Like earth can be trained simultaneously for degree and nprune, is there a way
to train ctree simultaneously for mincriterion and maxdepth?
Also, I notice there are separate methods ctree and ctree2, and if both options
are attempted to tune with one method, the summary averages the option it
doesn't support. The full log is attached, and notice these lines below for
method="ctree" where maxdepth=c(2,4) are averaged to maxdepth=3.
Fitting: maxdepth=2, mincriterion=0.95
Fitting: maxdepth=4, mincriterion=0.95
Fitting: maxdepth=2, mincriterion=0.99
Fitting: maxdepth=4, mincriterion=0.99
mincriterion Accuracy Kappa maxdepth Accuracy SD Kappa SD maxdepth SD
0.95 0.939 0.867 3 0.0156 0.0337 1.01
0.99 0.94 0.868 3 0.0157 0.0337 1.01
I use R 2.12.1 and caret 4.78.
Andrew
> require(party)
> require(mlbench)
> data(BreastCancer)
> BreastCancer$Id <- NULL
> grid <- expand.grid(.maxdepth=c(2:6), .mincriterion=c(0.95, 0.97, 0.99))
> print(grid)
.maxdepth .mincriterion
1 2 0.95
2 3 0.95
3 4 0.95
4 5 0.95
5 6 0.95
6 2 0.97
7 3 0.97
8 4 0.97
9 5 0.97
10 6 0.97
11 2 0.99
12 3 0.99
13 4 0.99
14 5 0.99
15 6 0.99
> ct.best <- train(Class ~ . , data=BreastCancer, method="ctree", tuneGrid=grid)
Loading required package: class
Attaching package: 'class'
The following object(s) are masked from 'package:reshape':
condense
Attaching package: 'e1071'
The following object(s) are masked from 'package:gtools':
permutations
Fitting: maxdepth=2, mincriterion=0.95
Fitting: maxdepth=3, mincriterion=0.95
Fitting: maxdepth=4, mincriterion=0.95
Fitting: maxdepth=5, mincriterion=0.95
Fitting: maxdepth=6, mincriterion=0.95
> grid <- expand.grid(.maxdepth=c(2, 4), .mincriterion=c(0.95, 0.99))
> print(grid)
.maxdepth .mincriterion
1 2 0.95
2 4 0.95
3 2 0.99
4 4 0.99
> ct.best <- train(Class ~ . , data=BreastCancer, method="ctree", tuneGrid=grid)
Fitting: maxdepth=2, mincriterion=0.95
Fitting: maxdepth=4, mincriterion=0.95
Fitting: maxdepth=2, mincriterion=0.99
Fitting: maxdepth=4, mincriterion=0.99
Aggregating results
Selecting tuning parameters
Fitting model on full training set
> print(ct.best)
683 samples
80 predictors
Pre-processing: None
Resampling: Bootstrap (25 reps)
Summary of sample sizes: 683, 683, 683, 683, 683, 683, ...
Resampling results across tuning parameters:
mincriterion Accuracy Kappa maxdepth Accuracy SD Kappa SD maxdepth SD
0.95 0.939 0.867 3 0.0156 0.0337 1.01
0.99 0.94 0.868 3 0.0157 0.0337 1.01
Accuracy was used to select the optimal model using the largest value.
The final value used for the model was mincriterion = 0.99.
>
>
>
>
> grid <- expand.grid(.maxdepth=c(2, 4), .mincriterion=c(0.95, 0.99))
> print(grid)
.maxdepth .mincriterion
1 2 0.95
2 4 0.95
3 2 0.99
4 4 0.99
> ct.best <- train(Class ~ . , data=BreastCancer, method="ctree2",
> tuneGrid=grid)
Fitting: maxdepth=2, mincriterion=0.95
Fitting: maxdepth=4, mincriterion=0.95
Fitting: maxdepth=2, mincriterion=0.99
Fitting: maxdepth=4, mincriterion=0.99
Aggregating results
Selecting tuning parameters
Fitting model on full training set
> print(ct.best)
683 samples
80 predictors
Pre-processing: None
Resampling: Bootstrap (25 reps)
Summary of sample sizes: 683, 683, 683, 683, 683, 683, ...
Resampling results across tuning parameters:
maxdepth Accuracy Kappa mincriterion Accuracy SD Kappa SD mincriterion
SD
2 0.935 0.858 0.97 0.0163 0.0343 0.0202
4 0.935 0.857 0.97 0.0142 0.0322 0.0202
Accuracy was used to select the optimal model using the largest value.
The final value used for the model was maxdepth = 4. ______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
