BigQuery ML](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/bigquery_ml/bqml-online-prediction.ipynb)
```
@@ -14,3 +15,25 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [BigQuery ML](https://cloud.google.com/vertex-ai/docs/beginner/bqml).
[Get started with BigQuery ML Training](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/bigquery_ml/get_started_with_bqml_training.ipynb)
```
Learn how to use `BigQueryML` for training with `Vertex AI`.
The steps performed include:
- Create a local BigQuery table in your project
- Train a BigQuery ML model
- Evaluate the BigQuery ML model
- Export the BigQuery ML model as a cloud model
- Upload the exported model as a `Vertex AI Model` resource
- Hyperparameter tune a BigQuery ML model with `Vertex AI Vizier`
- Automatically register a BigQuery ML model to `Vertex AI Model Registry`
```
&nbsp;&nbsp;&nbsp;Learn more about [BigQuery ML](https://cloud.google.com/vertex-ai/docs/beginner/bqml).
@@ -63,7 +63,7 @@
"\n",
"This notebook is aimed at data analysts and data scientists who have data in BigQuery, want to train a model using BigQuery ML, register the model to Vertex AI Model Registry, and deploy it to an endpoint for real-time prediction. \n",
"\n",
"Learn more about [BigQuery ML](https://cloud.google.com/bigquery-ml/docs/introduction)."
"Learn more about [BigQuery ML](https://cloud.google.com/vertex-ai/docs/beginner/bqml)."
]
},
{
File diff suppressed because it is too large Load Diff
File diff suppressed because it is too large Load Diff
+62 -41
View File
@@ -1,3 +1,24 @@
[Deploying Iris-detection model using FastAPI and Vertex AI custom container serving](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/custom/SDK_Custom_Container_Prediction.ipynb)
```
Learn how to create, deploy and serve a custom classification model on Vertex AI.
The steps performed include:
- Train a model that uses flower's measurements as input to predict the class of iris.
- Save the model and its serialized pre-processor.
- Build a FastAPI server to handle predictions and health checks.
- Build a custom container with model artifacts.
- Upload and deploy custom container to Vertex AI Endpoints.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Prediction](https://cloud.google.com/vertex-ai/docs/predictions/get-predictions).
[Training and deploying a sales forecasting model using FBProphet and Vertex AI](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/custom/SDK_FBProphet_Forecasting_Online.ipynb)
```
@@ -16,34 +37,9 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training).
[Custom training and batch prediction](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/custom/sdk-custom-image-classification-batch.ipynb)
```
Learn to use `Vertex AI Training` to create a custom trained model and use `Vertex AI Batch Prediction` to do a batch prediction on the trained model.
The steps performed include:
- Create a `Vertex AI` custom job for training a TensorFlow model.
- Upload the trained model artifacts as a `Model` resource.
- Make a batch prediction.
```
[Profile model training performance using Profiler](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/custom/custom_training_tensorboard_profiler.ipynb)
```
Learn how to enable Vertex AI TensorBoard Profiler for custom training jobs.
The steps performed include:
- Setup a service account and a Cloud Storage bucket
- Create a TensorBoard instance
- Create and run a custom training job
- View the TensorBoard Profiler dashboard
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Prediction](https://cloud.google.com/vertex-ai/docs/predictions/get-predictions).
[Training a TensorFlow model on BigQuery data](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/custom/custom-tabular-bq-managed-dataset.ipynb)
@@ -61,6 +57,43 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training).
[Profile model training performance using Profiler](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/custom/custom_training_tensorboard_profiler.ipynb)
```
Learn how to enable Vertex AI TensorBoard Profiler for custom training jobs.
The steps performed include:
- Setup a service account and a Cloud Storage bucket
- Create a TensorBoard instance
- Create and run a custom training job
- View the TensorBoard Profiler dashboard
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI TensorBoard Profiler](https://cloud.google.com/vertex-ai/docs/experiments/tensorboard-profiler).
[Custom training and batch prediction](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/custom/sdk-custom-image-classification-batch.ipynb)
```
Learn to use `Vertex AI Training` to create a custom trained model and use `Vertex AI Batch Prediction` to do a batch prediction on the trained model.
The steps performed include:
- Create a `Vertex AI` custom job for training a TensorFlow model.
- Upload the trained model artifacts as a `Model` resource.
- Make a batch prediction.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Batch Prediction](https://cloud.google.com/vertex-ai/docs/tabular-data/classification-regression/get-batch-predictions).
[Custom training and online prediction](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/custom/sdk-custom-image-classification-online.ipynb)
@@ -78,19 +111,7 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training).
[Deploying Iris-detection model using FastAPI and Vertex AI custom container serving](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/custom/SDK_Custom_Container_Prediction.ipynb)
```
Learn how to create, deploy and serve a custom classification model on Vertex AI.
The steps performed include:
- Train a model that uses flower's measurements as input to predict the class of iris.
- Save the model and its serialized pre-processor.
- Build a FastAPI server to handle predictions and health checks.
- Build a custom container with model artifacts.
- Upload and deploy custom container to Vertex AI Endpoints.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Prediction](https://cloud.google.com/vertex-ai/docs/predictions/get-predictions).
@@ -60,7 +60,7 @@
"\n",
"This tutorial walks you through building a custom container to serve a scikit-learn model on Vertex AI. You use the FastAPI Python web server framework to create a prediction and health endpoint. You also incorporate a pre-processor from training pipeline into your online serving application.\n",
"\n",
"Learn more about [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training) and [Vertex AI Prediction](https://cloud.google.com/vertex-ai/docs/predictions/get-predictions)."
"Learn more about [Custom training](https://cloud.google.com/vertex-ai/docs/training/custom-training) and [Vertex AI Prediction](https://cloud.google.com/vertex-ai/docs/predictions/get-predictions)."
]
},
{
@@ -64,7 +64,7 @@
"\n",
"Learn more about serving an FBProphet model from this [article on testdriven.io: Deploying and Hosting a Machine Learning Model with FastAPI and Heroku](https://testdriven.io/blog/fastapi-machine-learning/).\n",
"\n",
"Learn more about [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training) and [Vertex AI Prediction](https://cloud.google.com/vertex-ai/docs/predictions/get-predictions).\n"
"Learn more about [Custom training](https://cloud.google.com/vertex-ai/docs/training/custom-training) and [Vertex AI Prediction](https://cloud.google.com/vertex-ai/docs/predictions/get-predictions).\n"
]
},
{
File diff suppressed because it is too large Load Diff
@@ -64,7 +64,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK for Python to train and deploy a custom image classification model for batch prediction.\n",
"\n",
"Learn more about [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training) and [Vertex AI Batch Prediction](https://cloud.google.com/vertex-ai/docs/tabular-data/classification-regression/get-batch-predictions)."
"Learn more about [Custom training](https://cloud.google.com/vertex-ai/docs/training/custom-training) and [Vertex AI Batch Prediction](https://cloud.google.com/vertex-ai/docs/tabular-data/classification-regression/get-batch-predictions)."
]
},
{
@@ -64,7 +64,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK for Python to train and deploy a custom image classification model for online prediction.\n",
"\n",
"Learn more about [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training) and [Vertex AI Prediction](https://cloud.google.com/vertex-ai/docs/predictions/get-predictions)."
"Learn more about [Custom training](https://cloud.google.com/vertex-ai/docs/training/custom-training) and [Vertex AI Prediction](https://cloud.google.com/vertex-ai/docs/predictions/get-predictions)."
]
},
{
+38
View File
@@ -0,0 +1,38 @@
[Get started with BigQuery datasets](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/datasets/get_started_bq_datasets.ipynb)
```
Learn how to use `BigQuery` as a dataset for training with `Vertex AI`.
The steps performed include:
- Create a Vertex AI `Dataset` resource from `BigQuery` table -- compatible for `AutoML` training.
- Extract a copy of the dataset from `BigQuery` to a CSV file in Cloud Storage -- compatible for `AutoML` or custom training.
- Select rows from a `BigQuery` dataset into a `pandas` dataframe -- compatible for custom training.
- Select rows from a `BigQuery` dataset into a `tf.data.Dataset` -- compatible for custom training `TensorFlow` models.
- Select rows from extracted CSV files into a `tf.data.Dataset` -- compatible for custom training `TensorFlow` models.
- Create a `BigQuery` dataset from CSV files.
- Extract data from `BigQuery` table into a `DMatrix` -- compatible for custom training `XGBoost` models.
```
&nbsp;&nbsp;&nbsp;Learn more about [BigQuery Datasets](https://cloud.google.com/bigquery/docs/datasets-intro).
[Get started with Vertex AI Data Labeling](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/datasets/get_started_with_data_labeling.ipynb)
```
Learn how to use the `Vertex AI Data Labeling` service.
The steps performed include:
- Create a Specialist Pool for data labelers.
- Create a data labeling job.
- Submit the data labeling job.
- List data labeling jobs.
- Cancel a data labeling job.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Data Labeling](https://cloud.google.com/vertex-ai/docs/datasets/data-labeling-job).
@@ -29,7 +29,7 @@
"id": "title:generic,gcp"
},
"source": [
"# E2E ML on GCP: MLOps stage 1 : data management: get started with BigQuery datasets\n",
"# Get started with BigQuery datasets\n",
"\n",
"<table align=\"left\">\n",
" <td>\n",
@@ -61,7 +61,9 @@
"## Overview\n",
"\n",
"\n",
"This tutorial demonstrates how to use Vertex AI for E2E MLOps on Google Cloud in production. This tutorial covers stage 1 : data management: get started with BigQuery datasets."
"This tutorial demonstrates how to use Vertex AI in production. This tutorial covers data management: get started with BigQuery datasets.\n",
"\n",
"Learn more about [BigQuery Datasets](https://cloud.google.com/bigquery/docs/datasets-intro)."
]
},
{
File diff suppressed because it is too large Load Diff
+60 -8
View File
@@ -1,17 +1,25 @@
[Compare pipeline runs with Vertex AI Experiments](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/experiments/comparing_pipeline_runs.ipynb)
[Build Vertex AI Experiment lineage for custom training](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/experiments/build_model_experimentation_lineage_with_prebuild_code.ipynb)
```
Learn how to use `Vertex AI Experiments` to log a pipeline job and compare different pipeline jobs.
Learn how to integrate preprocessing code in a Vertex AI experiments.
The steps performed include:
* Formalize a training component
* Build a training pipeline
* Run several Pipeline jobs and log their results
* Compare different Pipeline jobs
- Execute module for preprocessing data
- Create a dataset artifact
- Log parameters
- Execute module for training the model
- Log parameters
- Create model artifact
- Assign tracking lineage to dataset, model and parameters
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Experiments](https://cloud.google.com/vertex-ai/docs/experiments/intro-vertex-ai-experiments).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex ML Metadata](https://cloud.google.com/vertex-ai/docs/ml-metadata).
[Track parameters and metrics for locally trained models](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/experiments/comparing_local_trained_models.ipynb)
@@ -26,13 +34,57 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Experiments](https://cloud.google.com/vertex-ai/docs/experiments/intro-vertex-ai-experiments).
[Build Vertex AI Experiment lineage for custom training](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/experiments/build_model_experimentation_lineage_with_prebuild_code.ipynb)
[Compare pipeline runs with Vertex AI Experiments](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/experiments/comparing_pipeline_runs.ipynb)
```
Learn how to integrate preprocessing code in a Vertex AI experiments.
Learn how to use `Vertex AI Experiments` to log a pipeline job and compare different pipeline jobs.
The steps performed include:
* Formalize a training component
* Build a training pipeline
* Run several Pipeline jobs and log their results
* Compare different Pipeline jobs
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Experiments](https://cloud.google.com/vertex-ai/docs/experiments/intro-vertex-ai-experiments).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Pipelines](https://cloud.google.com/vertex-ai/docs/pipelines/introduction).
[Get started with Vertex AI Experiments](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/experiments/get_started_with_vertex_experiments.ipynb)
```
Learn how to use `Vertex AI Experiments` when training with `Vertex AI`.
