You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
AutoGluon is an AutoML framework that also offers support for time series forecasting. Most if not all traditional models are supported (Arima, DeepAR) but also new models such as PatchTST and Chronos.
The useful part about that is that it makes fine tuning and comparison versy easy.
Would it b possible to enable AutoGluon support for your model? Or point out how a wrapper class for Lag LLama could be written?
The text was updated successfully, but these errors were encountered:
We do not have plans to integrate our model with AutoGluon at the moment. I am not sure how a wrapper class can be written too, but I think it shouldn't be very difficult given we use the GluonTS backend.
Maybe @kashif has ideas on the best ways to integrate Lag-Llama into AutoGluon.
AutoGluon is an AutoML framework that also offers support for time series forecasting. Most if not all traditional models are supported (Arima, DeepAR) but also new models such as PatchTST and Chronos.
The useful part about that is that it makes fine tuning and comparison versy easy.
Would it b possible to enable AutoGluon support for your model? Or point out how a wrapper class for Lag LLama could be written?
The text was updated successfully, but these errors were encountered: