Allow function as season_length argument in SeasonalNaivePredictor
#3033
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
jaheba
reviewed
Oct 30, 2023
| if isinstance(self.season_length, int): | ||
| season_length = self.season_length | ||
| else: | ||
| season_length = self.season_length(item["start"].freq) |
There was a problem hiding this comment.
This requires that there is a freq for the time series, which might not be the case in the future?
There was a problem hiding this comment.
I guess this mechanism is really only useful if you have frequency information attached to the series. If that's not the case, there's no point in using this option, and the model should be configured with an int value.
Also: I believe the assumption that item["start"] is a pd.Period (hence has frequency information) is already baked into this method since it uses forecast_start?
There was a problem hiding this comment.
At the moment we require that all time series have frequency information attached -- but that's an arbitrary requirement.
jaheba
approved these changes
Oct 30, 2023
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description of changes: The idea is to allow constructing the predictor with a "season length map" instead of a fixed integer, so that the model can be applied to datasets with different frequencies and expected seasonal patterns. For example, the predictor can be constructed by passing
season_length=gluonts.time_feature.seasonality.get_seasonality.By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.
Please tag this pr with at least one of these labels to make our release process faster: BREAKING, new feature, bug fix, other change, dev setup