Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Create a use case for the gallery #67

Open
mllg opened this issue Mar 10, 2020 · 4 comments
Open

Create a use case for the gallery #67

mllg opened this issue Mar 10, 2020 · 4 comments

Comments

@mllg
Copy link
Member

mllg commented Mar 10, 2020

No description provided.

@sebffischer
Copy link
Member

I suggest using this dataset (5000 obs, 1k features)

@mllg
Copy link
Member Author

mllg commented Aug 8, 2022

Are you suggesting it because of its dimension or because filtering works pretty well on it?

What about spam in mlr3?

@pat-s
Copy link
Member

pat-s commented Aug 8, 2022

spam would also work I'd say. In the end the question is what the use case should show:

  • compare filters against wrappers?
  • compare single filters against ensemble filters?
  • showcase the caching attribute of filters?

None of the above do really depend on the dataset unless you want to find a use case where features outperform wrappers.

@sebffischer
Copy link
Member

sebffischer commented Aug 8, 2022

Yes, my idea was because there are so many features.
I would prefer a regression task (because then I can use it for MaRDI), that would make spam less attractive for me.

Are there already benchmark studies on those questions @pat-s ?
In this case we could make a gallery-posts that gives some recommendations on what to do with filters?
I.e. giving the reader the answer to those questions :)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

3 participants