The steps performed include:
- Local (notebook) Training
- Create an experiment
- Create a first run in the experiment
- Log parameters and metrics
- Create artifact lineage
- Visualize the experiment results
- Execute a second run
- Compare the two runs in the experiment
- Cloud (`Vertex AI`) Training
- Within the training script:
- Create an experiment
- Log parameters and metrics
- Create artifact lineage
- Create a `Vertex AI Training` custom job
- Execute the custom job
- Visualize the experiment results
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Experiments](https://cloud.google.com/vertex-ai/docs/experiments/intro-vertex-ai-experiments).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex ML Metadata](https://cloud.google.com/vertex-ai/docs/ml-metadata).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training).
File diff suppressed because it is too large Load Diff
+59 -30
View File
@@ -1,18 +1,3 @@
[Custom training tabular regression model for batch prediction with explainabilty](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/explainable_ai/sdk_custom_tabular_regression_batch_explain.ipynb)
```
Learn how to use `Vertex AI Training and Explainable AI` to create a custom image classification model with explanations, and then you learn to use `Vertex AI Batch Prediction` to make a batch prediction request with explanations.
The steps performed include:
- Create a `Vertex AI` custom job for training a TensorFlow model.
- View the model evaluation for the trained model.
- Set explanation parameters for when the model is deployed.
- Upload the trained model artifacts and explanations as a `Model` resource.
- Make a batch prediction with explanations.
```
[AutoML training tabular binary classification model for batch explanation](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/explainable_ai/sdk_automl_tabular_binary_classification_batch_explain.ipynb)
@@ -28,6 +13,10 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [AutoML Tabular](https://cloud.google.com/vertex-ai/docs/tabular-data/overview).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex Explainable AI](https://cloud.google.com/vertex-ai/docs/explainable-ai/overview).
[AutoML training tabular classification model for online explanation](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/explainable_ai/sdk_automl_tabular_classification_online_explain.ipynb)
@@ -46,6 +35,30 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [AutoML Tabular](https://cloud.google.com/vertex-ai/docs/tabular-data/overview).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex Explainable AI](https://cloud.google.com/vertex-ai/docs/explainable-ai/overview).
[Custom training image classification model for batch prediction with explainabilty](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/explainable_ai/sdk_custom_image_classification_batch_explain.ipynb)
```
Learn to use `Vertex AI Training and Explainable AI` to create a custom image classification model with explanations, and then you learn to use `Vertex AI Batch Prediction` to make a batch prediction request with explanations.
The steps performed include:
- Create a `Vertex AI` custom job for training a TensorFlow model.
- View the model evaluation for the trained model.
- Set explanation parameters for when the model is deployed.
- Upload the trained model artifacts and explanation parameters as a `Model` resource.
- Make a batch prediction with explanations.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex Explainable AI](https://cloud.google.com/vertex-ai/docs/explainable-ai/overview).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Batch Prediction](https://cloud.google.com/vertex-ai/docs/tabular-data/classification-regression/get-batch-predictions).
[Custom training image classification model for online prediction with explainabilty](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/explainable_ai/sdk_custom_image_classification_online_explain.ipynb)
@@ -65,6 +78,30 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex Explainable AI](https://cloud.google.com/vertex-ai/docs/explainable-ai/overview).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Prediction](https://cloud.google.com/vertex-ai/docs/predictions/get-predictions).
[Custom training tabular regression model for batch prediction with explainabilty](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/explainable_ai/sdk_custom_tabular_regression_batch_explain.ipynb)
```
Learn how to use `Vertex AI Training and Explainable AI` to create a custom image classification model with explanations, and then you learn to use `Vertex AI Batch Prediction` to make a batch prediction request with explanations.
The steps performed include:
- Create a `Vertex AI` custom job for training a TensorFlow model.
- View the model evaluation for the trained model.
- Set explanation parameters for when the model is deployed.
- Upload the trained model artifacts and explanations as a `Model` resource.
- Make a batch prediction with explanations.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex Explainable AI](https://cloud.google.com/vertex-ai/docs/explainable-ai/overview).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Batch Prediction](https://cloud.google.com/vertex-ai/docs/tabular-data/classification-regression/get-batch-predictions).
[Custom training tabular regression model for online prediction with explainabilty](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/explainable_ai/sdk_custom_tabular_regression_online_explain.ipynb)
@@ -84,11 +121,15 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex Explainable AI](https://cloud.google.com/vertex-ai/docs/explainable-ai/overview).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Prediction](https://cloud.google.com/vertex-ai/docs/predictions/get-predictions).
[Custom training tabular regression model for online prediction with explainabilty using get_metadata](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/explainable_ai/sdk_custom_tabular_regression_online_explain_get_metadata.ipynb)
```
Learn how to create a custom model from a Python script in a Google prebuilt Docker container using the Vertex SDK, and then do a prediction with explanations on the deployed model by sending data.
Learn how to create a custom model from a Python script in a Google prebuilt Docker container using the Vertex AI SDK, and then do a prediction with explanations on the deployed model by sending data.
The steps performed include:
@@ -104,19 +145,7 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex Explainable AI](https://cloud.google.com/vertex-ai/docs/explainable-ai/overview).
[Custom training image classification model for batch prediction with explainabilty](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/explainable_ai/sdk_custom_image_classification_batch_explain.ipynb)
```
Learn to use `Vertex AI Training and Explainable AI` to create a custom image classification model with explanations, and then you learn to use `Vertex AI Batch Prediction` to make a batch prediction request with explanations.
The steps performed include:
- Create a `Vertex AI` custom job for training a TensorFlow model.
- View the model evaluation for the trained model.
- Set explanation parameters for when the model is deployed.
- Upload the trained model artifacts and explanation parameters as a `Model` resource.
- Make a batch prediction with explanations.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Prediction](https://cloud.google.com/vertex-ai/docs/predictions/get-predictions).
@@ -64,7 +64,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK to create tabular binary classification models and do batch prediction with explanation using a Google Cloud [AutoML](https://cloud.google.com/vertex-ai/docs/start/automl-users) model.\n",
"\n",
"Learn more about [AutoML Tabular](https://cloud.google.com/vertex-ai/docs/tabular-data/overview) and [Vertex Explainable AI](https://cloud.google.com/vertex-ai/docs/explainable-ai/overview)."
"Learn more about [Classification for tabular data](https://cloud.google.com/vertex-ai/docs/tabular-data/classification-regression/overview). Learn more about [Vertex Explainable AI](https://cloud.google.com/vertex-ai/docs/explainable-ai/overview)."
]
},
{
File diff suppressed because one or more lines are too long
@@ -1,3 +1,20 @@
[Streaming ingestion SDK](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/feature_store/feature_store_streaming_ingestion_sdk.ipynb)
```
Learn how to ingest features from a `Pandas DataFrame` into your Vertex AI Feature Store using `write_feature_values` method from the Vertex AI SDK.
The steps performed include:
- Create `Feature Store`
- Create new `Entity Type` for your `Feature Store`
- Ingest feature values from `Pandas DataFrame` into `Feature Store`'s `Entity Types`.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Feature Store](https://cloud.google.com/vertex-ai/docs/featurestore).
[Using Vertex AI Feature Store with Pandas Dataframe](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/feature_store/sdk-feature-store-pandas.ipynb)
```
@@ -15,6 +32,8 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Feature Store](https://cloud.google.com/vertex-ai/docs/featurestore).
[Online and Batch predictions using Vertex AI Feature Store](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/feature_store/sdk-feature-store.ipynb)
@@ -27,6 +46,9 @@ The steps performed include:
- Import feature data into `Vertex AI Feature Store` resource.
- Serve online prediction requests using the imported features.
- Access imported features in offline jobs, such as training jobs.
- Use streaming ingestion to ingest small amount of data.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Feature Store](https://cloud.google.com/vertex-ai/docs/featurestore).
+24 -17
View File
@@ -1,18 +1,3 @@
[Create Vertex AI Matching Engine index](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/matching_engine/sdk_matching_engine_for_indexing.ipynb)
```
Learn how to create Approximate Nearest Neighbor (ANN) Index, query against indexes, and validate the performance of the index.
The steps performed include:
* Create ANN Index and Brute Force Index
* Create an IndexEndpoint with VPC Network
* Deploy ANN Index and Brute Force Index
* Perform online query
* Compute recall
```
[Introduction to builtin Swivel embedding algorithm](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/matching_engine/intro-swivel.ipynb)
@@ -30,11 +15,31 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Matching Engine](https://cloud.google.com/vertex-ai/docs/matching-engine/overview).
[Introduction to builtin Two-towers embedding algorithm](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/matching_engine/two-tower-model-introduction.ipynb)
[Create Vertex AI Matching Engine index](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/matching_engine/sdk_matching_engine_for_indexing.ipynb)
```
Learn how to run the two-tower model.
Learn how to create Approximate Nearest Neighbor (ANN) Index, query against indexes, and validate the performance of the index.
The steps performed include:
* Create ANN Index and Brute Force Index
* Create an IndexEndpoint with VPC Network
* Deploy ANN Index and Brute Force Index
* Perform online query
* Compute recall
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Matching Engine](https://cloud.google.com/vertex-ai/docs/matching-engine/overview).
[Introduction to builtin Two-Towers embedding algorithm](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/matching_engine/two-tower-model-introduction.ipynb)
```
Learn how to run the Two-Tower model.
The steps performed include:
1. **Setup**: Importing the required libraries and setting your global variables.
@@ -47,3 +52,5 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Matching Engine](https://cloud.google.com/vertex-ai/docs/matching-engine/overview).
File diff suppressed because it is too large Load Diff
@@ -292,8 +292,7 @@
},
"outputs": [],
"source": [
"# VPC_NETWORK = \"[your-vpc-network-name]\" # @param {type:\"string\"}\n",
"VPC_NETWORK = \"matching-engine-test\" # @param {type:\"string\"}\n",
"VPC_NETWORK = \"[your-vpc-network-name]\" # @param {type:\"string\"}\n",
"\n",
"PEERING_RANGE_NAME = \"ann-haystack-range\""
]
@@ -544,7 +543,7 @@
" json.dumps(\n",
" {\n",
" \"id\": str(index),\n",
" \"embedding\": [str(value) for value in train[index]],\n",
" \"embedding\": [str(value) for value in embedding],\n",
" \"restricts\": [\n",
" {\n",
" \"namespace\": \"class\",\n",
@@ -646,7 +645,7 @@
"outputs": [],
"source": [
"tree_ah_index = aiplatform.MatchingEngineIndex.create_tree_ah_index(\n",
" display_name=DISPLAY_NAME_BRUTE_FORCE,\n",
" display_name=DISPLAY_NAME,\n",
" contents_delta_uri=EMBEDDINGS_INITIAL_URI,\n",
" dimensions=DIMENSIONS,\n",
" approximate_neighbors_count=150,\n",
@@ -712,7 +711,7 @@
"outputs": [],
"source": [
"brute_force_index = aiplatform.MatchingEngineIndex.create_brute_force_index(\n",
" display_name=DISPLAY_NAME,\n",
" display_name=DISPLAY_NAME_BRUTE_FORCE,\n",
" contents_delta_uri=EMBEDDINGS_INITIAL_URI,\n",
" dimensions=DIMENSIONS,\n",
" distance_measure_type=\"DOT_PRODUCT_DISTANCE\",\n",
File diff suppressed because it is too large Load Diff
+194 -20
View File
@@ -1,3 +1,117 @@
[AutoML Image Classification](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/migration/UJ1 Vertex SDK AutoML Image Classification.ipynb)
```
Learn to use `AutoML` to train an image model and use `Vertex AI Prediction` and `Vertex AI Batch Prediction` to do online and batch predictions.
The steps performed include:
- Train an AutoML image classification model.
- Make a batch prediction.
- Deploy model to a endpoint
- Make a online prediction
```
&nbsp;&nbsp;&nbsp;Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai).
