v2.1

What's Changed

• Explicitly set dtype when reading data by @dstansby in #172. This fixes a bug where all output data would be created with float64 dtype.
• Allow passing a custom downsample function by @dstansby in #173. This makes it easier to downsample label data, by passing the new stack_to_chunk.mode function as a downsample_funciton.

Full Changelog: v2.0...v2.1

v2.0

Merge pull request #167 from HiPCTProject/downsample-with-ts

Use tensorestore for downsampling

v2.0b2

Merge pull request #167 from HiPCTProject/downsample-with-ts

Use tensorestore for downsampling

v2.0b1

Merge pull request #164 from HiPCTProject/dependabot/github_actions/g…

v1.10

What's Changed

• Bump the python group across 1 directory with 4 updates by @dependabot[bot] in #136

Full Changelog: v1.9...v1.10

v1.9

What's Changed

• Bump ome-zarr-models version by @dstansby in #129. This prevents invalid OME-Zarr metadata from being written.

Full Changelog: v1.8...v1.9

v1.8

What's Changed

stack-to-chunk now uses ome-zarr-models to validate and write metadata. This has resulted in a breaking change in setting up datasets:

• When creating a MultiScaleGroup, an array specification must now be provided for the original full resolution data.
• MultiScaleGroup.create_initial_dataset no longer exists - the initial dataset is now automatically created when the MultiScaleGroup is created.

Other changes:

• Use ome-zarr-models for writing metadata by @dstansby in #99
• Raise helpful error on wrong voxel tuple size by @dstansby in #127
• Bump the python group across 1 directory with 2 updates by @dependabot in #128

Full Changelog: v1.7...v1.8

v1.7

What's Changed

• Add extra test config by @dstansby in #106
• Clean up pyproject.toml by @dstansby in #107
• Test on windows by @dstansby in #109
• Update dep numcodecs by @chourroutm in #121

New Contributors

• @chourroutm made their first contribution in #121

Full Changelog: v1.6...v1.7

v1.6

What's Changed

Documentation

• Add some more docs to add_downsample_level by @dstansby in #83

Dependencies

Updated:

• dask to 2025.2.0
• numpy to 2.2.3
• scikit-image to 0.25.1

Full Changelog: v1.5...v1.6

v1.5

What's Changed

‼️Data produced with older versions of stack-to-chunk did not have the correct order of transfroms. When opening a multiscale group using open_multiscale_group stack-to-chunk will now attempt to fix this by re-writing the transform metadata to be correct.

• Fix transform order in OME-zarr metadata by @dstansby in #86
• Allow opening arbitrary stores by @dstansby in #88

Dependencies:

• Bump zarr from 2.18.2 to 2.18.3 by @dependabot in #62
• Bump dask from 2024.6.2 to 2024.8.2 by @dependabot in #64
• Bump dask from 2024.8.2 to 2024.11.2 by @dependabot in #73
• Bump numpy from 2.1.2 to 2.1.3 by @dependabot in #69
• Bump loguru from 0.7.2 to 0.7.3 by @dependabot in #78
• Bump dask from 2024.11.2 to 2024.12.0 by @dependabot in #79
• Bump zarr from 2.18.3 to 2.18.4 by @dependabot in #82
• Bump numpy from 2.1.3 to 2.2.0 by @dependabot in #80

Documentation

• Use .ome.zarr for group suffix in tutorial by @dstansby in #74

Full Changelog: v1.4...v1.5