Sebastian Funk

Distinguish NA (missing) from NA (accumulated)

13

> Enabling this would, I think, require some sort of that marks dates explicitly as missing vs. NA. I think this would be my preferred option as it would be...

Efficiency of accumulating NAs

2

I guess I am saying that cases_time would become optionally an array with a sum. Another option might be to make this a matrix multiplication problem _Originally posted by @seabbs...

Move to `{cli}` for messaging

1

One reason to do so: https://github.com/epiforecasts/EpiNow2/pull/534#issuecomment-1943972293

Change all priors to new dist interface, not just delays

3

For a future PR this should probably be updated to use `normal()`. _Originally posted by @seabbs in https://github.com/epiforecasts/EpiNow2/pull/504#discussion_r1408162989_ This will require moving the conversion to discretised distribution to the processing...

repo size

9

The repo has grown fairly large (~1 GB), but the files currently in the repo are only 11 MB in size. It might be nice, particularly towards those on low...

streamline S3 classes

10

Currently the three models included in the package return S3 class with similar but slightly different contents. Whilst to some degree that's expected because of their different use cases it...

[Recent internal changes](https://github.com/epiforecasts/EpiNow2/pull/363) in the represenation of delay distributions have caused a decline in performance. A systematic investigation of possible performance improvements would be useful.

Model validation

6

At the moment in the tests we only validate the model itself in a few specificy ways (e.g. `update_infectiousness`, `generate_infections`). There is also the synthetic validation but it requires a...

Testing fails locally when done repeatedly

2

Result of repeated `devtools::test()` ```r `assert_input_interval(observed, lower, upper, 0.5)` did not throw the expected warning. ``` because the warning is only shown once by design.

At the moment column names are specified when calling `check.survey` as well as `contact.matrix`. It might be better to specify column names when constructing the survey object, i.e. in `survey`.