&nbsp;&nbsp;&nbsp;Learn more about [AutoML Image](https://cloud.google.com/vertex-ai/docs/tutorials/image-recognition-automl/training).
[Custom Scikit-Learn model with pre-built training container](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/migration/UJ10 Vertex SDK Custom Scikit-Learn with pre-built training container.ipynb)
```
Learn to use `Vertex AI Training` to create a custom trained model and use `Vertex AI Batch Prediction` to do a batch prediction on the trained model.
The steps performed include:
- Create a `Vertex AI` custom job for training a scikit-learn model.
- Upload the trained model artifacts as a `Model` resource.
- Make a batch prediction.
- Deploy model to a endpoint
- Make a online prediction
```
&nbsp;&nbsp;&nbsp;Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training).
[Hyperparameter Tuning](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/migration/UJ11 Vertex SDK Hyperparameter Tuning.ipynb)
```
Learn to use `Vertex AI Hyperparameter` to create and tune a custom trained model.
The steps performed include:
- Create a `Vertex AI` hyperparameter tuning job for training a TensorFlow model.
```
&nbsp;&nbsp;&nbsp;Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/training/hyperparameter-tuning-overview).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training).
[AutoML Video Classificaton](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/migration/UJ14 Vertex SDK AutoML Video Classification.ipynb)
```
Learn to use `AutoML` to train a video model and use `Vertex AI Batch Prediction` to do batch predictions.
The steps performed include:
- Train an AutoML video classification model.
- Make a batch prediction.
```
&nbsp;&nbsp;&nbsp;Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai).
&nbsp;&nbsp;&nbsp;Learn more about [AutoML Video](https://cloud.google.com/vertex-ai/docs/tutorials/video-classification-automl/training).
[AutoML Video Object Tracking](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/migration/UJ15 Vertex SDK AutoML Object Tracking.ipynb)
```
Learn to use `AutoML` to train a video model and use `Vertex AI Batch Prediction` to do batch predictions.
The steps performed include:
- Train an AutoML video object tracking model.
- Make a batch prediction.
```
&nbsp;&nbsp;&nbsp;Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai).
&nbsp;&nbsp;&nbsp;Learn more about [AutoML Video](https://cloud.google.com/video-intelligence/automl/object-tracking/docs/index-object-tracking).
[Custom Image Classification w/pre-built training container](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/migration/UJ2,12 Vertex SDK Custom Image Classification with pre-built training container.ipynb)
```
Learn how to train a tensorflow image classification model using a prebuilt container and Vertex AI training.
The steps performed include:
- *Package the training code into a python application.*
- *Containerize the training application using Cloud Build and Artifact Registry.*
- *Create a custom container training job in Vertex AI and run it.*
- *Evaluate the model generated from the training job.*
- *Create a model resource for the trained model in Vertex AI Model Registry.*
- *Run a Vertex AI batch prediction job.*
- *Deploy the model resource to a Vertex AI Endpoint.*
- *Run a online prediction job on the model resource.*
- *Clean up the resources created.*
```
&nbsp;&nbsp;&nbsp;Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training).
[Custom Image Classification w/custom training container](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/migration/UJ3 Vertex SDK Custom Image Classification with custom training container.ipynb)
```
@@ -17,6 +131,10 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training).
[AutoML Tabular Binary Classification](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/migration/UJ4 Vertex SDK AutoML Tabular Binary Classification.ipynb)
@@ -34,39 +152,28 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai).
[Custom Scikit-Learn model with pre-built training container](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/migration/UJ10 Vertex SDK Custom Scikit-Learn with pre-built training container.ipynb)
&nbsp;&nbsp;&nbsp;Learn more about [AutoML Tabular](https://cloud.google.com/vertex-ai/docs/start/automl-users#tables).
[AutoML Image Object Detection](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/migration/UJ5 Vertex SDK AutoML Image Object Detection.ipynb)
```
Learn to use `Vertex AI Training` to create a custom trained model and use `Vertex AI Batch Prediction` to do a batch prediction on the trained model.
Learn to use `AutoML` to train an image model and use `Vertex AI Prediction` and `Vertex AI Batch Prediction` to do online and batch predictions.
The steps performed include:
- Create a `Vertex AI` custom job for training a scikit-learn model.
- Upload the trained model artifacts as a `Model` resource.
- Train an AutoML object detection model.
- Make a batch prediction.
- Deploy model to a endpoint
- Make a online prediction
```
&nbsp;&nbsp;&nbsp;Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai).
[AutoML Text Entity Extraction](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/migration/UJ7 Vertex SDK AutoML Text Entity Extraction.ipynb)
```
The objective of this notebook is to build a AutoML Text Entity Extrasction Model.
The steps performed include the following:
* Set your task name, and GCS prefix
* Copy AutoML video demo train data for creating managed dataset
* Create a dataset on Vertex AI.
* Configure a training job
* Launch a training job and create a model on Vertex AI
* Copy AutoML Video Demo Prediction Data for creating batch prediction job
* Perform batch prediction job on the model
```
&nbsp;&nbsp;&nbsp;Learn more about [AutoML Image](https://cloud.google.com/vertex-ai/docs/tutorials/image-recognition-automl/training).
[AutoML Text Classification](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/migration/UJ6 Vertex SDK AutoML Text Classification.ipynb)
@@ -86,3 +193,70 @@ The steps performed include the following:
```
&nbsp;&nbsp;&nbsp;Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai).
&nbsp;&nbsp;&nbsp;Learn more about [AutoML Text](https://cloud.google.com/vertex-ai/docs/text-data/classification/prepare-data).
[AutoML Text Entity Extraction](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/migration/UJ7 Vertex SDK AutoML Text Entity Extraction.ipynb)
```
The objective of this notebook is to build a AutoML Text Entity Extraction Model.
The steps performed include the following:
* Set your task name, and GCS prefix
* Copy AutoML video demo train data for creating managed dataset
* Create a dataset on Vertex AI.
* Configure a training job
* Launch a training job and create a model on Vertex AI
* Copy AutoML Video Demo Prediction Data for creating batch prediction job
* Perform batch prediction job on the model
```
&nbsp;&nbsp;&nbsp;Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai).
&nbsp;&nbsp;&nbsp;Learn more about [AutoML Text](https://cloud.google.com/vertex-ai/docs/text-data/entity-extraction/prepare-data).
[AutoML Text Sentiment Analysis](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/migration/UJ8 Vertex SDK AutoML Text Sentiment Analysis.ipynb)
```
The objective of this notebook is to build a AutoML Text Sentiment Analysis model.
The steps performed include the following:
* Copy AutoML video demo train data for creating managed dataset
* Create a dataset on Vertex AI.
* Configure a training job
* Launch a training job and create a model on Vertex AI
* Copy AutoML Video Demo Prediction Data for creating batch prediction job
* Perform batch prediction job on the model
```
&nbsp;&nbsp;&nbsp;Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai).
&nbsp;&nbsp;&nbsp;Learn more about [AutoML Text](https://cloud.google.com/vertex-ai/docs/text-data/sentiment-analysis/prepare-data).
[Custom XGBoost model with pre-built training container](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/migration/UJ9 Vertex SDK Custom XGBoost with pre-built training container.ipynb)
```
Learn to use `Vertex AI Training` to create a custom trained model and use `Vertex AI Batch Prediction` to do a batch prediction on the trained model.
The steps performed include:
- Create a `Vertex AI` custom job for training a scikit-learn model.
- Upload the trained model artifacts as a `Model` resource.
- Make a batch prediction.
- Deploy model to a endpoint
- Make a online prediction
```
&nbsp;&nbsp;&nbsp;Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training).
@@ -64,7 +64,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK for Python to train and deploy an AutoML image classification model.\n",
"\n",
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [AutoML Image](https://cloud.google.com/vertex-ai/docs/tutorials/image-recognition-automl/training)."
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [Classification for image data](https://cloud.google.com/vertex-ai/docs/training-overview#classification_for_images)."
]
},
{
@@ -64,7 +64,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK for Python to train and deploy a custom tabular classification scikit-learn model for batch prediction.\n",
"\n",
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training)."
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [Custom training](https://cloud.google.com/vertex-ai/docs/training/custom-training)."
]
},
{
@@ -65,7 +65,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK for Python to hyperparamer tune a custom tabular classification TemsorFlow model.\n",
"\n",
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/training/hyperparameter-tuning-overview) and [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training)."
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/training/hyperparameter-tuning-overview) and [Custom training](https://cloud.google.com/vertex-ai/docs/training/custom-training)."
]
},
{
@@ -568,7 +568,7 @@
},
"outputs": [],
"source": [
"IMPORT_FILE = \"gs://automl-video-demo-data/hmdb_split1_5classes_train_inf.csv\""
"IMPORT_FILE = \"gs://automl-video-demo-data/hmdb_split1_train_40_mp4_gs.csv\""
]
},
{
@@ -604,6 +604,19 @@
"! gsutil cat $FILE | head"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "f483d8d80f64"
},
"outputs": [],
"source": [
"import datetime\n",
"\n",
"start = datetime.datetime.now()"
]
},
{
"cell_type": "markdown",
"metadata": {
@@ -757,7 +770,19 @@
"\n",
"The `run` method when completed returns the `Model` resource.\n",
"\n",
"The execution of the training pipeline will take upto 20 minutes."
"The execution of the training pipeline may take over 24 hrs."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "8f135100f5d9"
},
"outputs": [],
"source": [
"if os.getenv(\"IS_TESTING\"):\n",
" sys.exit(0)"
]
},
{
@@ -806,6 +831,18 @@
" INFO:google.cloud.aiplatform.training_jobs:Model available at projects/759209241365/locations/us-central1/models/1899701006099283968"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "48ca7f0980e0"
},
"outputs": [],
"source": [
"end = datetime.datetime.now()\n",
"end - start"
]
},
{
"cell_type": "markdown",
"metadata": {
@@ -64,7 +64,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK for Python to train a AutoML video object tracking model and do a batch prediction.\n",
"\n",
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [AutoML Video](https://cloud.google.com/video-intelligence/automl/object-tracking/docs/index-object-tracking)."
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [Object tracking for video data](https://cloud.google.com/vertex-ai/docs/training-overview#object_tracking_for_videos)."
]
},
{
@@ -64,7 +64,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK for Python to train using a pre-built container and deploy a custom image classification model for online and batch prediction.\n",
"\n",
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training)."
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [Custom training](https://cloud.google.com/vertex-ai/docs/training/custom-training)."
]
},
{
@@ -63,7 +63,7 @@
"\n",
"This notebook demonstrates training a custom image classification model using Tensorflow and Vertex AI SDK by creating a custom training container. Additionally, the notebooks also deploys the trained model to Vertex AI and predictions are generated from it.\n",
"\n",
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training)."
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [Custom training](https://cloud.google.com/vertex-ai/docs/training/custom-training)."
]
},
{
@@ -64,7 +64,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK to create tabular binary classification models and do online prediction using a Google Cloud [AutoML](https://cloud.google.com/vertex-ai/docs/start/automl-users) model.\n",
"\n",
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [AutoML Tabular](https://cloud.google.com/vertex-ai/docs/start/automl-users#tables)."
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai). Learn more about [Classification for tabular data](https://cloud.google.com/vertex-ai/docs/tabular-data/classification-regression/overview)."
]
},
{
@@ -65,7 +65,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK for Python to train and deploy an AutoML object detection model.\n",
"\n",
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [AutoML Image](https://cloud.google.com/vertex-ai/docs/tutorials/image-recognition-automl/training)."
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [Object detection for image data](https://cloud.google.com/vertex-ai/docs/training-overview#object_detection_for_images)."
]
},
{
@@ -61,11 +61,11 @@
"source": [
"## Overview\n",
"\n",
"This notebook demonstrates how to create an AutoML Video Classification Model, with a Vertex AI video dataset, and how to serve the model for batch prediction. It requires you provide a bucket where the dataset will be stored.\n",
"This notebook demonstrates how to create an AutoML Text Classification Model, with a Vertex AI text dataset, and how to serve the model for batch prediction. It requires you provide a bucket where the dataset will be stored.\n",
"\n",
"Note: you may incur charges for training, prediction, storage or usage of other GCP products in connection with testing this SDK.\n",
"\n",
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [AutoML Text](https://cloud.google.com/vertex-ai/docs/text-data/classification/prepare-data)."
