iid
Interface for enriched iid feature sampling methods on sequence-like collections of feature vectors.
Interface for enriched iid feature sampling methods on sequence-like collections of feature vectors. A feature vector is some sequential collection of Double values, whose values may be iid sampled to generate new synthetic feature vectors having the same marginal distributions as the input features. The underlying feature vector representation is not assumed by this interface; multiple representations might be supported
import io.radanalytics.silex.sample.iid.implicits._ rdd.iidFeatureSeqRDD(100000)
A package that provides 'iid' sampling functionality on various collections whose elements constitute some kind of feature vector representation, eg Seq, FeatureSeq, etc. The acronym 'iid' stands for 'independent, identically distributed', and refers to the property that sampled columns share the identical marginal distribution with corresponding input colums, but are statistically independent of any other columns, which is not necessarily true of the input.