K Hi @here, this is a quick announcement that Kedro 0.18.4 is out! K

TLDR: In this release, we introduce the option of using

kedro-datasets

as its own package, easier ways to access and update project configurations, new datasets to work with svmlight/libsvm files and video objects and more!

Kedro 0.18.4 is a non-breaking release in the 0.18.x series, which means you can upgrade with:

pip install kedro==0.18.4

or

conda install -c conda-forge kedro==0.18.4

🚀 Major Features/Changes • We have released

kedro-datasets

as its own package (Check it out here!) and Kedro now instantiates datasets from

kedro_datasets

with a higher priority than

kedro.extras.datasets

. • The config loader now implements

UserDict

and the configuration can now be accessed through

conf_loader['catalog']

. • You can configure config file patterns through

settings.py

without having to create a custom config loader. • Support for a functional definition of schema added to

dask.ParquetDataSet

to work with

<http://dask.to|dask.to>_parquet

API. • Our documentation has had a revamp, including a new and improved Spaceflights tutorial. • New datasets added to our collection - ◦

svmlight.SVMLightDataSet

◦

video.VideoDataSet

◦

video.video_dataset.SequenceVideo

◦

video.video_dataset.GeneratorVideo

🪲 Bug fixes and other changes • Fixed

kedro micropkg pull

for packages on PyPI. • Fixed

format

in

save_args

for

SparkHiveDataSet

, previously it didn’t allow you to save it as delta format. • Fixed save errors in

TensorFlowModelDataset

when used without versioning; previously, it wouldn’t overwrite an existing model. • The Kedro IPython extension now surfaces errors when it cannot load a Kedro project. • And more … You can find the full list of changes in the release notes: https://github.com/kedro-org/kedro/releases/tag/0.18.4

Thanks for great release! BTW, I’ve created a custom dataset for geopandas to access data from PostGIS sql database. It is initially based on quite similar

pandas.SQLQueryDataSet

. Can this contribution be of any value for Kedro/Your team and if yes what are the contribution guidelines? thanks!

Hi Eugene! A contribution for a dataset would now go into the `kedro-datasets` repository. Some useful docs for contributions: • How to contribute a custom dataset • Developer contributor guidelines