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [Classification for text data](https://cloud.google.com/vertex-ai/docs/training-overview#classification_for_text)."
]
},
{
@@ -76,7 +76,7 @@
"source": [
"### Objective\n",
"\n",
"The objective of this notebook is to build a AutoML Video Classification Model. The following steps have been followed:\n",
"The objective of this notebook is to build a AutoML Text Classification Model. The following steps have been followed:\n",
"This tutorial uses the following Google Cloud ML services :\n",
"\n",
"* Vertex AI Dataset resource\n",
@@ -87,11 +87,11 @@
"The steps performed include the following:\n",
"\n",
"* Set your task name, and GCS prefix\n",
"* Copy AutoML video demo train data for creating managed dataset\n",
"* Copy AutoML text demo train data for creating managed dataset\n",
"* Create a dataset on Vertex AI.\n",
"* Configure a training job\n",
"* Launch a training job and create a model on Vertex AI\n",
"* Copy AutoML Video Demo Prediction Data for creating batch prediction job\n",
"* Copy AutoML Text Demo Prediction Data for creating batch prediction job\n",
"* Perform batch prediction job on the model"
]
},
@@ -65,7 +65,7 @@
"\n",
"Note: you may incur charges for training, prediction, storage or usage of other Google Cloud products in connection with testing this SDK.\n",
"\n",
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [AutoML Text](https://cloud.google.com/vertex-ai/docs/text-data/entity-extraction/prepare-data)."
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [Entity extraction for text data](https://cloud.google.com/vertex-ai/docs/training-overview#entity_extraction_for_text).\n"
]
},
{
@@ -76,7 +76,7 @@
"source": [
"### Objective\n",
"\n",
"The objective of this notebook is to build a AutoML Text Entity Extraction Model. The following steps have been followed:\n",
"The objective of this notebook is to build a AutoML Text Entity Extraction model. The following steps have been followed:\n",
"This tutorial uses the following Google Cloud ML services :\n",
"\n",
"* Vertex AI Dataset resource\n",
@@ -66,7 +66,7 @@
"\n",
"Note: you may incur charges for training, prediction, storage or usage of other Google Cloud products in connection with testing this SDK.\n",
"\n",
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [AutoML Text](https://cloud.google.com/vertex-ai/docs/text-data/sentiment-analysis/prepare-data)."
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [Sentiment analysis for text data](https://cloud.google.com/vertex-ai/docs/training-overview#sentiment_analysis_for_text)."
]
},
{
@@ -64,7 +64,7 @@
"\n",
"This tutorial demonstrates how to use the Vertex AI SDK for Python to train and deploy a custom tabular classification XGBoost model for batch prediction.\n",
"\n",
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training)."
"Learn more about [Migrate to Vertex AI](https://cloud.google.com/vertex-ai/docs/start/migrating-to-vertex-ai) and [Custom training](https://cloud.google.com/vertex-ai/docs/training/custom-training)."
]
},
{
+22 -11
View File
@@ -1,3 +1,20 @@
[Track parameters and metrics for custom training jobs](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/ml_metadata/sdk-metric-parameter-tracking-for-custom-jobs.ipynb)
```
Learn how to use Vertex AI SDK for Python to:
The steps performed include:
- Track training parameters and prediction metrics for a custom training job.
- Extract and perform analysis for all parameters and metrics within an Experiment.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex ML Metadata](https://cloud.google.com/vertex-ai/docs/ml-metadata).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training).
[Track parameters and metrics for locally trained models](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/ml_metadata/sdk-metric-parameter-tracking-for-locally-trained-models.ipynb)
```
@@ -10,17 +27,7 @@ The steps performed include:
```
[Track parameters and metrics for custom training jobs](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/ml_metadata/sdk-metric-parameter-tracking-for-custom-jobs.ipynb)
```
Learn how to use Vertex AI SDK for Python to:
The steps performed include:
- Track training parameters and prediction metrics for a custom training job.
- Extract and perform analysis for all parameters and metrics within an Experiment.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex ML Metadata](https://cloud.google.com/vertex-ai/docs/ml-metadata).
[Track artifacts and metrics across Vertex AI Pipelines runs using Vertex ML Metadata](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/ml_metadata/vertex-pipelines-ml-metadata.ipynb)
@@ -39,3 +46,7 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex ML Metadata](https://cloud.google.com/vertex-ai/docs/ml-metadata).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Pipelines](https://cloud.google.com/vertex-ai/docs/pipelines/introduction).
@@ -63,7 +63,7 @@
"\n",
"This notebook demonstrates how to track metrics and parameters for Vertex AI custom training jobs, and how to perform detailed analysis using this data.\n",
"\n",
"Learn more about [Vertex ML Metadata](https://cloud.google.com/vertex-ai/docs/ml-metadata) and [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training)."
"Learn more about [Vertex ML Metadata](https://cloud.google.com/vertex-ai/docs/ml-metadata) and [Custom training](https://cloud.google.com/vertex-ai/docs/training/custom-training)."
]
},
{
+58 -33
View File
@@ -1,3 +1,4 @@
[Evaluating batch prediction results from an AutoML Tabular classification model](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_evaluation/automl_tabular_classification_model_evaluation.ipynb)
```
@@ -14,22 +15,9 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Model Evaluation](https://cloud.google.com/vertex-ai/docs/evaluation/introduction).
[Evaluating batch prediction results from AutoML Video classification model](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_evaluation/automl_video_classification_model_evaluation.ipynb)
```
Learn how to train a Vertex AI AutoML Video classification model and learn how to evaluate it through a Vertex AI pipeline job using `google_cloud_pipeline_components`:
The steps performed include:
- Create a `Vertex AI Dataset`.
- Train a Automl Video Classification model on the `Vertex AI Dataset` resource.
- Import the trained `AutoML Vertex AI Model resource` into the pipeline.
- Run a batch prediction job inside the pipeline.
- Evaulate the AutoML model using the classification evaluation component.
- Import the classification metrics to the AutoML Vertex AI Model resource.
```
&nbsp;&nbsp;&nbsp;Learn more about [AutoML Tabular](https://cloud.google.com/vertex-ai/docs/start/automl-users#tables).
[Evaluating batch prediction results from AutoML Tabular regression model](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_evaluation/automl_tabular_regression_model_evaluation.ipynb)
@@ -49,25 +37,9 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Model Evaluation](https://cloud.google.com/vertex-ai/docs/evaluation/introduction).
[Evaluating batch prediction results from custom tabular regression model](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_evaluation/custom_tabular_regression_model_evaluation.ipynb)
```
Learn how to evaluate a Vertex AI model resource through a Vertex AI pipeline job using `google_cloud_pipeline_components`:
The steps performed include:
- Create a Vertex AI `CustomTrainingJob` for training a model.
- Run the `CustomTrainingJob`
- Retrieve and load the model artifacts.
- View the model evaluation.
- Upload the model as a Vertex AI Model resource.
- Import a pre-trained `Vertex AI model resource` into the pipeline.
- Run a `batch prediction` job in the pipeline.
- Evaulate the model using the `regression evaluation component`.
- Import the Regression Metrics to the Vertex AI model resource.
```
&nbsp;&nbsp;&nbsp;Learn more about [AutoML Tabular](https://cloud.google.com/vertex-ai/docs/start/automl-users#tables).
[AutoML text classification pipelines using google-cloud-pipeline-components](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_evaluation/automl_text_classification_model_evaluation.ipynb)
@@ -86,6 +58,31 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Model Evaluation](https://cloud.google.com/vertex-ai/docs/evaluation/introduction).
&nbsp;&nbsp;&nbsp;Learn more about [AutoML Text](https://cloud.google.com/vertex-ai/docs/text-data/classification/prepare-data).
[Evaluating batch prediction results from AutoML Video classification model](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_evaluation/automl_video_classification_model_evaluation.ipynb)
```
Learn how to train a Vertex AI AutoML Video classification model and learn how to evaluate it through a Vertex AI pipeline job using `google_cloud_pipeline_components`:
The steps performed include:
- Create a `Vertex AI Dataset`.
- Train a Automl Video Classification model on the `Vertex AI Dataset` resource.
- Import the trained `AutoML Vertex AI Model resource` into the pipeline.
- Run a batch prediction job inside the pipeline.
- Evaulate the AutoML model using the classification evaluation component.
- Import the classification metrics to the AutoML Vertex AI Model resource.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Model Evaluation](https://cloud.google.com/vertex-ai/docs/evaluation/introduction).
&nbsp;&nbsp;&nbsp;Learn more about [AutoML Video](https://cloud.google.com/vertex-ai/docs/video-data/classification/prepare-data).
[Evaluating BatchPrediction results from a Custom Tabular classification model](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_evaluation/custom_tabular_classification_model_evaluation.ipynb)
@@ -109,3 +106,31 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Model Evaluation](https://cloud.google.com/vertex-ai/docs/evaluation/introduction).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training).
[Evaluating batch prediction results from custom tabular regression model](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_evaluation/custom_tabular_regression_model_evaluation.ipynb)
```
Learn how to evaluate a Vertex AI model resource through a Vertex AI pipeline job using `google_cloud_pipeline_components`:
The steps performed include:
- Create a Vertex AI `CustomTrainingJob` for training a model.
- Run the `CustomTrainingJob`
- Retrieve and load the model artifacts.
- View the model evaluation.
- Upload the model as a Vertex AI Model resource.
- Import a pre-trained `Vertex AI model resource` into the pipeline.
- Run a `batch prediction` job in the pipeline.
- Evaulate the model using the `regression evaluation component`.
- Import the Regression Metrics to the Vertex AI model resource.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Model Evaluation](https://cloud.google.com/vertex-ai/docs/evaluation/introduction).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training).
@@ -63,7 +63,7 @@
"\n",
"This notebook demonstrates how to use the Vertex AI classification model evaluation component to evaluate an AutoML Tabular classification model. Model evaluation helps determine your model's performance based on the evaluation metrics and improve the model whenever necessary. \n",
"\n",
"Learn more about [Vertex AI Model Evaluation](https://cloud.google.com/vertex-ai/docs/evaluation/introduction) and [AutoML Tabular](https://cloud.google.com/vertex-ai/docs/start/automl-users#tables)."
"Learn more about [Vertex AI Model Evaluation](https://cloud.google.com/vertex-ai/docs/evaluation/introduction). Learn more about [Classification for tabular data](https://cloud.google.com/vertex-ai/docs/tabular-data/classification-regression/overview)."
]
},
{
@@ -63,7 +63,7 @@
"\n",
"This notebook demonstrates how to use Vertex AI regression model evaluation component to evaluate an AutoML Tabular regression model. Model evaluation helps you determine your model performance based on the evaluation metrics and improve the model if necessary. \n",
"\n",
"Learn more about [Vertex AI Model Evaluation](https://cloud.google.com/vertex-ai/docs/evaluation/introduction) and [AutoML Tabular](https://cloud.google.com/vertex-ai/docs/start/automl-users#tables)."
"Learn more about [Vertex AI Model Evaluation](https://cloud.google.com/vertex-ai/docs/evaluation/introduction). Learn more about [Regression for tabular data](https://cloud.google.com/vertex-ai/docs/tabular-data/classification-regression/overview)."
