add contant baseline model #88
Conversation
| @@ -0,0 +1,117 @@ | |||
| if (baseline == "constant baseline") { | |||
| # get last observed value for median | |||
|
|
|||
There was a problem hiding this comment.
file name should no include name of model or R. its in the constant-baseline folder so this is repetition.
There was a problem hiding this comment.
please update name to be update.R or similar having constant-baseline in both folder and filename makes me sad.
There was a problem hiding this comment.
I don't want to make you sad! :(
It is just that all other files are called update-timeseries.R and update-rt.R etc...
|
|
||
| formatted_forecasts <- dplyr::bind_rows(formatted, | ||
| point_forecast) %>% | ||
| dplyr::relocate(forecast_date, submission_date, target, target_end_date, location, type, quantile, value) |
There was a problem hiding this comment.
need to hook into weekly scheduled run time (i.e files in models)
There was a problem hiding this comment.
done - I think. Needs checking whether there are any other places this needs to go
There was a problem hiding this comment.
needs to be in the aggregated R script in order for the results to be loaded into the submissions folder.
|
Thanks @nikosbosse looks good and few stylistic and usage concerns. |
|
@nikosbosse, what's happening with this one - is the PR ready for another round of review, or are we rethinking adding this at all (@seabbs)? no rush, just checking in. |
|
So the constant baseline model 'works' - the problem is more with the QRA (that fails with too many models and sometimes runs indefinitely). So I think ideally we need to do the following: look through what is happening in detail, set a maximum to the quantgen runtime, look at the horizon grid we want to have qra for, implement a fallback to equal weights, have a flag somewhere that something goes wrong |
|
Perhaps drop the NULL model from the QRA and we can instead think about using the NULL model as a stand in if our algorithmic approaches are producing nonsense. @nikosbosse can you make the QRA issues into an issue that someone can then pick up if not able to do in this PR. |
|
created an issue for QRA: #94 |
|
I think this can be merged now |
as mentioned in #87