MDL
Transform a column of continuous labelled features to n columns of binned categorical features.
The optimum number of bins is computed using Minimum Description Length (MDL), which is an
entropy measurement between the values and the targets.
The optimum number of bins is computed using Minimum Description Length (MDL), which is an
entropy measurement between the values and the targets.
The transformer expects an MDLRecord where the first field is a label and the second value
is the scalar that will be transformed into buckets.
is the scalar that will be transformed into buckets.
MDL is an iterative algorithm so all of the data needed to compute the buckets will be pulled
into memory. If you run into memory issues the
into memory. If you run into memory issues the
sampleRate parameter should be lowered.References:
-
Fayyad, U., & Irani, K. (1993). "Multi-interval discretization of continuous-valued attributes
for classification learning."
Value members
Methods
def apply[T](name: String, sampleRate: Double, stoppingCriterion: Double, minBinPercentage: Double, maxBins: Int, seed: Int)(evidence$2: ClassTag[T]): Transformer[MDLRecord[T], B[T], C]
Create an MDL Instance.
- Value Params
- maxBins
-
maximum number of thresholds per feature
- minBinPercentage
-
minimum percent of total data allowed in a single bin
- sampleRate
-
percentage of records to keep to compute the buckets
- seed
-
seed for the sampler
- stoppingCriterion
-
stopping criterion for MDL