]
},
{
@@ -92,7 +92,7 @@
"- Run the `AutoMLTabularTrainingJob` which returns a model\n",
"- Import a pre-trained `AutoML model resource` into the pipeline\n",
"- Run a `batch prediction` job in the pipeline\n",
"- Evaulate the AutoML model using the `regression evaluation component`\n",
"- Evaluate the AutoML model using the `regression evaluation component`\n",
"- Import the Regression Metrics to the AutoML model resource"
]
},
@@ -63,7 +63,7 @@
"\n",
"This notebook demonstrates how to use the Vertex AI classification model evaluation component to evaluate an AutoML video classification model. Model evaluation helps you determine your model performance based on the evaluation metrics and improve the model if necessary. \n",
"\n",
"Learn more about [Vertex AI Model Evaluation](https://cloud.google.com/vertex-ai/docs/evaluation/introduction) and [AutoML Video](https://cloud.google.com/vertex-ai/docs/video-data/classification/prepare-data)."
"Learn more about [Vertex AI Model Evaluation](https://cloud.google.com/vertex-ai/docs/evaluation/introduction) and [Classification for video data](https://cloud.google.com/vertex-ai/docs/training-overview#classification_for_videos)."
]
},
{
@@ -792,7 +792,19 @@
"\n",
"The `run` method when completed returns the `Model` resource.\n",
"\n",
"The execution of the training pipeline can take over 3 hours to complete."
"The execution of the training pipeline can take over 24 hours to complete."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "3eaba926cdfa"
},
"outputs": [],
"source": [
"if os.getenv(\"IS_TESTING\"):\n",
" sys.exit(0)"
]
},
{
@@ -63,7 +63,7 @@
"\n",
"This notebook demonstrates how to use the Vertex AI regression model evaluation component to evaluate a custom regression model. Model evaluation helps you determine your model performance based on the evaluation metrics and improve the model if necessary. \n",
"\n",
"Learn more about [Vertex AI Model Evaluation](https://cloud.google.com/vertex-ai/docs/evaluation/introduction) and [Vertex AI Training](https://cloud.google.com/vertex-ai/docs/training/custom-training)."
"Learn more about [Vertex AI Model Evaluation](https://cloud.google.com/vertex-ai/docs/evaluation/introduction) and [Custom training](https://cloud.google.com/vertex-ai/docs/training/custom-training)."
]
},
{
@@ -0,0 +1,865 @@
{
"cells": [
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "ur8xi4C7S06n"
},
"outputs": [],
"source": [
"# Copyright 2023 Google LLC\n",
"#\n",
"# Licensed under the Apache License, Version 2.0 (the \"License\");\n",
"# you may not use this file except in compliance with the License.\n",
"# You may obtain a copy of the License at\n",
"#\n",
"# https://www.apache.org/licenses/LICENSE-2.0\n",
"#\n",
"# Unless required by applicable law or agreed to in writing, software\n",
"# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
"# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
"# See the License for the specific language governing permissions and\n",
"# limitations under the License."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "JAPoU8Sm5E6e"
},
"source": [
"# Get started with importing a custom model evaluation to the Vertex AI Model Registry\n",
"\n",
"<table align=\"left\">\n",
"\n",
" <td>\n",
" <a href=\"https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_evaluation/get_started_with_custom_model_evaluation_import.ipynb\">\n",
" <img src=\"https://cloud.google.com/ml-engine/images/colab-logo-32px.png\" alt=\"Colab logo\"> Run in Colab\n",
" </a>\n",
" </td>\n",
" <td>\n",
" <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_evaluation/get_started_with_custom_model_evaluation_import.ipynb\">\n",
" <img src=\"https://cloud.google.com/ml-engine/images/github-logo-32px.png\" alt=\"GitHub logo\">\n",
" View on GitHub\n",
" </a>\n",
" </td>\n",
" <td>\n",
" <a href=\"https://console.cloud.google.com/vertex-ai/workbench/deploy-notebook?download_url=https://raw.githubusercontent.com/GoogleCloudPlatform/vertex-ai-samples/main/notebooks/official/model_evaluation/get_started_with_custom_model_evaluation_import.ipynb\">\n",
" <img src=\"https://lh3.googleusercontent.com/UiNooY4LUgW_oTvpsNhPpQzsstV5W8F7rYgxgGBD85cWJoLmrOzhVs_ksK_vgx40SHs7jCqkTkCk=e14-rj-sc0xffffff-h130-w32\" alt=\"Vertex AI logo\">\n",
" Open in Vertex AI Workbench\n",
" </a>\n",
" </td> \n",
"</table>"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "24743cf4a1e1"
},
"source": [
"**_NOTE_**: This notebook has been tested in the following environment:\n",
"\n",
"* Python version = 3.9"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "tvgnzT1CKxrO"
},
"source": [
"## Overview\n",
"\n",
"This tutorial shows how to use Vertex AI Model Evaluation to import a custom model evaluation to an existing Vertex AI Model Registry entry."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "d975e698c9a4"
},
"source": [
"### Objective\n",
"\n",
"In this tutorial, you learn how to construct and upload a custom model evaluation, and upload the custom model evaluation to a Model resource entry in Vertex AI Model Registry.\n",
"\n",
"This tutorial uses the following Google Cloud ML services and resources:\n",
"\n",
"- Vertex AI Model Evaluation\n",
"- Vertex AI Model Registry\n",
"\n",
"The steps performed include:\n",
"\n",
"- Import a pretrained (blessed) model to the Vertex AI Model Registry.\n",
"- Construct a custom model evaluation.\n",
"- Import the model evaluation metrics to the corresponding model in the Vertex AI Model Registry.\n",
"- List the model evaluation for the corresponding model in the Vertex AI Model Registry.\n",
"- Construct a second custom model evaluation.\n",
"- Import the second model evaluation metrics to the corresponding model in the Vertex AI Model Registry.\n",
"- List the second model evaluation for the corresponding model in the Vertex AI Model Registry.\n",
"\n",
"Learn more about [Model Evaluation in Vertex AI](https://cloud.google.com/vertex-ai/docs/evaluation/introduction)."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "08d289fa873f"
},
"source": [
"### Model\n",
"\n",
"This tutorial uses a pre-trained image classification model from TensorFlow Hub, which is trained on ImageNet dataset.\n",
"\n",
"Learn more about [ResNet V2 pretained model](https://tfhub.dev/google/imagenet/resnet_v2_101/classification/5). "
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "aed92deeb4a0"
},
"source": [
"### Costs \n",
"\n",
"This tutorial uses billable components of Google Cloud:\n",
"\n",
"* Vertex AI\n",
"* Cloud Storage\n",
"\n",
"Learn about [Vertex AI pricing](https://cloud.google.com/vertex-ai/pricing),\n",
"and [Cloud Storage pricing](https://cloud.google.com/storage/pricing), \n",
"and use the [Pricing Calculator](https://cloud.google.com/products/calculator/)\n",
"to generate a cost estimate based on your projected usage."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "i7EUnXsZhAGF"
},
"source": [
"## Installation\n",
"\n",
"Install the following packages required to execute this notebook. "
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "2b4ef9b72d43"
},
"outputs": [],
"source": [
"# Install the packages\n",
"USER=''\n",
"! pip3 install {USER} --upgrade google-cloud-aiplatform \\\n",
" tensorflow==2.5 \\\n",
" tensorflow-hub"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "58707a750154"
},
"source": [
"### Colab only: Uncomment the following cell to restart the kernel."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "f200f10a1da3"
},
"outputs": [],
"source": [
"# Automatically restart kernel after installs so that your environment can access the new packages\n",
"# import IPython\n",
"\n",
"# app = IPython.Application.instance()\n",
"# app.kernel.do_shutdown(True)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "BF1j6f9HApxa"
},
"source": [
"## Before you begin\n",
"\n",
"### Set up your Google Cloud project\n",
"\n",
"**The following steps are required, regardless of your notebook environment.**\n",
"\n",
"1. [Select or create a Google Cloud project](https://console.cloud.google.com/cloud-resource-manager). When you first create an account, you get a $300 free credit towards your compute/storage costs.\n",
"\n",
"2. [Make sure that billing is enabled for your project](https://cloud.google.com/billing/docs/how-to/modify-project).\n",
"\n",
"3. [Enable the Vertex AI API]\n",
"\n",
"4. If you are running this notebook locally, you need to install the [Cloud SDK](https://cloud.google.com/sdk)."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "WReHDGG5g0XY"
},
"source": [
"#### Set your project ID\n",
"\n",
"**If you don't know your project ID**, try the following:\n",
"* Run `gcloud config list`.\n",
"* Run `gcloud projects list`.\n",
"* See the support page: [Locate the project ID](https://support.google.com/googleapi/answer/7014113)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "oM1iC_MfAts1"
},
"outputs": [],
"source": [
"PROJECT_ID = \"[your-project-id]\" # @param {type:\"string\"}\n",
"\n",
"# Set the project id\n",
"! gcloud config set project {PROJECT_ID}"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "region"
},
"source": [
"#### Region\n",
"\n",
"You can also change the `REGION` variable used by Vertex AI. Learn more about [Vertex AI regions](https://cloud.google.com/vertex-ai/docs/general/locations)."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "region"
},
"outputs": [],
"source": [
"REGION = \"us-central1\" # @param {type: \"string\"}"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "sBCra4QMA2wR"
},
"source": [
"### Authenticate your Google Cloud account\n",
"\n",
"Depending on your Jupyter environment, you may have to manually authenticate. Follow the relevant instructions below."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "74ccc9e52986"
},
"source": [
"**1. Vertex AI Workbench**\n",
"* Do nothing as you are already authenticated."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "de775a3773ba"
},
"source": [
"**2. Local JupyterLab instance, uncomment and run:**"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "254614fa0c46"
},
"outputs": [],
"source": [
"# ! gcloud auth login"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "ef21552ccea8"
},
"source": [
"**3. Colab, uncomment and run:**"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "603adbbf0532"
},
"outputs": [],
"source": [
"# from google.colab import auth\n",
"# auth.authenticate_user()"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "f6b2ccc891ed"
},
"source": [
"**4. Service account or other**\n",
"* See how to grant Cloud Storage permissions to your service account at https://cloud.google.com/storage/docs/gsutil/commands/iam#ch-examples."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "zgPO1eR3CYjk"
},
"source": [
"### Create a Cloud Storage bucket\n",
"\n",
"Create a storage bucket to store intermediate artifacts such as datasets.\n",
"\n",
"- *{Note to notebook author: For any user-provided strings that need to be unique (like bucket names or model ID's), append \"-unique\" to the end so proper testing can occur}*"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "MzGDU7TWdts_"
},
"outputs": [],
"source": [
"BUCKET_URI = f\"gs://your-bucket-name-unique-{PROJECT_ID}\" # @param {type:\"string\"}"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "-EcIXiGsCePi"
},
"source": [
"**Only if your bucket doesn't already exist**: Run the following cell to create your Cloud Storage bucket."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "NIq7R4HZCfIc"
},
"outputs": [],
"source": [
"! gsutil mb -l $REGION -p $PROJECT_ID $BUCKET_URI"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "960505627ddf"
},
"source": [
"### Import libraries"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "PyQmSRbKA8r-"
},
"outputs": [],
"source": [
"import tensorflow as tf\n",
"import tensorflow_hub as hub\n",
"from google.cloud import aiplatform\n",
"from google.cloud.aiplatform import gapic"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "init_aip:mbsdk,all"
},
"source": [
"### Initialize Vertex AI SDK for Python\n",
"\n",
"Initialize the Vertex AI SDK for Python for your project."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "init_aip:mbsdk,all"
},
"outputs": [],
"source": [
"aiplatform.init(project=PROJECT_ID, location=REGION, staging_bucket=BUCKET_URI)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "accelerators:training,cpu,prediction,cpu,mbsdk"
},
"source": [
"#### Set hardware accelerators\n",
"\n",
"You can set hardware accelerators for prediction.\n",
"\n",
"Set the variables `DEPLOY_GPU/DEPLOY_NGPU` to use a container image supporting a GPU and the number of GPUs allocated to the virtual machine (VM) instance. For example, to use a GPU container image with 4 Nvidia Telsa K80 GPUs allocated to each VM, you would specify:\n",
