[ENH] Added GluonTS datasets as sktime datatypes#6530
Conversation
|
@fkiraly, I started work on adding in the GluonTS datatypes! Could you please let me know if I am making changes in the correct place and if so, what a rough roadmap could look like to add complete support for the |
|
@fkiraly, sorry about the repeated messages! I am trying to utilize the Is it possible the function is possibly outdated? Thank you for the help! |
fkiraly
left a comment
There was a problem hiding this comment.
I think you are in the right module, but for time series, you need to use the _panel and _series directories, the Panel, and Series scitypes.
Series is single time series, Panel is collection of time series.
It is very possible that the existing functions are outdated - I do not think any tests have been added since the converters were contributed in 2022, so it is very much possible that they need to be updated.
|
@fkiraly, could you please review the PR and let me know if it all looks good? Thank you so much! 😄 |
fkiraly
left a comment
There was a problem hiding this comment.
Looks good to me! Great!
Reference Issues/PRs
Fixes #2860
What does this implement/fix? Explain your changes.
This is an implementation that adds support for GluonTS implemented data types in the sktime library to make it easier for users to utilize both libraries together.
Does your contribution introduce a new dependency? If yes, which one?
I believe GluonTS was added as soft dependency in an earlier PR, but I will check this.
What should a reviewer concentrate their feedback on?
A reviewer should focus on my implementations in the
sktime/datatypes/_tabledirectory, specifically in the_registry.pyand_check.pyfiles.Did you add any tests for the change?
Not yet, but planning on doing so!
Any other comments?
Hello everyone! I have tried to add data types on my own, but I am not sure if I am doing it correctly. I have a feeling that I will need to add more functionality to fully add support for
ListDataset(planning on addingFileDatasetas well to support larger datasets which likely might be common for many use cases) but I would really appreciate feedback on if I have the right idea of how to add a new datatype to sktime.PR checklist
For all contributions
How to: add yourself to the all-contributors file in the
sktimeroot directory (not theCONTRIBUTORS.md). Common badges:code- fixing a bug, or adding code logic.doc- writing or improving documentation or docstrings.bug- reporting or diagnosing a bug (get this pluscodeif you also fixed the bug in the PR).maintenance- CI, test framework, release.See here for full badge reference