Added 'data_dict' attribute (DataDictDataset) to AbstractVersionedDat…

[noamgoldberg]

…aset

Description

Development notes

Developer Certificate of Origin

We need all contributions to comply with the Developer Certificate of Origin (DCO). All commits must be signed off by including a Signed-off-by line in the commit message. See our wiki for guidance.

If your PR is blocked due to unsigned commits, then you must follow the instructions under "Rebase the branch" on the GitHub Checks page for your PR. This will retroactively add the sign-off to all unsigned commits and allow the DCO check to pass.

Checklist

• Read the contributing guidelines
• Signed off each commit with a Developer Certificate of Origin (DCO)
• Opened this PR as a 'Draft Pull Request' if it is work-in-progress
• Updated the documentation to reflect the code changes
• Added a description of this change in the RELEASE.md file
• Added tests to cover my changes
• Checked if this change will affect Kedro-Viz, and if so, communicated that with the Viz team

[astrojuanlu]

Hi @noamgoldberg, thanks for your PR ! Could you explain the rationale behind this? What problem does it solve?

[noamgoldberg]

Hi @noamgoldberg, thanks for your PR ! Could you explain the rationale behind this? What problem does it solve?

I use kedro a lot for personal projects, and it's helpful to have a data dictionary attached to large datasets. For example, I like to create data_dict.yml with the feature descriptions, ranges, and general source information, to be referenced in Jupyter notebooks and used dynamically in the code (i.e. visualizations, reports). The DataDictDataset class is rather straightforward AbstractDataset, but the unique and helpful change in this PR is the enablement of the attachment of an instance of DataDictDataset to other datasets inheriting from AbstractVersionedDataset (i.e. pandas.CSVDataSet). For example, this would enable the following entry in catalog.yml:

    stocks_data:
        type: pandas.CSVDataSet
        filepath: data/01_raw/stocks.csv
        data_dict:
            dataset: yaml.YAMLDataSet
            filepath: data/01_raw/data_dict.yml

This would create a dataset stocks_data with an attached data dictionary.

[merelcht]

@noamgoldberg so this data_dict basically contains metadata about the dataset?

[noamgoldberg]

@merelcht yes :) I mainly use it for feature definitions and basic dataset information (i.e. author, source, location/date created)

[astrojuanlu]

Hi @noamgoldberg, sorry it took us so long to get back to you.

IIUC, the data_dict you propose here already exists and it's called metadata. See an example here:

https://docs.kedro.org/projects/kedro-viz/en/latest/kedro-viz_visualisation.html#visualise-layers

Please confirm if that would suit your needs. Arguably we could do a better job at documenting it, most likely here: https://docs.kedro.org/en/stable/data/data_catalog.html