search

sol-eng / bike_predict

A demo of an end-to-end machine learning pipeline, using Posit Connect
92 stars 31 forks source link

Suggestions for tidymodels + vetiver code #31

Open juliasilge opened 2 years ago

juliasilge commented 2 years ago

Hello! 👋 This demo is looking so great; thank you for creating this to show folks how to use our tools. I have a couple of minor suggestions for tidymodels and vetiver code in 01-train-and-deploy-model:

Can you use initial_time_split() here, instead of the manual splitting? Then you can use training() and testing() from tidymodels:

https://github.com/sol-eng/bike_predict/blob/2a18bbb608ba06d52d4a1020ac3ad3e1f9f7ea94/content/02-model/01-train-and-deploy-model/document.qmd#L55-L58

I don't believe you need to manually save the ptype here (or save versioned = TRUE). This should be grabbed automatically from the model_fit:

https://github.com/sol-eng/bike_predict/blob/2a18bbb608ba06d52d4a1020ac3ad3e1f9f7ea94/content/02-model/01-train-and-deploy-model/document.qmd#L172-L175

If you'd like a PR for either of these, I would be happy to do it!

juliasilge commented 2 years ago

I edited because I realized that we are predicting on the endpoint here and it doesn't have an augment() method yet, as tracked in rstudio/vetiver-r#10. Apologies!

SamEdwardes commented 2 years ago

Thank you Julia - this is great, I did not know about these functions. I will update the code to include them :)