Rest API: provide example for dataset with with row-id and ignored features.

Not sure where to put this issue, since it is documentation-related on the one hand but not part of the docs.openml.org page on the other.

The json response to /task/{id} may sometimes contain the ignore_attribute and row_id_attribute elements, but they are missing for most tasks. This is currently not documented anywhere on the only kind of reference documentation about the REST API that I could find, the "REST" tab of this page. It would be ideal to have this in the "schema" section, and to also provide an example ID for which they are present in the "Actions" Example Value display.

One dataset that I found that had a row_id_attribute was dataset 164. For ignore_attribute, one example is dataset 185, and one example with multiple ignore_attribute is dataset 940.

(It would be nice to have a set of IDs for datasets and probably also other entities that cover all of these special issues, so that someone trying to develop against the API could test against these and then be reasonably sure that the implementation is robust.)

Thanks! I added a section on special attributes in the new BootCamp docs: https://docs.openml.org/#special-attributes I also included your examples there.

In the spec it is originally part of the data_features schema. We also return it (for convenience) in the dataset and task descriptions. You're right that this should also be documented there. @PGijsbers Can we include this in the next version of the API schema?

@PGijsbers Can we include this in the next version of the API schema?

At this point I can't say anything about the particular fields of tasks and other entities as we will re-evaluate which attributes are appropriate for each entity, so it may be subject to change. However, the new API will be completely documented (and almost surely be accompanied with JSON Schema).