Generate a new synthetic RDD whose rows are iid sampled from input feature vectors
Generate a new synthetic RDD whose rows are iid sampled from input feature vectors
The number of iid samples to generate.
The input sample size. Input is periodically sampled and the sample is used to generate iid output data. Defaults to 10000.
The output sample size. Each input sample is used to generate this number of output samples. Defaults to 10000.
An RDD of FeatureSeq where each 'column' in the feature sequence is statistically independent of the others, but shares the marginal distribution of the corresponding input column.
Implementation-specific subclass of IIDFeatureSamplingMethods for RDDs