Steven Paul Sanderson II, MPH comments

Results 53 comments of


                                            Steven Paul Sanderson II, MPH

extract_nested_future_forecast() fails when adding multiple predictors in recipe() with +

I too have this issue

Add recipe name to model name in modeltime table

I will do my best to try to get something of this working and paste my code as a comment even if it is just a skeleton

Add recipe name to model name in modeltime table

@AlbertoAlmuinha that would be awesome! Then you know which recipe_model combo is the winner without having to lookup the model id and trying to figure out which recipe went with...

Add recipe name to model name in modeltime table

This should also work with panel data correct? Meaning it would play nice with panel date as well? Sent from my iPhone www.spsanderson.com > On Aug 9, 2021, at 2:57...

modeltime_nested_refit overwriting issue

For me I keep getting an error of the parallel backend failing Started like two weeks ago

modeltime_nested_refit overwriting issue

I wouldn't worry for now, could be local to my machine. I have a script I run here: https://github.com/spsanderson/healthyverse_tsa/blob/master/README.Rmd

Modeltime in Production

I think this would be an amazing idea, I also think maybe being able to export the model to sql would be good Sent from my iPhone www.spsanderson.com > On...

Modeltime in Production

@AlbertoAlmuinha thanks for the link

`modeltime_calibrate()` All models fail, quite = FALSE provides no direction

using `modeltime::modeltime_calibrate(new_data = rsample::testing(splits), quiet = FALSE)` provides no extra information.

`modeltime_calibrate()` All models fail, quite = FALSE provides no direction

It seems that if I drop `recipes::step_corr()` then the model completes. Even though this fixes it, I'll keep the issue open until I can fully understand the why?