&rigvedmanoj, benheid, geetu040 [ENH] interface to Chronos (zero-shot) forecaster #7001
Conversation
|
also @fkiraly do we still need this? I don't think so. def __init__():
self.model_pipeline = None
def _fit(self, y, X=None, fh=None):
if self.model_pipeline is not None:
return
... |
No, probably not, then. Another point, if |
yes it is present in the CI runs, as it is a requirement for |
|
The failures are due to the segmentation fault, not relevant to this PR |
There was a problem hiding this comment.
Thanks for completing - I have added @geetu040 and @RigvedManoj (for the initial PR #6205) as author and contributor for this PR, and @Z-Fran and @RigvedManoj to the contributors register.
…) forecaster (#7001) Adds an interface to the Chronos zero-shot forecaster, including a vendored version of the library. Usage example: ```python from sktime.datasets import load_airline from sktime.forecasting.chronos import ChronosForecaster from sktime.split import temporal_train_test_split from sktime.forecasting.base import ForecastingHorizon y = load_airline() y_train, y_test = temporal_train_test_split(y) fh = ForecastingHorizon(y_test.index, is_relative=False) forecaster = ChronosForecaster("amazon/chronos-t5-tiny") forecaster.fit(y_train) y_pred = forecaster.predict(fh) ```
Adds an interface to the Chronos zero-shot forecaster, including a vendored version of the library.
Usage example: