Validate Metric Output

Validates that metric output conforms to the expected schema. Metric output must be a named list where all names are non-empty strings and values are either scalar atomic types or named numeric vectors.

Usage

validate_metric_output(output, metric_name)

Arguments

output

The output list to validate.

metric_name

Character string identifying the metric (for error messages).

Value

Invisible output if validation passes. Otherwise, an error is raised.

Validation Rules

• output must be a list

• all elements must have non-empty names

• values must be one of:

  • scalar logical, integer, double, or character

  • named numeric vector

• no NULL values allowed

• no nested lists, data frames, or matrices allowed

Examples

# Valid output
validate_metric_output(list(rmse = 0.5, n_obs = 100L), "rmse")

# Valid output with named vector
validate_metric_output(
  list(params = c(alpha = 0.1, beta = 0.2)),
  "param_estimates"
)

if (FALSE) { # \dontrun{
# Invalid: unnamed element
validate_metric_output(list(0.5), "rmse")  # Error

# Invalid: nested list
validate_metric_output(list(nested = list(a = 1)), "rmse")  # Error
} # }