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 support for TimesFM #6408

Open
benHeid opened this issue May 11, 2024 · 3 comments
Open

[ENH] Add support for TimesFM #6408

benHeid opened this issue May 11, 2024 · 3 comments
Labels
enhancement Adding new functionality interfacing algorithms Interfacing existing algorithms/estimators from third party packages module:forecasting forecasting module: forecasting, incl probabilistic and hierarchical forecasting
Projects
Workstream: Deep Learning based Forec...

Comments

@benHeid
Copy link
Contributor

benHeid commented May 11, 2024

Support for TimesFM, the time series foundation model from google research:
@benHeid benHeid added module:forecasting forecasting module: forecasting, incl probabilistic and hierarchical forecasting enhancement Adding new functionality labels May 11, 2024
@fkiraly fkiraly mentioned this issue May 11, 2024
Open
@fkiraly
Copy link
Collaborator

fkiraly commented May 11, 2024

added it to the list here: #6177

One of these days, we should perhaps create our own?

@fkiraly fkiraly added the interfacing algorithms Interfacing existing algorithms/estimators from third party packages label May 11, 2024
@geetu040
Copy link
Contributor

I have a question: What is the right way to implement this interface?

  1. add the package timesfm as a dependency to sktime and create the interface on top of that?
  2. repeat the model implementation using the same libraries and code that is in the actual source? - that will put jax and related libraries as new dependencies in sktime
  3. convert the model code from jax to pytorch and build an interface on top of the pytorch adapter in sktime?

@benHeid
Copy link
Contributor Author

benHeid commented May 24, 2024 •
edited

I would prefer option 1 if possible. If this is not possible we can discuss in more detailed how to proceed

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 interfacing algorithms Interfacing existing algorithms/estimators from third party packages module:forecasting forecasting module: forecasting, incl probabilistic and hierarchical forecasting
Projects
Workstream: Deep Learning based Forec...
ToDo
Milestone
No milestone
Development

No branches or pull requests
3 participants
@fkiraly @benHeid @geetu040