"\n",
" (aip.gapic.AcceleratorType.NVIDIA_TESLA_K80, 4)\n",
"\n",
"\n",
"Otherwise specify `(None, None)` to use a container image to run on a CPU.\n",
"\n",
"Learn more about [hardware accelerator support for your region](https://cloud.google.com/vertex-ai/docs/general/locations#accelerators), and [GPU pricing](https://cloud.google.com/compute/gpus-pricing)."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "accelerators:training,cpu,prediction,cpu,mbsdk"
},
"outputs": [],
"source": [
"DEPLOY_GPU, DEPLOY_NGPU = (None, None)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "container:training,prediction"
},
"source": [
"#### Set pre-built containers\n",
"\n",
"Set the pre-built Docker container image for prediction.\n",
"\n",
"For the latest list, see [Pre-built containers for prediction](https://cloud.google.com/ai-platform-unified/docs/predictions/pre-built-containers)."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "container:training,prediction"
},
"outputs": [],
"source": [
"TF = \"2.5\".replace(\".\", \"-\")\n",
"\n",
"if DEPLOY_GPU:\n",
" DEPLOY_VERSION = \"tf2-gpu.{}\".format(TF)\n",
"else:\n",
" DEPLOY_VERSION = \"tf2-cpu.{}\".format(TF)\n",
"\n",
"DEPLOY_IMAGE = \"{}-docker.pkg.dev/vertex-ai/prediction/{}:latest\".format(\n",
" REGION.split(\"-\")[0], DEPLOY_VERSION\n",
")\n",
"\n",
"print(\"Deployment:\", DEPLOY_IMAGE, DEPLOY_GPU, DEPLOY_NGPU)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "d8128b8ff025"
},
"source": [
"## Get pretrained model from TensorFlow Hub\n",
"\n",
"For demonstration purposes, this tutorial uses a pretrained model from TensorFlow Hub (TFHub), which is then uploaded to a `Vertex AI Model` resource. Once you have a `Vertex AI Model` resource, the model can be deployed to a `Vertex AI Endpoint` resource.\n",
"\n",
"### Download the pretrained model\n",
"\n",
"First, you download the pretrained model from TensorFlow Hub. The model gets downloaded as a TF.Keras layer. To finalize the model, in this example, you create a `Sequential()` model with the downloaded TFHub model as a layer, and specify the input shape to the model."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "c55fa4c826f7"
},
"outputs": [],
"source": [
"tfhub_model = tf.keras.Sequential(\n",
" [hub.KerasLayer(\"https://tfhub.dev/google/imagenet/resnet_v2_101/classification/5\")]\n",
")\n",
"\n",
"tfhub_model.build([None, 32, 32, 3])\n",
"\n",
"tfhub_model.summary()"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "64618c713db9"
},
"outputs": [],
"source": [
"MODEL_DIR = BUCKET_URI + \"/model\"\n",
"tfhub_model.save(MODEL_DIR)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "e8ce91147c93"
},
"source": [
"### Upload the TensorFlow Hub model to a `Vertex AI Model` resource\n",
"\n",
"Finally, you upload the model artifacts from the TFHub model into a `Vertex AI Model` resource using the method `upload()`, with the following parameters:\n",
"\n",
"- `display_name`: A human readable name for the `Model` resource.\n",
"- `artifact_uri`: The Cloud Storage location of the model package.\n",
"- `serving_container_image_uri`: The serving container image.\n",
"\n",
"Uploading a model into a Vertex AI Model resource returns a long running operation, since it may take a few moments. \n",
"\n",
"*Note:* When you upload the model artifacts to a `Vertex AI Model` resource, you specify the corresponding deployment container image."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "ad61e1429512"
},
"outputs": [],
"source": [
"model = aiplatform.Model.upload(\n",
" display_name=\"resnet\",\n",
" artifact_uri=MODEL_DIR,\n",
" serving_container_image_uri=DEPLOY_IMAGE,\n",
" is_default_version=True,\n",
" version_aliases=[\"v1\"],\n",
")\n",
"\n",
"print(model)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "c11e98ef5391"
},
"source": [
"## Introduction to custom model evaluations\n",
"\n",
"When training a custom model, one generally performs some form of an evaluation of the trained model. Your custom model evaluation can then be imported to the corresponding model in the Vertex AI Model Registry using the `import_model_evaluation()` method. Once imported, the custom model evaluation can be subsequently retreived with the `list_model_evaluations()` method. \n",
"\n",
"The Vertex AI Model Registry supports importing multiple model evaluations for a model where each evaluation is distinquished by a unique `display_name`.\n",
"\n",
"\n",
"### Create a model evaluation\n",
"\n",
"First, you create a model evaluation in a format that corresponds to one of the predefined schemas for model evaluations. In this example, you use the schema for a classification metric, and specify the following subset of evaluation metrics as a dictionary:\n",
"\n",
"- `logLoss`: The log loss.\n",
"- `auPrc`: The accuracy.\n",
"\n",
"You then construct the `ModelEvaluation` object with the following parameters:\n",
"\n",
"- `display_name`: The human readable name for the evaluation metric.\n",
"- `metrics_schema_uri`: The schema for the specific type of evaluation metrics.\n",
"- `metrics`: The dictionary with the evaluation metrics.\n",
"\n",
"Learn more about [Schemas for evaluation metrics](https://cloud.google.com/vertex-ai/docs/evaluation/introduction#features)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "e9af222db292"
},
"outputs": [],
"source": [
"metrics = {\"logLoss\": 1.4, \"auPrc\": 0.85}\n",
"print(metrics)\n",
"\n",
"model_eval = gapic.ModelEvaluation(\n",
" display_name=\"eval\",\n",
" metrics_schema_uri=\"gs://google-cloud-aiplatform/schema/modelevaluation/classification_metrics_1.0.0.yaml\",\n",
" metrics=metrics,\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "68870bd8194d"
},
"source": [
"### Upload the evaluation metrics to the Model Registry\n",
"\n",
"Next, upload the model's evaluation from the custom training job to the corresponding entry in the Vertex AI Model Registry.\n",
"\n",
"Currently, there is not yet support for this method in the SDK. Instead, you use the lower level GAPIC API interface."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "3044788848af"
},
"outputs": [],
"source": [
"API_ENDPOINT = f\"{REGION}-aiplatform.googleapis.com\"\n",
"client = gapic.ModelServiceClient(client_options={\"api_endpoint\": API_ENDPOINT})\n",
"\n",
"client.import_model_evaluation(parent=model.resource_name, model_evaluation=model_eval)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "337319a19efe"
},
"source": [
"### List the custom model evaluation\n",
"\n",
"Now that your custom metric has been uploaded to the corresponding model in the Vertex AI Model Registry, you can retriv"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "1b3374e988cf"
},
"outputs": [],
"source": [
"evaluation = model.list_model_evaluations()[0]\n",
"print(evaluation.gca_resource)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "68870bd8194d"
},
"source": [
"### Upload a second evaluation metrics to the Model Registry\n",
"\n",
"Next, upload a second model evaluation to the corresponding entry in the Vertex AI Model Registry. In this example, we refer to first evaluation metric as `eval` (from training) and the second as `prod` (from production data)."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "9b864c0ba07f"
},
"outputs": [],
"source": [
"metrics = {\"logLoss\": 1.2, \"auPrc\": 0.87}\n",
"print(metrics)\n",
"\n",
"model_prod = gapic.ModelEvaluation(\n",
" display_name=\"prod\",\n",
" metrics_schema_uri=\"gs://google-cloud-aiplatform/schema/modelevaluation/classification_metrics_1.0.0.yaml\",\n",
" metrics=metrics,\n",
")\n",
"\n",
"client.import_model_evaluation(parent=model.resource_name, model_evaluation=model_prod)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "337319a19efe"
},
"source": [
"### List a specific custom model evaluation\n",
"\n",
"Now that your custom second metric has been uploaded to the corresponding model in the Vertex AI Model Registry, you can retrieve this specific evaluation by filtering using the `display_name`:"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "1b3374e988cf"
},
"outputs": [],
"source": [
"evaluations = model.list_model_evaluations()\n",
"for evaluation in evaluations:\n",
" if evaluation.display_name == \"prod\":\n",
" print(evaluation.gca_resource)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "b3a3993dc813"
},
"source": [
"### Upload version 2 of the TFHub model to the `Vertex AI Model Registry`\n",
"\n",
"Next, you upload the second version of the TFHub model as a `Model` resource in the `Vertex AI Model Registry`, with the additional following parameters:\n",
"\n",
"- `parent_model`: The existing `Model` resource for which to add this model as the next model version.\n",
"- `is_default_version`: Whether this will be the default version for the `Model` resource. In this example, you change from the default from the first version to the second version of the model.\n",
"- `version_ailiases`: User defined list of alternative alias names for the model version, such as `production`.\n",
"- `version_description`: User description of the model version.\n",
"\n",
"When a subsequent model version is created in the `Vertex AI Model Registry`, the property `version_id` will automatically be incremented. In this example, it will be set to 2 (2nd version)."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "852be5e5f964"
},
"outputs": [],
"source": [
"model_v2 = aiplatform.Model.upload(\n",
" display_name=\"resnet\",\n",
" artifact_uri=MODEL_DIR,\n",
" serving_container_image_uri=DEPLOY_IMAGE,\n",
" parent_model=model.resource_name,\n",
" is_default_version=True,\n",
" version_aliases=[\"v2\"],\n",
" version_description=\"This is the second version of the model\",\n",
")\n",
"\n",
"print(model_v2)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "68870bd8194d"
},
"source": [
"### Upload an evaluation metrics for version 2 of the model to the Model Registry\n",
"\n",
"Next, upload a model evaluation to the corresponding model version in the Vertex AI Model Registry. Note, you referenced `model_v2.resource_name` to refer to version 2 of this model."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "9b864c0ba07f"
},
"outputs": [],
"source": [
"metrics = {\"logLoss\": 1.0, \"auPrc\": 0.91}\n",
"print(metrics)\n",
"\n",
"model_eval = gapic.ModelEvaluation(\n",
" display_name=\"eval\",\n",
" metrics_schema_uri=\"gs://google-cloud-aiplatform/schema/modelevaluation/classification_metrics_1.0.0.yaml\",\n",
" metrics=metrics,\n",
")\n",
"\n",
"client.import_model_evaluation(\n",
" parent=model_v2.resource_name, model_evaluation=model_eval\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "d4e72b21b06b"
},
"source": [
"### List the evaluations for both versions of the model\n",
"\n",
"Finally, list the number of evaluations for both versions of the model."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "369a23d392a4"
},
"outputs": [],
"source": [
"evaluations = model.list_model_evaluations()\n",
"print(\"Model v1 no. of evaluations\", len(evaluations))\n",
"evaluations = model_v2.list_model_evaluations()\n",
"print(\"Model v2 no. of evaluations\", len(evaluations))"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "TpV-iwP9qw9c"
},
"source": [
"## Cleaning up\n",
"\n",
"To clean up all Google Cloud resources used in this project, you can [delete the Google Cloud\n",
"project](https://cloud.google.com/resource-manager/docs/creating-managing-projects#shutting_down_projects) you used for the tutorial.\n",
"\n",
"Otherwise, you can delete the individual resources you created in this tutorial:\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "sx_vKniMq9ZX"
},
"outputs": [],
"source": [
"import os\n",
"\n",
"# Delete model resource\n",
"model.delete()\n",
"\n",
"# Delete Cloud Storage objects that were created\n",
"delete_bucket = False\n",
"if delete_bucket or os.getenv(\"IS_TESTING\"):\n",
" ! gsutil -m rm -r $BUCKET_URI"
]
}
],
"metadata": {
"colab": {
"collapsed_sections": [],
"name": "get_started_with_custom_model_evaluation_import.ipynb",
"toc_visible": true
},
"kernelspec": {
"display_name": "Python 3",
"name": "python3"
}
},
"nbformat": 4,
"nbformat_minor": 0
}
+126 -3
View File
@@ -1,7 +1,4 @@
### model_monitoring
[Vertex AI Batch Prediction with Model Monitoring](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_monitoring/batch_prediction_model_monitoring.ipynb)
```
@@ -15,6 +12,130 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Model Monitoring for batch predictions](https://cloud.google.com/vertex-ai/docs/model-monitoring/model-monitoring-batch-predictions).
