Cran updates

NEWS.md

@@ -11,7 +11,7 @@ of our [original](https://doi.org/10.48550/arXiv.2205.07090) `scoringutils` pape
 
 ### `score()`
 - The main function of the package is still the function `score()`. However, we reworked the function and updated and clarified its input requirements. 
-  - The previous columns "true_value" and "prediction" were renamed. `score()` now requires columns called "observed" and "predicted" and "model". The column `quantile` was renamed to `quantile_level` and `sample` was renamed to `sample_id`
+  - The previous columns "true_value" and "prediction" were renamed. `score()` now requires columns called "observed" and "predicted" (some functions still assume the existence of a `model` column as default but this is not a strict requirement). The column `quantile` was renamed to `quantile_level` and `sample` was renamed to `sample_id`
   - `score()` is now a generic. It has S3 methods for the classes `forecast_point`, `forecast_binary`, `forecast_quantile`,  `forecast_sample`, and `forecast_nominal`, which correspond to the different forecast types that can be scored with `scoringutils`. 
   - `score()` now calls `na.omit()` on the data, instead of only removing rows with missing values in the columns `observed` and `predicted`. This is because `NA` values in other columns can also mess up e.g. grouping of forecasts according to the unit of a single forecast.
   - `score()` and many other functions now require a validated `forecast` object. `forecast` objects can be created using the functions `as_forecast_point()`, `as_forecast_binary()`, `as_forecast_quantile()`, and `as_forecast_sample()` (which replace the previous `check_forecast()`). A forecast object is a data.table with class `forecast` and an additional class corresponding to the forecast type (e.g. `forecast_quantile`).

R/class-forecast.R

@@ -75,10 +75,13 @@ assert_forecast.default <- function(
   forecast, forecast_type = NULL, verbose = TRUE, ...
 ) {
   cli_abort(
+    #nolint start: keyword_quote_linter
     c(
       "!" = "The input needs to be a valid forecast object.",
-      "i" = "Please convert to `forecast` object first (see {.fn as_forecast})." # nolint
+      "i" = "Please convert to `forecast` object first by calling the
+      appropriate {.fn as_forecast_<type>} function)."
     )
+    #nolint end
   )
 }
 

R/score.R

@@ -100,10 +100,13 @@ score <- function(forecast, metrics, ...) {
 #' @export
 score.default <- function(forecast, metrics, ...) {
   cli_abort(
+    #nolint start: keyword_quote_linter
     c(
       "!" = "The input needs to be a valid forecast object.",
-      "i" = "Please convert to `forecast` object first (see {.fn as_forecast})." # nolint
+      "i" = "Please convert to `forecast` object first by calling the
+      appropriate {.fn as_forecast_<type>} function)."
     )
+    #nolint end
   )
 }
 

README.Rmd

@@ -87,12 +87,13 @@ remotes::install_github("epiforecasts/scoringutils", dependencies = TRUE)
 - `point`: a forecast for a continuous or discrete outcome variable that is represented by a single number.
 - `quantile`: a probabilistic forecast for a continuous or discrete outcome variable, with the forecast distribution represented by a set of predictive quantiles.
 - `sample`: a probabilistic forecast for a continuous or discrete outcome variable, with the forecast represented by a finite set of samples drawn from the predictive distribution.
+- `nominal` categorical forecast with unordered outcome possibilities (generalisation of binary forecasts to multiple outcomes)
 
 ### Input formats and input validation
 
-The expected input format is generally a `data.frame` (or similar) with required columns `observed`, `predicted`, and `model` that holds the forecasts and observed values. Exact requirements depend on the forecast type. For more information, have a look at the [paper](https://drive.google.com/file/d/1URaMsXmHJ1twpLpMl1sl2HW4lPuUycoj/view?usp=drive_link), call `?as_forecast()`, or have a look at the example data provided in the package (`example_binary`, `example_point`, `example_quantile`, `example_sample_continuous`, `example_sample_discrete`). 
+The expected input format is generally a `data.frame` (or similar) with required columns `observed`, and `predicted` that holds the forecasts and observed values. Exact requirements depend on the forecast type. For more information, have a look at the [paper](https://drive.google.com/file/d/1URaMsXmHJ1twpLpMl1sl2HW4lPuUycoj/view?usp=drive_link), call `?as_forecast_binary`, `?as_forecast_quantile` etc., or have a look at the example data provided in the package (`example_binary`, `example_point`, `example_quantile`, `example_sample_continuous`, `example_sample_discrete`, `example_nominal`). 
 
-Before scoring, input data needs to be validated and transformed into a forecast object using the function `as_forecast()`. 
+Before scoring, input data needs to be validated and transformed into a forecast object using one of the `as_forecast_<type>()` functions. 
 
 ```{r}
 forecast_quantile <- example_quantile |>