NimbusML icon indicating copy to clipboard operation
NimbusML copied to clipboard

Quantile Regression using fast forest regressor

Open hschap opened this issue 3 years ago • 0 comments

The nimbusml documentation says it's a quantile regression forest implementation, but I don't see where to input different quantiles for training or prediction and I don't see a quantiles arg in the prediction fn: https://docs.microsoft.com/en-us/python/api/nimbusml/nimbusml.ensemble.fastforestregressor?view=nimbusml-py-latest

I've used scikit-garden implementation of quantile regression forest before which includes a quantiles= arg in the predict fn: https://github.com/scikit-garden/scikit-garden/blob/master/skgarden/quantile/ensemble.py#L103

Are quantile predictions possible using nimbusml fast forest regressor or any other regression model type?

Thanks for your help!

hschap avatar Nov 16 '20 17:11 hschap