I'm working on improving the usefulness of the report. Some of this is limited by issues I've opened in pointblank, but might have to eventually work around.
So far I've:
Added a short description of how to interpret pointblank tables
Given the validation tables informative titles
Hidden explanatory code
Removed variables from forecast-based validations that did not meet assumptions of timeseries models (precip, wind direction, solar radiation)
Set GitHub actions to run the targets pipeline every day at noon
Experimented with date-range sliders to adjust range in a shiny app version of the validation report (shiny-report.qmd)
I wasn't able to:
customize the pointblank reports (requires either changes to pointblank or building the tables from the tibble output of the validation. Either is outside of the scope of the incubator, or at least for another PR).
I'm working on improving the usefulness of the report. Some of this is limited by issues I've opened in
pointblank, but might have to eventually work around.So far I've:
I wasn't able to:
pointblankor building the tables from the tibble output of the validation. Either is outside of the scope of the incubator, or at least for another PR).