[Vertex AI Model Monitoring for AutoML tabular models](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_monitoring/get_started_with_model_monitoring_automl.ipynb)
```
Learn to use the `Vertex AI Model Monitoring` service to detect feature skew and drift in the input predict requests, for AutoML tabular models.
The steps performed include:
- Train an `AutoML` model.
- Deploy the `Model` resource to the `Endpoint` resource.
- Configure the `Endpoint` resource for model monitoring.
- Generate synthetic prediction requests for skew.
- Wait for email alert notification.
- Generate synthetic prediction requests for drift.
- Wait for email alert notification.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Model Monitoring](https://cloud.google.com/vertex-ai/docs/model-monitoring).
[Vertex AI Model Monitoring for batch prediction in AutoML image models](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_monitoring/get_started_with_model_monitoring_automl_image_batch.ipynb)
```
Learn how to use `Vertex AI Model Monitoring` with `Vertex AI Batch Prediction` with an AutoML image classification model to detect an out of distribution image.
The steps performed include:
1. Train an AutoML image classification model.
2. Submit a batch prediction containing both in and out of distribution images.
3. Use Model Monitoring to calculate anomaly score on each image.
4. Identify the images in the batch prediction request that are out of distribution.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Model Monitoring](https://cloud.google.com/vertex-ai/docs/model-monitoring).
[Vertex AI Model Monitoring for custom tabular models](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_monitoring/get_started_with_model_monitoring_custom.ipynb)
```
Learn to use the `Vertex AI Model Monitoring` service to detect feature skew and drift in the input predict requests, for custom tabular models.
The steps performed include:
- Download a pre-trained custom tabular model.
- Upload the pre-trained model as a `Model` resource.
- Deploy the `Model` resource to the `Endpoint` resource.
- Configure the `Endpoint` resource for model monitoring.
- Generate synthetic prediction requests for skew.
- Wait for email alert notification.
- Generate synthetic prediction requests for drift.
- Wait for email alert notification.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Model Monitoring](https://cloud.google.com/vertex-ai/docs/model-monitoring).
[Vertex AI Model Monitoring for custom tabular models with TensorFlow Serving container](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_monitoring/get_started_with_model_monitoring_custom_tf_serving.ipynb)
```
Learn to use the `Vertex AI Model Monitoring` service to detect feature skew and drift in the input predict requests, for custom tabular models, using a custom deployment container.
The steps performed include:
- Download a pre-trained custom tabular model.
- Upload the pre-trained model as a `Model` resource.
- Deploying the `Model` resource to an `Endpoint` resource with `TensorFlow Serving` serving binary.
- Configure the `Endpoint` resource for model monitoring.
- Generate synthetic prediction requests for skew.
- Wait for email alert notification.
- Generate synthetic prediction requests for drift.
- Wait for email alert notification.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Model Monitoring](https://cloud.google.com/vertex-ai/docs/model-monitoring).
[Vertex AI Model Monitoring for setup for tabular models](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_monitoring/get_started_with_model_monitoring_setup.ipynb)
```
Learn to setup the `Vertex AI Model Monitoring` service to detect feature skew and drift in the input predict requests.
The steps performed include:
- Download a pre-trained custom tabular model.
- Upload the pre-trained model as a `Model` resource.
- Deploy the `Model` resource to the `Endpoint` resource.
- Configure the `Endpoint` resource for model monitoring.
- Skew and drift detection for feature inputs.
- Skew and drift detection for feature attributions.
- Automatic generation of the `input schema` by sending 1000 prediction request.
- List, pause, resume and delete monitoring jobs.
- Restart monitoring job with predefined `input schema`.
- View logged monitored data.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Model Monitoring](https://cloud.google.com/vertex-ai/docs/model-monitoring).
[Vertex AI Model Monitoring for XGBoost models](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_monitoring/get_started_with_model_monitoring_xgboost.ipynb)
```
Learn to use the `Vertex AI Model Monitoring` service to detect feature skew and drift in the input predict requests for XGBoost models.
The steps performed include:
- Download a pre-trained XGBoost model.
- Upload the pre-trained model as a `Model` resource.
- Deploy the `Model` resource to the `Endpoint` resource.
- Configure the `Endpoint` resource for model monitoring:
- drift detection only -- no access to training data.
- predefine the input schema to map feature alias names to the unnamed array input to the model.
- Generate synthetic prediction requests for drift.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Model Monitoring](https://cloud.google.com/vertex-ai/docs/model-monitoring).
[Vertex AI Model Monitoring with Explainable AI Feature Attributions](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_monitoring/model_monitoring.ipynb)
@@ -33,3 +154,5 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Model Monitoring](https://cloud.google.com/vertex-ai/docs/model-monitoring).
@@ -61,7 +61,9 @@
"source": [
"## Overview\n",
"\n",
"In this notebook, you will learn how to use Model Monitoring with batch prediction requests on a deployed Vertex AI Model resource. In a companion notebook, <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_monitoring/model_monitoring.ipynb\" target=\"_blank\">Vertex AI Model Monitoring with Explainable AI Feature Attributions</a>, you can learn about how to apply model monitoring to streaming, real-time predictions."
"In this notebook, you will learn how to use Model Monitoring with batch prediction requests on a deployed Vertex AI Model resource. In a companion notebook, <a href=\"https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_monitoring/model_monitoring.ipynb\" target=\"_blank\">Vertex AI Model Monitoring with Explainable AI Feature Attributions</a>, you can learn about how to apply model monitoring to streaming, real-time predictions.\n",
"\n",
"Learn more about [Vertex AI Model Monitoring for batch predictions](https://cloud.google.com/vertex-ai/docs/model-monitoring/model-monitoring-batch-predictions)."
]
},
{
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
@@ -80,7 +80,9 @@
"\n",
"[Vertex Explainable AI](https://cloud.google.com/vertex-ai/docs/explainable-ai/overview) adds another facet to model monitoring, which we call feature attribution monitoring. Explainable AI enables you to understand the relative contribution of each feature to a resulting prediction. In essence, it assesses the magnitude of each feature's influence.\n",
"\n",
"If production traffic differs from training data, or varies substantially over time, **either in terms of model predictions or feature attributions**, that's likely to impact the quality of the answers your model produces. When that happens, you'd like to be alerted automatically and responsively, so that **you can anticipate problems before they affect your customer experiences or your revenue streams**."
"If production traffic differs from training data, or varies substantially over time, **either in terms of model predictions or feature attributions**, that's likely to impact the quality of the answers your model produces. When that happens, you'd like to be alerted automatically and responsively, so that **you can anticipate problems before they affect your customer experiences or your revenue streams**.\n",
"\n",
"Learn more about [Vertex AI Model Monitoring](https://cloud.google.com/vertex-ai/docs/model-monitoring)."
]
},
{
+4 -3
View File
@@ -1,7 +1,4 @@
### model_registry
[Deploy BiqQuery ML Model on Vertex AI Model Registry and make predictions](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_registry/bqml_vertexai_model_registry.ipynb)
```
@@ -18,3 +15,7 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Model Registry](https://cloud.google.com/vertex-ai/docs/model-registry/introduction).
&nbsp;&nbsp;&nbsp;Learn more about [BigQuery ML](https://cloud.google.com/vertex-ai/docs/beginner/bqml).
File diff suppressed because it is too large Load Diff
+263 -173
View File
@@ -1,76 +1,4 @@
### pipelines
[AutoML image classification pipelines using google-cloud-pipeline-components](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/google_cloud_pipeline_components_automl_images.ipynb)
```
Learn how to use `Vertex AI Pipelines` and `Google Cloud Pipeline Components` to build an `AutoML` image classification model.
The steps performed include:
- Create a KFP pipeline:
- Create a `Dataset` resource.
- Train an AutoML image classification `Model` resource.
- Create an `Endpoint` resource.
- Deploys the `Model` resource to the `Endpoint` resource.
- Compile the KFP pipeline.
- Execute the KFP pipeline using `Vertex AI Pipelines`
```
[Metrics visualization and run comparison using the KFP SDK](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/metrics_viz_run_compare_kfp.ipynb)
```
Learn how to use the KFP SDK to build pipelines that generate evaluation metrics.
The steps performed include:
- Create KFP components:
- Generate ROC curve and confusion matrix visualizations for classification results
- Write metrics
- Create KFP pipelines.
- Execute KFP pipelines
- Compare metrics across pipeline runs
```
[Lightweight Python function-based components, and component I/O](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/lightweight_functions_component_io_kfp.ipynb)
```
Learn to use the KFP SDK to build lightweight Python function-based components, and then you learn to use `Vertex AI Pipelines` to execute the pipeline.
The steps performed include:
- Build Python function-based KFP components.
- Construct a KFP pipeline.
- Pass *Artifacts* and *parameters* between components, both by path reference and by value.
- Use the `kfp.dsl.importer` method.
- Compile the KFP pipeline.
- Execute the KFP pipeline using `Vertex AI Pipelines`
```
[Custom training with pre-built Google Cloud Pipeline Components](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/custom_model_training_and_batch_prediction.ipynb)
```
Learn to use `Vertex AI Pipelines` and `Google Cloud Pipeline Components` to build a custom model.
The steps performed include:
- Create a KFP pipeline:
- Train a custom model.
- Upload the trained model as a `Model` resource.
- Create an `Endpoint` resource.
- Deploy the `Model` resource to the `Endpoint` resource.
- Make a batch prediction request.
```
[AutoML Tabular pipelines using google-cloud-pipeline-components](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/automl_tabular_classification_beans.ipynb)
```
@@ -88,68 +16,9 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Pipelines](https://cloud.google.com/vertex-ai/docs/pipelines/introduction).
[Training an acquisition-prediction model using Swivel, BigQuery ML and Vertex AI Pipelines](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/google_cloud_pipeline_components_bqml_text.ipynb)
```
Learn how to build a simple BigQuery ML pipeline using Vertex AI pipelines in order to calculate text embeddings of content from articles and classify them
into the *corporate acquisitions* category.
The steps performed include:
- Creating a component for Dataflow job that ingests data to BigQuery.
- Creating a component for preprocessing steps to run on the data in BigQuery.
- Creating a component for training a logistic regression model using BigQuery ML.
- Building and configuring a Kubeflow DSL pipeline with all the created components.
- Compiling and running the pipeline in Vertex AI Pipelines.
```
[Pipelines introduction for KFP](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/pipelines_intro_kfp.ipynb)
```
Learn how to use the KFP SDK to build pipelines that generate evaluation metrics.
The steps performed include:
- Define and compile a `Vertex AI` pipeline.
- Specify which service account to use for a pipeline run.
```
[AutoML tabular regression pipelines using google-cloud-pipeline-components](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/google_cloud_pipeline_components_automl_tabular.ipynb)
```
Learn to use `Vertex AI Pipelines` and `Google Cloud Pipeline Components` to build an `AutoML` tabular regression model.
The steps performed include:
- Create a KFP pipeline:
- Create a `Dataset` resource.
- Train an AutoML tabular regression `Model` resource.
- Create an `Endpoint` resource.
- Deploys the `Model` resource to the `Endpoint` resource.
- Compile the KFP pipeline.
- Execute the KFP pipeline using `Vertex AI Pipelines`
```
[Model upload, predict, and evaluate using google-cloud-pipeline-components](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/google_cloud_pipeline_components_model_upload_predict_evaluate.ipynb)
```
Learn how to evaluate a custom model using a pipeline with components from `google_cloud_pipeline_components` and a custom pipeline component you build.
The steps performed include:
- Upload a pre-trained model as a `Model` resource.
- Run a `BatchPredictionJob` on the `Model` resource with ground truth data.
- Generate evaluation `Metrics` artifact about the `Model` resource.
- Compare the evaluation metrics to a threshold.
