v0.5.0

What’s Changed

🔧 Maintenance

• Remove sphinx_markdown_tables @benfred (#106)
• Don't git pull origin main in unit tests, use container version @karlhigley (#105)

v0.4.0

What’s Changed

🐜 Bug Fixes

• Update merlin imports to specify modules to import @oliverholworthy (#100)
• Update imports in dispatch module to handle missing GPU @oliverholworthy (#99)
• Update GPU detection in merlin.core.utils for Distributed class @oliverholworthy (#98)

📄 Documentation

• Update Distributed docstring class - Rename new_cluster to cluster_type @oliverholworthy (#95)
• chore: Add release-drafter @mikemckiernan (#88)

🔧 Maintenance

• Add a GA workflow that requires labels on PR's @benfred (#101)
• Revert "Add --python argument to conda build" @oliverholworthy (#97)
• Update setup.py to include only the merlin namespace @oliverholworthy (#96)
• Request adding labels to PRs @mikemckiernan (#90)
• Adds a common triage workflow @benfred (#92)
• Add conda builds to our github actions workflow @benfred (#87)

v0.3.0

What's Changed

• Upgrading black by @marcromeyn in #62
• Improve schema representation in a notebook by @marcromeyn in #69
• Standardize our pre-commit hooks by @karlhigley in #71
• Add mypy to the commit hooks and address typing issues by @karlhigley in #72
• Add Github issue templates by @karlhigley in #73
• Schema apply to select refactor by @jperez999 in #74
• update distributed and dask to new version by @jperez999 in #77
• docs: Add ext-toc and copydirs by @mikemckiernan in #79
• Change dask and distributed deps to ==2021.11.2 by @karlhigley in #78
• req fix slim down by @jperez999 in #81
• Fix for cudf 22.04 by @benfred in #82
• arrow downgrade in requirements by @jperez999 in #84

Full Changelog: v0.2.0...v0.3.0

v0.2.0

What's Changed

• Add a conda recipe for uploading to anaconda by @benfred in #32
• Add a basic sphinx documentation site by @benfred in #31
• Automate pushing package to pypi by @benfred in #30
• Set up automated docstring coverage checks by @karlhigley in #34
• Make BaseOperator.dependencies a property by @karlhigley in #33
• Add docstrings to merlin.schema by @karlhigley in #35
• chore(docs): configure doc preview for PRs by @mikemckiernan in #36
• Fix reading from S3 by @benfred in #37
• Update requirements.txt by @albert17 in #41
• Adds container testing by @albert17 in #42
• Fix JSON serialization of schema's by @benfred in #43
• Split schema tests from dag tests by @benfred in #44
• Move interrogate docstring coverage settings to pyproject.toml by @karlhigley in #45
• Fix an issue with fetching leaf dtypes from multiply nested lists by @karlhigley in #46
• Exclude additional directories from docstring coverage checks by @karlhigley in #49
• Additional docstrings from bash by @karlhigley in #50
• docs: Add nightly multi-version build by @mikemckiernan in #51
• Move merlin.core.worker to merlin.io.worker by @karlhigley in #52
• Switch to packaging.version from distutils.LooseVersion by @benfred in #54
• README updates by @benfred in #53
• Fix for cudf 22.02 by @albert17 in #57
• Generalize the fix for Pandas extension dtypes by @karlhigley in #58
• Restrict running on pandas 1.4.x by @benfred in #59
• Exclude merlin/io from docstring coverage checks by @karlhigley in #56
• Avoid nullable pandas dtypes in CPUParquetEngine by @rjzamora in #60
• Fix series_has_nulls dispatch by @rjzamora in #61
• Fix type-hints of schema.select_by_tag & remove_by_tag by @marcromeyn in #63
• Avoid creating pd.StringDtype in CPUParquetEngine by @rjzamora in #66
• Remove use of kind property in ColumnSchema by @karlhigley in #65
• Add pd.StringDtype check back to schema post_init by @rjzamora in #67
• docs: Add redirect page by @mikemckiernan in #64
• Report skipped tests in blossom by @benfred in #68

New Contributors

• @mikemckiernan made their first contribution in #36
• @albert17 made their first contribution in #41
• @marcromeyn made their first contribution in #63

Full Changelog: v0.1.1...v0.2.0

v0.1.0

What's Changed

• Add CI with github actions by @benfred in #1
• Make utils._pynvml_mem_size public by @karlhigley in #2
• Migrate Schema class to use better-proto for IO by @karlhigley in #3
• Check Schema roundtrip to/from TensorflowMetadata by @benfred in #7
• Allow JSON serialization for TensorflowMetadata by @benfred in #8
• Fixes for TensorflowMetadata JSON compatability with merlin models by @benfred in #9
• Add properties and tags for compatability with merlin-models by @benfred in #10
• Move Schema from merlin.graph to merlin.schema by @benfred in #11
• Store dtype precision as itemsize in schema files by @jperez999 in #14
• Use implicit namespace package for merlin by @benfred in #15
• Update setup.py to use find_namespace_packages() by @karlhigley in #16
• Add Distributed and Serial execution-manager utilities by @rjzamora in #21
• Rename merlin.graph to merlin.dag by @karlhigley in #22
• Make it possible to pass device through pull_apart_list by @karlhigley in #26

New Contributors

• @benfred made their first contribution in #1
• @jperez999 made their first contribution in #14
• @rjzamora made their first contribution in #21

Full Changelog: https://github.com/NVIDIA-Merlin/core/commits/v0.1.0