20.23 Complexity Parameter Table
We can look at the raw data to have a more precise and detailed view of the data. Here we only list specific rows from the complexity parameter table.
## CP nsplit rel error xerror xstd
## 1 1.690000e-01 0 1.0000000 1.0000000 0.004003739
## 2 2.704940e-02 1 0.8310000 0.8360073 0.003742385
## 3 3.801427e-03 3 0.7769012 0.7782501 0.003638129
## 4 3.535735e-03 5 0.7692984 0.7612050 0.003606007
## 5 3.453984e-03 10 0.7458971 0.7601627 0.003604022
## 22 4.189744e-04 61 0.7008931 0.7153222 0.003516271
## 23 4.019430e-04 63 0.7000552 0.7148113 0.003515244
## 24 3.883178e-04 66 0.6988494 0.7148930 0.003515408
## 25 3.780989e-04 70 0.6972961 0.7145864 0.003514792
## 26 3.678800e-04 72 0.6965399 0.7144025 0.003514422
## 27 3.474422e-04 73 0.6961720 0.7141981 0.003514011
## 28 3.372233e-04 77 0.6947822 0.7144229 0.003514463
## 29 3.270044e-04 80 0.6934742 0.7139529 0.003513517
## 80 6.715270e-05 967 0.5876678 0.7176521 0.003520946
## 81 6.688727e-05 985 0.5863394 0.7176521 0.003520946
## 82 6.540089e-05 1009 0.5845204 0.7176521 0.003520946
## 83 6.471963e-05 1019 0.5838664 0.7176521 0.003520946%$
See how the relative error continues to decrease as the tree becomes more complex, but the cross validated error decreases and then starts to increase! We might choose a sensible value of from this table.
Your donation will support ongoing availability and give you access to the PDF version of this book. Desktop Survival Guides include Data Science, GNU/Linux, and MLHub. Books available on Amazon include Data Mining with Rattle and Essentials of Data Science. Popular open source software includes rattle, wajig, and mlhub. Hosted by Togaware, a pioneer of free and open source software since 1984. Copyright © 1995-2022 Graham.Williams@togaware.com Creative Commons Attribution-ShareAlike 4.0
