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

Kedro

I would like to bring `polars` to our `kedro` pipeline, but I don't want to make a data catalog load files only with polars (`type: polars.CSVDataSet`); doing so, everyone else must load it as `polars` too.
Is there a way in `kedro` to be able to use both formats in some way or should I create a custom dataset?

You could create 2 entries in the data catalog, one for `polars` and one for (let's say) `pandas` but both pointing to the same `.csv` file.

Then the user can choose which one to use when loading data to a notebook (for instance)

Nonetheless, for the actual pipelines you will need to choose which one to use as default, your node's code will depend on which dataframe library you are using.

<https://docs.kedro.org/en/stable/data/data_catalog_yaml_examples.html#read-the-same-file-using-two-different-datasets>

```my_dataframe@pandas:
  type: pandas.CSVDataSet
  filepath: data.csv

my_dataframe@polars:
  type: POLARS
  filepath: data.csv```