0.5.7

0.5.7

ZenML 0.5.7 is here 💯 and it brings not one, but 🔥TWO🔥 brand new integrations 🚀! ZenML now support MLFlow for tracking pipelines as experiments and Evidently for detecting drift in your ML pipelines in production!

New Features

• Introducing the MLFLow Tracking Integration, a first step towards our complete MLFlow Integration as described in the #115 poll. Full example found here.

• Introducing the Evidently integration. Use the standard Evidently drift detection step to calculate drift automatically in your pipeline. Full example found here.

Bugfixes

• Prevent KFP install timeouts during stack up by @stefannica in #299
• Prevent naming parameters same name as inputs/outputs to prevent kwargs-errors by @bcdurak in #300

What's Changed

• Force pull overwrites local examples without user confirmation by @AlexejPenner in #278
• Updated README with latest features by @htahir1 in #280
• Integration test the examples within ci pipeline by @AlexejPenner in #282
• Add exception for missing system requirements by @kamalesh0406 in #281
• Examples are automatically pulled if not present before any example command is run by @AlexejPenner in #279
• Add pipeline error for passing the same step object twice by @kamalesh0406 in #283
• Create pytest fixture to use a temporary zenml repo in tests by @htahir1 in #287
• Additional example run implementations for standard interfaces, functional and class based api by @AlexejPenner in #286
• Make pull_request.yaml actually use os.runner instead of ubuntu by @htahir1 in #288
• In pytest return to previous workdir before tearing down tmp_dir fixture by @AlexejPenner in #289
• Don't raise an exception during integration installation if system requirement is not installed by @schustmi in #291
• Update starting page for the API docs by @alex-zenml in #294
• Add stack up failure prompts by @alex-zenml in #290
• Spelling fixes by @alex-zenml in #295
• Remove instructions to git init from docs by @bcdurak in #293
• Fix the stack up and orchestrator up failure prompts by @stefannica in #297
• Prevent KFP install timeouts during stack up by @stefannica in #299
• Add stefannica to list of internal github users by @stefannica in #303
• Improve KFP UI daemon error messages by @schustmi in #292
• Replaced old diagrams with new ones in the docs by @AlexejPenner in #306
• Fix broken links & text formatting in docs by @alex-zenml in #302
• Run KFP container as local user/group if local by @stefannica in #304
• Add james to github team by @jwwwb in #308
• Implement integration of mlflow tracking by @AlexejPenner in #301
• Bugfix integration tests on windows by @jwwwb in #296
• Prevent naming parameters same name as inputs/outputs to prevent kwargs-errors by @bcdurak in #300
• Add tests for fileio by @alex-zenml in #298
• Evidently integration (standard steps and example) by @alex-zenml in #307
• Implemented evidently integration by @stefannica in #310
• Make mlflow example faster by @AlexejPenner in #312

New Contributors

• @kamalesh0406 made their first contribution in #281
• @stefannica made their first contribution in #297
• @jwwwb made their first contribution in #308