Ranjana Girish
2016-Oct-08 11:23 UTC
[R] In SOM package all entities are predicted to the same class
From: "Ranjana Girish" <ranjanagirish30 at gmail.com> Date: Oct 7, 2016 3:39 PM Subject: Re:In SOM package all entities are predicted to the same class Cc: <r-help at r-project.org>> Even after trying with different parameters of SOM still all entities aregetting predicted to same class..> > Note: for each run, class are different because nrow considers train seteach time randomly.> > 1)som.prediction$unit.classif > som.wines <- som(ScaledNonNAtraining, grid = somgrid(5, 5, "hexagonal")) > [1] 4 4 4 4 23 4 4 4 4 4 4 4 4 4 4 4 21 4 4 4 4 4 44 4 4 4 4 4 4> [31] 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 44 4 4 4 4 4 4> [61] 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 44 4 4 4 4 4 4> [91] 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 44 4 4 4 4 4 4> [121] 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 44 4 4 4 4 4 4 4> [151] 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 44 4 4 4 4 4 4 4> [181] 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 44 4 4 4 4 4 4 4> [211] 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 44 4 4 4 4 4 4 4> [241] 4 4 4 4 4 4 4 4 4 > Accuracy 2.811245 > > 2)som.wines <- som(ScaledNonNAtraining, grid = somgrid(5, 5,"rectangular")) som.prediction$unit.classif [1] 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 [35] 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 2 15 15 15 15 15 15 15 15 15 15 15 15 15 15 [69] 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 [103] 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 [137] 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 1 15 [171] 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 2 25 25 15 15 15 15 15 15 15 15 15 [205] 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 2 15 15 15 15 15 15 15 15 2 15 15 15 15 15 15 15 15 15 [239] 15 15 15 15 15 15 15 15 15 15 15 accuracy [1] 1.204819> > 3)som.wines <- som(ScaledNonNAtraining, grid = somgrid(5, 5,"rectangular"), rlen = 5000) som.prediction$unit.classif [1] 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 [35] 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 [69] 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 [103] 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 [137] 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 [171] 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 4 25 25 25 25 25 25 25 25 25 25 25 [205] 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 4 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 [239] 25 25 25 25 25 25 25 25 25 25 25 accuracy [1] 0> > 4)om.wines <- som(ScaledNonNAtraining, grid = somgrid(5, 5,"rectangular"), alpha = 0.15, rlen = 5000) som.prediction$unit.classif [1] 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 [52] 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 [103] 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 [154] 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 [205] 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1> >[[alternative HTML version deleted]]