-
-
Notifications
You must be signed in to change notification settings - Fork 1.3k
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
[ENH] Support Global Time Series Forecasting #4651
Comments
Addendum: this might be the same request as by @romanlutz some time earlier: |
Yes, I think the design above still makes sense.
The one thing that bothers me is that ordinary panel forecasting can be considered a special case, in that case we fit to what, in the code snippet, is Should this be bothering us? Ultimately, it would result in an interface wrapper class for the naive case where we ignore |
should we maybe circle back to finalizing this design, @benHeid? Perhaps in the Fri meeting? |
Yes that is a good idea. I think we can split the pipeline meeting and talk about that topic in the first or second half of the meeting |
Questions for discussion:
|
Regarding the first point, there is a problem that I cannot seem to satisfy both conditions at the same time:
options to do in the general case, from discussion with @benHeid:
|
from meeting on August 4, decisions (please comment, @benHeid):
yes, as described above.
same class with extended signature. As @benHeid also suggested, would use tags to differentiate, not new class
already supported by current interface, no action needed.
|
hackmd notebook for Aug 4 meeting here for reference. To view, click the dropdown arrow. Design notes for global forecasting APIFK questionsQuestions for discussion:
FK - modelling global math/interfaceDefinition of "global" is often fuzzy, and how it maps onto fit/predict. perspective: without fit/predictIn a single input-output perspective, we have: A. "panel" forecasting: given n instances of time series produce forecasts of B. "global" forecasting with forecasted instances different from training instances given n instances of time series produce forecasts of notes from discussion with BH
mapping on fit/predictA -> in sktime, we've mapped this as follows:
B -> we could map this two different ways! B1 - as currently in issue #4651
B2 - pass all data to
What is the difference between B1 and B2?B1 gets additional instances in In B2, "which series" is Difference is similar to passing notes from discussion with BH
BH's undersanding of global forecasting:You have a training dataset containing n time series During inference you have a set of m time series, Technically noteThe forecasts for each of this Probable Definition in Literature:tentative decisions?
FK: yes - after discusison, favours option B1 above B2
FK: thinks, same class with extended signature. As BH also suggested, would use tags to differentiate, not new class
FK - can you explain? BH - different vs same exogenous data per instance FK: currently, exogenous data nees to have same instance indexing. We cannot pass exogenous data that has no instance index if the y does (but could be extended in the future) -> current state is fine
|
FYI @olerch |
Actions:
|
@Xinyu-Wu-0000, can you kindly comment so I can assign you this issue? |
Yeah of course. |
Is your feature request related to a problem? Please describe.
In deep learning-based time series forecasting, a model is often trained on multiple time series (e.g. Hourly Subset of the M4 Competition or UCI ElectricityLoadDiagramDataset) and applied to time series. Thereby, only one model for all time series exists. Furthermore, the set of time series for which the prediction is made does not have to be equal to the time series on which the model is trained.
Describe the solution you'd like
Based on this consideration, I would like to make forecasts on time series on which a model is not fitted. Based on a proposal of @fkiraly and @ahmedgc in the pysf Repo a solution may look at follows:
This would require adding an optional argument y to the
predict
method where y would comprise all required historical values of the time series that should be predicted.Note depending on the use case, different time series also get different values of exogenous variables. E.g., consider a set of time series describing the electrical load of different buildings at different locations. Thereby, for each location exist a different temperature time series. Then a solution may require a mapping from the target time series exogenous time series. Theoretically, this could be solved by adding a further dimension to the exogenous data, whereby the first dimension has to equal the number of considered time series and then the mapping could be performed via the index.
Required Decisions
predict
?The text was updated successfully, but these errors were encountered: