Features / Enhancements
- Dependencies: Move all sqlalchemy code to framework to prevent sqlachemy dependency in mlrun, #8251, @adamdelman
- Pipelines: Enhance retry pipeline api to create rerunrunner for remote workflow retries, #8239, @yaelgen
- Model Monitoring: Add
function-summaryapi, #8241, @Eyal-Danieli - Requirements-automated: Upgrade lock files, #8263, @iguazio-cicd
- Dependabot-automated: Update mlrun-pipelines-kfp-common requirement from ~=0.5.6 to ~=0.5.8, #8260, @dependabot[bot]
- Dependabot-automated: Bump anchore/scan-action from 6.3.0 to 6.4.0, #8261, @dependabot[bot]
- Dependabot-automated: Update mlrun-pipelines-kfp-v1-8 requirement from ~=0.5.4 to ~=0.5.7, #8259, @dependabot[bot]
- Serving: Implement serving graph as a job, #8209, @gtopper
- Model Serving: Move execution mechanism from
modeltoadd_model, #8172, @gtopper - Model Monitoring: Support all model endpoints in
evaluate, #8218, @jond01 - Serving: Added special plot for
modelrunnerstep, #8235, @royischoss - Datastore: Change get_or_create to be more flexible, #8214, @tomerm-iguazio
- Datastore: Make spark s3 credentials bucket aware, #8240, @alxtkr77
- Unknown: Revert "[k8s] configure retries for kubernetes client using urllib3", #8242, @liranbg
- DB: Improve db initialization, #8160, @adamdelman
- Tests: Make postgresql support optional for testing, #8161, @adamdelman
- Dependabot-automated: Bump ncipollo/release-action from 1.16.0 to 1.18.0, #8236, @dependabot[bot]
- DB: Properly create db partitions when creating the db from scratch, #8205, @adamdelman
- Serving: Allow
modelrunnerstep:add_modelto get model_class asmodelobject, #8183, @royischoss - K8S: Configure retries for kubernetes client using urllib3, #8220, @adamdelman
- Docs: Add databricks runtime to cheat-sheet.md, #8227, @tomerm-iguazio
- Requirements-automated: Upgrade lock files, #8224, @iguazio-cicd
- Requirements: Temporarily pin storey version, #8225, @gtopper
- Requirements: Add remote models requirements, including openai sdk, #8215, @tomerm-iguazio
- DB: Remove unused tables, #8217, @adamdelman
- UI: Features & enhancement
Bug fixes
- Bugfix: Mlrun k8s client raises an ssl error, #8237, @adamdelman
- Frameworks:
pre_compilehorovod to tensorflow adjustments bug fixes, #8258, @guy1992l - Model Monitoring: Fix model endpoint tsdb data deletion, #8254, @royischoss
- Tests: Fix
test_app_flowsystem test withwith_training_set=false, #8249, @jond01 - Serving: Fix allow multiple deployment using copy in
modelobj:from_dict, #8247, @royischoss - Tests: Fix postgres conditional test skip based on import, #8245, @adamdelman
- DB: Fix return uid value from db as hex, #8243, @royischoss
- Artifacts: Fix artifact deletion by uid, #8233, @moranbental
- Bugfix: Don't flood mlrun-chief log with "failed to load kube config", #8232, @adamdelman
- UI: Bug fixes
Pull requests:
fe4d073 [Bugfix] MLRun K8S client raises an SSL error (#8237)
5d3890f [Dependencies] Move all sqlalchemy code to framework to prevent sqlachemy dependency in mlrun (#8251)
bb98e45 [Pipelines] Enhance retry pipeline API to create RerunRunner for remote workflow retries (#8239)
bb8f6c0 [Model Monitoring] Add function-summary API (#8241)
1417f75 [Requirements-automated] Upgrade lock files (#8263)
579bd85 [Dependabot-automated] Update mlrun-pipelines-kfp-common requirement from ~=0.5.6 to ~=0.5.8 (#8260)
f3697e4 [Dependabot-automated] Bump anchore/scan-action from 6.3.0 to 6.4.0 (#8261)
d50ca12 [Frameworks] pre_compile Horovod to TensorFlow adjustments bug fixes (#8258)
be1d6cd [Dependabot-automated] Update mlrun-pipelines-kfp-v1-8 requirement from ~=0.5.4 to ~=0.5.7 (#8259)
229b4f5 [Serving] Implement serving graph as a job (#8209)
4781b51 [Model Serving] Move execution mechanism from Model to add_model (#8172)
22fdc68 [Model Monitoring] Fix model endpoint tsdb data deletion (#8254)
dd0a52f [Model Monitoring] Support all model endpoints in evaluate (#8218)
5cb5543 [Tests] Fix test_app_flow system test with with_training_set=False (#8249)
de88a95 [Serving] Fix allow multiple deployment using copy in ModelObj:from_dict (#8247)
f242741 [Serving] Added special plot for ModelRunnerStep (#8235)
f4f3002 [Datastore] change get_or_create to be more flexible (#8214)
71a1d34 [Datastore] Make Spark S3 credentials bucket aware (#8240)
b8be9f7 [Tests] Fix postgres conditional test skip based on import (#8245)
dee8e46 Revert "[K8S] Configure retries for Kubernetes client using urllib3" (#8242)
a7257a3 [DB] Fix return uid value from db as hex (#8243)
b78b287 [DB] Improve DB initialization (#8160)
8c458ff [Tests] Make PostgreSQL support optional for testing (#8161)
1c292a6 [Dependabot-automated] Bump ncipollo/release-action from 1.16.0 to 1.18.0 (#8236)
df6e1a8 [DB] Properly create DB partitions when creating the DB from scratch (#8205)
ae8daf9 [Serving] Allow ModelRunnerStep:add_model to get model_class as Model object (#8183)
548880b [Artifacts] Fix Artifact Deletion by UID (#8233)
258cece [Bugfix] Don't flood mlrun-chief log with "Failed to load kube config" (#8232)
4241cb8 [K8S] Configure retries for Kubernetes client using urllib3 (#8220)
9ca7724 [Docs] Add databricks runtime to cheat-sheet.md (#8227)
03489d2 [Requirements-automated] Upgrade lock files (#8224)
c524e35 [Requirements] Temporarily pin storey version (#8225)
66ae141 [Requirements] Add Remote models requirements, including OpenAI SDK (#8215)
59ec753 [DB] Remove unused tables (#8217)