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
I'm trying to use Ray to do a Hyperparameter search via the sklearn API. In there, one of the callback function RayTrainReportCallback requires the key of the XGBoost metric. Since I'm using XGBoostRegressor, I would think the key should be validation_0-rmse. But this, and all other variations of that throw a key error.
Can someone tell me what the correct key should be, or at least how to reveal the dictionary of metrics?
The text was updated successfully, but these errors were encountered:
I don't believe xgboost_ray is still supported. I have been having trouble using that package (which you'll see the latest issue being created by me too). I've since moved to use xgboost (with the sklearn API) with ray/ray-tune. Since that requires access to a key for a metric in xgboost, it should belong to the xgboost repo? I can open an issue in ray too, but I don't know if this belongs there more.
I'm trying to use Ray to do a Hyperparameter search via the sklearn API. In there, one of the callback function
RayTrainReportCallback
requires the key of the XGBoost metric. Since I'm usingXGBoostRegressor
, I would think the key should bevalidation_0-rmse
. But this, and all other variations of that throw a key error.Can someone tell me what the correct key should be, or at least how to reveal the dictionary of metrics?
The text was updated successfully, but these errors were encountered: