Packages

object ConfusionFit

The ConfusionFit companion object records the indicies and labels for the base Quality of Fit (QoF) measures for the classification techniques.

Linear Supertypes
AnyRef, Any
Ordering
  1. Alphabetic
  2. By Inheritance
Inherited
  1. ConfusionFit
  2. AnyRef
  3. Any
  1. Hide All
  2. Show All
Visibility
  1. Public
  2. All

Value Members

  1. val MIN_FOLDS: Int
  2. def fitLabel: Seq[String]

    Return the labels for the Scalar Quality of Fit (QoF) measures.

  3. def fitLabel_v: Seq[String]

    Return the labels for the Vector Quality of Fit (QoF) micro-measures.

  4. def help: String

    Return the help string that describes the Quality of Fit (QoF) measures provided by the ConfusionFit class.

    Return the help string that describes the Quality of Fit (QoF) measures provided by the ConfusionFit class. The QoF measures are divided into four groups: general, ordinary, micro (per class) vectors and means of the micro vectors. Ordinary are values of the last element in the micro vectors and can be interpreted as, say the precision for the last class value/label, e.g., y = hasCancer in {no, yes}, is the prcision of the 'yes' prediction and is most meaningful when the number of class values/labels (k) is 2.

    See also

    en.wikipedia.org/wiki/Precision_and_recall

    en.wikipedia.org/wiki/Cohen%27s_kappa

  5. val index_acc: Int
  6. val index_f1: Int
  7. val index_f1_m: Int
  8. val index_f1_v: Int
  9. val index_kappa: Int
  10. val index_p: Int
  11. val index_p_m: Int
  12. val index_p_rSq: Int
  13. val index_p_v: Int
  14. val index_r: Int
  15. val index_r_m: Int
  16. val index_r_v: Int
  17. val index_s: Int
  18. val index_s_m: Int
  19. val index_s_v: Int
  20. val index_sse: Int
  21. val index_sst: Int
  22. def qofStatTable: Array[Statistic]

    Create a table to store statistics for QoF measures, where each row corresponds to the statistics on a particular QoF measure, e.g., 'acc'

  23. def test(y: VectoI, yp: VectoI, k: Int = 2): Unit

    Test and report the confusion matrix and associate QoF measures.

    Test and report the confusion matrix and associate QoF measures.

    y

    the actual class values

    yp

    the predicted class values

    k

    the number of class labels {0, 1, ... , k-1}