Kedro is an open-sourced Python framework for creating maintainable and modular data science code.

Kedro

image.png

Hi all, do you know if we will have support for using `partition_cols` when saving parquet datasets? At the moment custom save and load methods are required which is quite cumbersome.

P.S. in the documentation <https://docs.kedro.org/projects/kedro-datasets/en/kedro-datasets-2.1.0/api/kedro_datasets.pandas.ParquetDataset.html|here> (pictured below) it has YAML examples using `partition_on` even though this is not supported.

It’s just a thin wrapper over `<http://pd.DataFrame.to|pd.DataFrame.to>_parquet` isn’t this available ?
<https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.to_parquet.html>

but is this possible to use without defining a custom dataset?

It should’ve possible today do you get an error?

hi <@U054YHLTY4E>, as far as I understand, you could specify

```ds:
  type: pandas.ParquetDataset
  save_args:
    partition_cols: ...```
could you try?

&gt;  in the documentation <https://docs.kedro.org/projects/kedro-datasets/en/kedro-datasets-2.1.0/api/kedro_datasets.pandas.ParquetDataset.html|here> (pictured below) it has YAML examples using `partition_on` even though this is not supported.
what do you mean? does it give an error?