dot-check_outcome_type_plausibility: Internal function for checking if the outcome type fits well...

In familiar: End-to-End Automated Machine Learning and Model Evaluation

Internal function for checking if the outcome type fits well to the data

Description

This function may help identify if the outcome type is plausible given the outcome data. In practice it also tests whether the outcome column is actually correct given the outcome type.

Usage

.check_outcome_type_plausibility(
  data,
  outcome_type,
  outcome_column,
  censoring_indicator,
  event_indicator,
  competing_risk_indicator,
  check_stringency = "strict"
)

Arguments

data	Data set as loaded using the .load_data function.
outcome_type	Character string indicating the type of outcome being assessed.
outcome_column	Name of the outcome column in the data set.
censoring_indicator	Name of censoring indicator.
event_indicator	Name of event indicator.
competing_risk_indicator	Name of competing risk indicator.
check_stringency	Specifies stringency of various checks. This is mostly: strict: default value used for summon_familiar. Thoroughly checks input data. Used internally for checking development data. external_warn: value used for extract_data and related methods. Less stringent checks, but will warn for possible issues. Used internally for checking data for evaluation and explanation. external: value used for external methods such as predict. Less stringent checks, particularly for identifier and outcome columns, which may be completely absent. Used internally for predict.

familiar documentation built on Sept. 30, 2024, 9:18 a.m.