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.

Sign up for GitHub

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

Jump to bottom

[ENH] Add various Kolmogorov-Arnold Network using PyKAN #6390

Open
3 tasks
benHeid opened this issue May 5, 2024 · 5 comments
Open
3 tasks

[ENH] Add various Kolmogorov-Arnold Network using PyKAN #6390

benHeid opened this issue May 5, 2024 · 5 comments
Labels
enhancement Adding new functionality implementing algorithms Implementing algorithms, estimators, objects native to sktime module:classification classification module: time series classification module:forecasting forecasting module: forecasting, incl probabilistic and hierarchical forecasting
Projects
Workstream: Deep Learning based Forec...

Comments

@benHeid
Copy link
Contributor

benHeid commented May 5, 2024 •
edited

Is your feature request related to a problem? Please describe.
This issue proposes to add support for pyKAN based models

Potential Estimators:

  •  PyKAN based forecaster (simple forward KAN) [ENH] pykan based forecaster #6386
  • PyKAN recurrent forecaster (should be simple to implement)
  • PyKAN classifier (should probably be a recurrent network?)
@benHeid benHeid added implementing algorithms Implementing algorithms, estimators, objects native to sktime module:classification classification module: time series classification module:forecasting forecasting module: forecasting, incl probabilistic and hierarchical forecasting enhancement Adding new functionality labels May 5, 2024
@benHeid benHeid added this to ToDo in Workstream: Deep Learning based Forecasters via automation May 5, 2024
@benHeid benHeid mentioned this issue May 5, 2024
Open
6 tasks
@benHeid benHeid moved this from ToDo to In Progress in Workstream: Deep Learning based Forecasters May 5, 2024
@fkiraly
Copy link
Collaborator

fkiraly commented May 6, 2024

we will also need to discuss how to deal with the high runtimes, e.g., in tests

@astrogilda
Copy link
Contributor

@fkiraly is this potentially solvable by using uv?

@fkiraly
Copy link
Collaborator

fkiraly commented May 6, 2024

no, I think the runtime comes from the pykan estimators directly, not the environment setup.

@benHeid
Copy link
Contributor Author

benHeid commented May 8, 2024

we will also need to discuss how to deal with the high runtimes, e.g., in tests

I suppose, we can control the runtime by specifying the parameters in the get_params_test. I.e., selecting small grid sizes and small number of steps.

@fkiraly
Copy link
Collaborator

fkiraly commented May 8, 2024

that might work, if the parameters allow that kind of control

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
enhancement Adding new functionality implementing algorithms Implementing algorithms, estimators, objects native to sktime module:classification classification module: time series classification module:forecasting forecasting module: forecasting, incl probabilistic and hierarchical forecasting
Projects
Workstream: Deep Learning based Forec...
In Progress
Milestone
No milestone
Development

No branches or pull requests
3 participants
@fkiraly @benHeid @astrogilda