Perform a benchmark between algorithm adaptation and multilabel and binary relevance wrappers

Perform a benchmark between algorithm adaptation and multilabel and binary relevance wrappers

multilabelBenchmark(
  trainingData,
  validationHumanReadingDTM,
  model_type,
  scale_type = "location",
  aggregated_labels = FALSE,
  obs_threshold = 10
)

Arguments

trainingData	training data
validationHumanReadingDTM	training data from human reading, used to extract target column names
model_type	type of predictions
scale_type	One of "location", "spatial", "temporal", default to "location"
aggregated_labels	logical
obs_threshold	remove columns with less than this threshold

Value

benchmark object