search

epiforecasts / EpiNow2

Estimate Realtime Case Counts and Time-varying Epidemiological Parameters
https://epiforecasts.io/EpiNow2/dev/
Other
116 stars 33 forks source link

Document/surface what settings are applied when `rt = NULL`. #761

Closed jamesmbaazam closed 2 months ago

jamesmbaazam commented 3 months ago

I've been thinking we should actually surface what settings are applied when rt = NULL. We only say that they can set it to NULL but we could add what settings are used when that is done. What do you think?

_Originally posted by @jamesmbaazam in https://github.com/epiforecasts/EpiNow2/pull/747#discussion_r1734718474_

seabbs commented 3 months ago

I think we might also want to move all the random walk + differencing etc settings. elsewhere (to i.e a latent argument). Then all models can use those features and we don't use anything by stealth. This would also mean getting rid of the gp arg (and passing it to the latent_opts as the model)

jamesmbaazam commented 2 months ago

I think we might also want to move all the random walk + differencing etc settings. elsewhere (to i.e a latent argument). Then all models can use those features and we don't use anything by stealth. This would also mean getting rid of the gp arg (and passing it to the latent_opts as the model)

I'm all for exposing all settings that have been set under the hood. Looks like it deserves its own issue thread.