Do you know the History of Data Science? READ MORE

Bulk import managed datasets

Add Dataset > Internal > Dataset from another project is a great feature. It allows users to quickly search the data catalog for a dataset to then import. However, it's common to need to import many datasets at once. It would be great if I could select multiple datasets from the data catalog screen and import them at once, similar to what's possible with the connection explorer. This is especially nice when the datasets happen to be managed datasets from other projects.

For example, general processing pipelines to turn raw data into semantic data might exist in one project. Other projects would like to use the resulting data. Optimally, the semantic datasets can remain managed datasets and projects remain connected to one another this way. Thus having a feature set for bulk importing managed datasets like is currently available for importing unmanaged datasets would be really helpful. My team relies pretty heavily on data that uses a process like this, and we're getting bottlenecked when we need to bring in 50-100 datasets from our semantic layer project.

1 Comment
fsergot
Dataiker
Dataiker
Status changed to: In Backlog

I have added the multi-dataset import in the create dataset context, it is indeed interesting in your context where you have built a master data project and want all your projects to use the pre-processed data from there.