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

Kedro

I was curious if we needed to do anything when using kedro viz 6.4.0 to get the data set previews to show up?  Is it restricted to any specific dataset types?

Kedro-Viz version 6.3.0 currently supports *preview* of two types of *dataset*s: `pandas.CSV*Dataset*` and `pandas.Excel*Dataset*`

is that out of scope for what you’re expecting?

we had `PartitionedDataset` of `pickled.PickledDataset` that had the pickled dataframes and then tried converting them over to  `PartitionedDataset` of `pandas.ParquetDataset`

Then tried just a single `pandas.ParquetDataset`

but if it is just the two datasets right now that makes sense

but one of the nodes has a CSV input.  Is there a size restriction?

<https://docs.kedro.org/en/stable/visualisation/preview_datasets.html#configure-the-data-catalog>

No worries.  You guys are doing great in our eyes.  The product gets better and better :slightly_smiling_face:

image.png

Thanks!  that worked :slightly_smiling_face:

Do you happen to know if there is a roadmap that has if/when the other datasets might be available?  do you know if `PartitionedDataset`  will ever show up with a preview?

So the best thing you can do is a raise an issue on `kedro-viz`  repo this is exactly the kind of feature we’ve held off fleshing out since we wanted to gauge demand first

when it comes to partitioned dataset I think we need to design it a bit first - what would you like to see? The latest partition? All partitions? etc

Similar to the current preview based on the dataset type, but maybe have a list of the files and when you select one the preview is loaded.

ill get an issue opened tonight when i get home with details for this and other data sets too.

Thanks <@U05CTGE082K> for the feedback and opening the issue. I have some follow up questions :slightly_smiling_face:
1. How does the dataset preview feature impact your/team's workflow?
2. Besides `PartitionedDataset` which other datasets do you commonly use in your workflow?
3. For the `n_rows` , what size would be ideal for you?
It would great to have a brief user feedback interview to explore your experience with Kedro-Viz, this feature, and any other pain points :slightly_smiling_face:.