```
&nbsp;&nbsp;&nbsp;Learn more about [AutoML components](https://cloud.google.com/vertex-ai/docs/pipelines/vertex-automl-component).
[Pipeline control structures using the KFP SDK](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/control_flow_kfp.ipynb)
@@ -166,58 +35,29 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Pipelines](https://cloud.google.com/vertex-ai/docs/pipelines/introduction).
[Loan eligibility prediction using `google-cloud-pipeline-components` and Spark ML](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/google_cloud_pipeline_components_dataproc_tabular.ipynb)
[Custom training with pre-built Google Cloud Pipeline Components](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/custom_model_training_and_batch_prediction.ipynb)
```
Learn how to build a Vertex AI pipeline and train a Random-forest model using Spark ML for loan-eligibility classification problem.
The steps performed include:
* Use the `DataprocPySparkBatchOp` to preprocess data.
* Create a Vertex AI dataset resource on the training data.
* Train a random forest model using Pyspark.
* Build a Vertex AI pipeline and run the training job.
* Use the Spark serving image in order to deploy a Spark model on Vertex AI Endpoint.
```
[Model train, upload, and deploy using Google Cloud Pipeline Components](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/google_cloud_pipeline_components_model_train_upload_deploy.ipynb)
```
Learn how to use `Vertex AI Pipelines` and `Google Cloud Pipeline Components` to build and deploy a custom model.
Learn to use `Vertex AI Pipelines` and `Google Cloud Pipeline Components` to build a custom model.
The steps performed include:
- Create a KFP pipeline:
- Train a custom model.
- Uploads the trained model as a `Model` resource.
- Creates an `Endpoint` resource.
- Deploys the `Model` resource to the `Endpoint` resource.
- Compile the KFP pipeline.
- Execute the KFP pipeline using `Vertex AI Pipelines`
```
[AutoML text classification pipelines using google-cloud-pipeline-components](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/google_cloud_pipeline_components_automl_text.ipynb)
```
Learn to use `Vertex AI Pipelines` and `Google Cloud Pipeline Components` to build an `AutoML` text classification model.
The steps performed include:
- Create a KFP pipeline:
- Create a `Dataset` resource.
- Train an AutoML text classification `Model` resource.
- Upload the trained model as a `Model` resource.
- Create an `Endpoint` resource.
- Deploys the `Model` resource to the `Endpoint` resource.
- Compile the KFP pipeline.
- Execute the KFP pipeline using `Vertex AI Pipelines`
- Deploy the `Model` resource to the `Endpoint` resource.
- Make a batch prediction request.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Pipelines](https://cloud.google.com/vertex-ai/docs/pipelines/introduction).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Training components](https://cloud.google.com/vertex-ai/docs/training/create-training-pipeline).
[Training and batch prediction with BigQuery source and destinantion for a custom tabular classification model](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/custom_tabular_train_batch_pred_bq_pipeline.ipynb)
@@ -240,3 +80,253 @@ The steps performed include:
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Pipelines](https://cloud.google.com/vertex-ai/docs/pipelines/introduction).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Batch Prediction components](https://cloud.google.com/vertex-ai/docs/pipelines/batchprediction-component).
[AutoML image classification pipelines using google-cloud-pipeline-components](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/google_cloud_pipeline_components_automl_images.ipynb)
```
Learn how to use `Vertex AI Pipelines` and `Google Cloud Pipeline Components` to build an `AutoML` image classification model.
The steps performed include:
- Create a KFP pipeline:
- Create a `Dataset` resource.
- Train an AutoML image classification `Model` resource.
- Create an `Endpoint` resource.
- Deploys the `Model` resource to the `Endpoint` resource.
- Compile the KFP pipeline.
- Execute the KFP pipeline using `Vertex AI Pipelines`
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Pipelines](https://cloud.google.com/vertex-ai/docs/pipelines/introduction).
&nbsp;&nbsp;&nbsp;Learn more about [AutoML components](https://cloud.google.com/vertex-ai/docs/pipelines/vertex-automl-component).
[AutoML tabular regression pipelines using google-cloud-pipeline-components](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/google_cloud_pipeline_components_automl_tabular.ipynb)
```
Learn to use `Vertex AI Pipelines` and `Google Cloud Pipeline Components` to build an `AutoML` tabular regression model.
The steps performed include:
- Create a KFP pipeline:
- Create a `Dataset` resource.
- Train an AutoML tabular regression `Model` resource.
- Create an `Endpoint` resource.
- Deploys the `Model` resource to the `Endpoint` resource.
- Compile the KFP pipeline.
- Execute the KFP pipeline using `Vertex AI Pipelines`
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Pipelines](https://cloud.google.com/vertex-ai/docs/pipelines/introduction).
&nbsp;&nbsp;&nbsp;Learn more about [AutoML components](https://cloud.google.com/vertex-ai/docs/pipelines/vertex-automl-component).
[AutoML text classification pipelines using google-cloud-pipeline-components](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/google_cloud_pipeline_components_automl_text.ipynb)
```
Learn to use `Vertex AI Pipelines` and `Google Cloud Pipeline Components` to build an `AutoML` text classification model.
The steps performed include:
- Create a KFP pipeline:
- Create a `Dataset` resource.
- Train an AutoML text classification `Model` resource.
- Create an `Endpoint` resource.
- Deploys the `Model` resource to the `Endpoint` resource.
- Compile the KFP pipeline.
- Execute the KFP pipeline using `Vertex AI Pipelines`
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Pipelines](https://cloud.google.com/vertex-ai/docs/pipelines/introduction).
&nbsp;&nbsp;&nbsp;Learn more about [AutoML components](https://cloud.google.com/vertex-ai/docs/pipelines/vertex-automl-component).
[Training an acquisition-prediction model using Swivel, BigQuery ML and Vertex AI Pipelines](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/google_cloud_pipeline_components_bqml_text.ipynb)
```
Learn how to build a simple BigQuery ML pipeline using Vertex AI pipelines in order to calculate text embeddings of content from articles and classify them
into the *corporate acquisitions* category.
The steps performed include:
- Creating a component for Dataflow job that ingests data to BigQuery.
- Creating a component for preprocessing steps to run on the data in BigQuery.
- Creating a component for training a logistic regression model using BigQuery ML.
- Building and configuring a Kubeflow DSL pipeline with all the created components.
- Compiling and running the pipeline in Vertex AI Pipelines.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Pipelines](https://cloud.google.com/vertex-ai/docs/pipelines/introduction).
&nbsp;&nbsp;&nbsp;Learn more about [BigQuery ML components](https://cloud.google.com/vertex-ai/docs/pipelines/bigqueryml-component).
[Loan eligibility prediction using `google-cloud-pipeline-components` and Spark ML](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/google_cloud_pipeline_components_dataproc_tabular.ipynb)
```
Learn how to build a Vertex AI pipeline and train a random-forest model using Spark ML for loan-eligibility classification problem.
The steps performed include:
* Use the `DataprocPySparkBatchOp` to preprocess data.
* Create a Vertex AI dataset resource on the training data.
* Train a random forest model using PySpark.
* Build a Vertex AI pipeline and run the training job.
* Use the Spark serving image in order to deploy a Spark model on Vertex AI Endpoint.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Pipelines](https://cloud.google.com/vertex-ai/docs/pipelines/introduction).
&nbsp;&nbsp;&nbsp;Learn more about [Dataproc components](https://cloud.google.com/vertex-ai/docs/pipelines/dataproc-component).
[Model train, upload, and deploy using Google Cloud Pipeline Components](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/google_cloud_pipeline_components_model_train_upload_deploy.ipynb)
```
Learn how to use `Vertex AI Pipelines` and `Google Cloud Pipeline Components` to build and deploy a custom model.
The steps performed include:
- Create a KFP pipeline:
- Train a custom model.
- Uploads the trained model as a `Model` resource.
- Creates an `Endpoint` resource.
- Deploys the `Model` resource to the `Endpoint` resource.
- Compile the KFP pipeline.
- Execute the KFP pipeline using `Vertex AI Pipelines`
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Pipelines](https://cloud.google.com/vertex-ai/docs/pipelines/introduction).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Training components](https://cloud.google.com/vertex-ai/docs/pipelines/customjob-component).
[Model upload, predict, and evaluate using google-cloud-pipeline-components](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/google_cloud_pipeline_components_model_upload_predict_evaluate.ipynb)
```
Learn how to evaluate a custom model using a pipeline with components from `google_cloud_pipeline_components` and a custom pipeline component you build.
The steps performed include:
- Upload a pre-trained model as a `Model` resource.
- Run a `BatchPredictionJob` on the `Model` resource with ground truth data.
- Generate evaluation `Metrics` artifact about the `Model` resource.
- Compare the evaluation metrics to a threshold.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Pipelines](https://cloud.google.com/vertex-ai/docs/pipelines/introduction).
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Model components](https://cloud.google.com/vertex-ai/docs/pipelines/model-endpoint-component).
[Lightweight Python function-based components, and component I/O](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/lightweight_functions_component_io_kfp.ipynb)
```
Learn to use the KFP SDK to build lightweight Python function-based components, and then you learn to use `Vertex AI Pipelines` to execute the pipeline.
The steps performed include:
- Build Python function-based KFP components.
- Construct a KFP pipeline.
- Pass *Artifacts* and *parameters* between components, both by path reference and by value.
- Use the `kfp.dsl.importer` method.
- Compile the KFP pipeline.
- Execute the KFP pipeline using `Vertex AI Pipelines`
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Pipelines](https://cloud.google.com/vertex-ai/docs/pipelines/introduction).
[Metrics visualization and run comparison using the KFP SDK](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/metrics_viz_run_compare_kfp.ipynb)
```
Learn how to use the KFP SDK to build pipelines that generate evaluation metrics.
The steps performed include:
- Create KFP components:
- Generate ROC curve and confusion matrix visualizations for classification results
- Write metrics
- Create KFP pipelines.
- Execute KFP pipelines
- Compare metrics across pipeline runs
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Pipelines](https://cloud.google.com/vertex-ai/docs/pipelines/introduction).
[Pipelines introduction for KFP](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/pipelines_intro_kfp.ipynb)
```
Learn how to use the KFP SDK to build pipelines that generate evaluation metrics.
The steps performed include:
- Define and compile a `Vertex AI` pipeline.
- Specify which service account to use for a pipeline run.
```
&nbsp;&nbsp;&nbsp;Learn more about [Vertex AI Pipelines](https://cloud.google.com/vertex-ai/docs/pipelines/introduction).
[BQML and AutoML - Experimenting with Vertex AI](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/pipelines/rapid_prototyping_bqml_automl.ipynb)
```
Learn how to use `Vertex AI Predictions` for rapid prototyping a model.
The steps performed include:
- Creating a BigQuery and Vertex AI training dataset.
- Training a BigQuery ML and AutoML model.
- Extracting evaluation metrics from the BigQueryML and AutoML models.
- Selecting the best trained model.
- Deploying the best trained model.
- Testing the deployed model infrastructure.
```
&nbsp;&nbsp;&nbsp;Learn more about [AutoML components](https://cloud.google.com/vertex-ai/docs/pipelines/vertex-automl-component).
[Train custom tabular ML models with many frameworks and import to Vertex AI using Vertex Pipelines](https://github.com/GoogleCloudPlatform/vertex-ai-samples/tree/main/notebooks/official/pipelines/Train_tabular_models_with_many_frameworks_and_import_to_Vertex_AI_using_Pipelines)
Learn how to build a pipeline that does the following:
* Ingest data
* Transform data
* Clean up data
* Split data into train/test subsets
* Configure model
* Train model using multiple ML frameworks
* Import model into Vertex Model Registry
* [Optional] Deploy model to Vertex Endpoints for serving
Included pipelines:
* Train ML model
* * Tabular classification
* * * TensorFlow
* * * PyTorch
* * * XGBoost
* * * Scikit-learn
* * Tabular regression
* * * TensorFlow
* * * PyTorch
* * * XGBoost
* * * Scikit-learn

Some files were not shown because too many files have changed in this diff Show More