[X] Make sure this PR is against "dev", not "main".
[x] Request a review from one of the current epipredict main reviewers:
dajmcdon.
[x] Make sure to bump the version number in DESCRIPTION and NEWS.md.
Always increment the patch version number (the third number), unless you are
making a release PR from dev to main, in which case increment the minor
version number (the second number).
[x] Describe changes made in NEWS.md, making sure breaking changes
(backwards-incompatible changes to the documented interface) are noted.
Collect the changes under the next release number (e.g. if you are on
0.7.2, then write your changes under the 0.8 heading).
[x] Consider pinning the epiprocess version in the DESCRIPTION file if
You anticipate breaking changes in epiprocess soon
You want to co-develop features in epipredict and epiprocess
Change explanations for reviewer
In the process of using both grf_quantile for trees and actually using step_adjust_latency in the context of the flusion dataset, I ran into a number of issues, which this addresses (actually separated by commit, so they're fairly atomic).
For step_adjust_latency:
epi_keys_checked had a bug when there was more than one key
the epi_df was incorrectly being reconstructed during the bake step because of epiprocess0.9
I had some key values (a different source) which stopped in 2020, but was still useful to include in the dataset. To ignore this for latency adjustment purposes, I added the keys_to_ignore parameter, which is a named list, where the name is a key column name from epi_keys_checked, and the values are the key values to ignore. In my use case, this is list(source = c("flusurv"))
For grf_quantile, I just ran into quantiles that were out of order, filed as https://github.com/cmu-delphi/epipredict/issues/407. For now I'm just dropping the order requirement, and sorting the values. We may want to investigate why it's producing out of order coefficients, or add a check to see if it's actully meaningfully out of order
Checklist
Please:
DESCRIPTIONandNEWS.md. Always increment the patch version number (the third number), unless you are making a release PR from dev to main, in which case increment the minor version number (the second number).epiprocessversion in theDESCRIPTIONfile ifepiprocesssoonepipredictandepiprocessChange explanations for reviewer
In the process of using both
grf_quantilefor trees and actually usingstep_adjust_latencyin the context of the flusion dataset, I ran into a number of issues, which this addresses (actually separated by commit, so they're fairly atomic). Forstep_adjust_latency:epi_keys_checkedhad a bug when there was more than one keyepi_dfwas incorrectly being reconstructed during the bake step because of epiprocess0.9keys_to_ignoreparameter, which is a named list, where the name is a key column name fromepi_keys_checked, and the values are the key values to ignore. In my use case, this islist(source = c("flusurv"))For
grf_quantile, I just ran into quantiles that were out of order, filed as https://github.com/cmu-delphi/epipredict/issues/407. For now I'm just dropping the order requirement, and sorting the values. We may want to investigate why it's producing out of order coefficients, or add a check to see if it's actully meaningfully out of order