GlobalEvaluation class — GlobalEvaluation • SticsREval

The GlobalEvaluation class is responsible for evaluating the global performance of the STICS model across multiple USMs (User Simulation Models). It computes global statistics and performs a comparison of the relative Root Mean Square Error (rRMSE) for the specified variables.

Details

The evaluation process involves the following steps:

Generating global statistics by comparing simulated and observed data.
Performing a comparison of rRMSE values for the specified variables.
Logging the results and providing a summary of the evaluation.

Active bindings

success: A logical value indicating whether the global evaluation was successful. The evaluation is considered successful if the rRMSE comparison was performed and there are no critical variables identified in the comparison.

Methods

`GlobalEvaluation$new()`

Create a new GlobalEvaluation object.

Usage

GlobalEvaluation$new(config, workspace = NULL, logger = default_logger)

Arguments

config: A Configuration object containing the necessary parameters for the evaluation.
workspace: An optional EvalWorkspace object. If not provided, a new EvalWorkspace will be created using the provided configuration.
logger: An optional logger object for logging messages. If not provided, the default logger will be used.

`GlobalEvaluation$run()`

Run the global evaluation. This method performs the global evaluation by generating global statistics and comparing the rRMSE values for the specified variables.

Usage

GlobalEvaluation$run()

`GlobalEvaluation$summary()`

Print a summary of the global evaluation results.

Usage

GlobalEvaluation$summary()

`GlobalEvaluation$export()`

Export the global evaluation results to CSV files. This method exports the global statistics and rRMSE comparison results to CSV files in the specified output directory.

Usage

GlobalEvaluation$export()

`GlobalEvaluation$clone()`

The objects of this class are cloneable with this method.

Usage

GlobalEvaluation$clone(deep = FALSE)

Arguments

deep: Whether to make a deep clone.

Examples

if (FALSE) { # \dontrun{
config <- Configuration$new(
  eval_workspace = "path/to/eval_workspace",
  usms_workspace = "path/to/usms_workspace",
  usms = c("USM1", "USM2"),
  var2exclude = c("var1", "var2"),
  percentage = 0.1,
  output_dir = "path/to/output_dir"
)
global_eval <- GlobalEvaluation$new(config = config)
global_eval$run()
global_eval$summary()
global_eval$export()
} # }