Datasets
Models
Diff of
/partyMod/tests/RandomForest-regtest.Rout.save
[000000]
..
[fbf06f]
Switch to unified view
| a | b/partyMod/tests/RandomForest-regtest.Rout.save | ||
|---|---|---|---|
| 1 | |||
| 2 | R Under development (unstable) (2014-06-29 r66051) -- "Unsuffered Consequences" |
||
| 3 | Copyright (C) 2014 The R Foundation for Statistical Computing |
||
| 4 | Platform: x86_64-unknown-linux-gnu (64-bit) |
||
| 5 | |||
| 6 | R is free software and comes with ABSOLUTELY NO WARRANTY. |
||
| 7 | You are welcome to redistribute it under certain conditions. |
||
| 8 | Type 'license()' or 'licence()' for distribution details. |
||
| 9 | |||
| 10 | R is a collaborative project with many contributors. |
||
| 11 | Type 'contributors()' for more information and |
||
| 12 | 'citation()' on how to cite R or R packages in publications. |
||
| 13 | |||
| 14 | Type 'demo()' for some demos, 'help()' for on-line help, or |
||
| 15 | 'help.start()' for an HTML browser interface to help. |
||
| 16 | Type 'q()' to quit R. |
||
| 17 | |||
| 18 | > |
||
| 19 | > set.seed(290875) |
||
| 20 | > library("party") |
||
| 21 | Loading required package: grid |
||
| 22 | Loading required package: zoo |
||
| 23 | |||
| 24 | Attaching package: 'zoo' |
||
| 25 | |||
| 26 | The following objects are masked from 'package:base': |
||
| 27 | |||
| 28 | as.Date, as.Date.numeric |
||
| 29 | |||
| 30 | Loading required package: sandwich |
||
| 31 | Loading required package: strucchange |
||
| 32 | Loading required package: modeltools |
||
| 33 | Loading required package: stats4 |
||
| 34 | > if (!require("TH.data")) |
||
| 35 | + stop("cannot load package TH.data") |
||
| 36 | Loading required package: TH.data |
||
| 37 | > if (!require("coin")) |
||
| 38 | + stop("cannot load package coin") |
||
| 39 | Loading required package: coin |
||
| 40 | Loading required package: survival |
||
| 41 | Loading required package: splines |
||
| 42 | > |
||
| 43 | > data("GlaucomaM", package = "TH.data") |
||
| 44 | > rf <- cforest(Class ~ ., data = GlaucomaM, control = cforest_unbiased(ntree = 30)) |
||
| 45 | > stopifnot(mean(GlaucomaM$Class != predict(rf)) < |
||
| 46 | + mean(GlaucomaM$Class != predict(rf, OOB = TRUE))) |
||
| 47 | > |
||
| 48 | > data("GBSG2", package = "TH.data") |
||
| 49 | > rfS <- cforest(Surv(time, cens) ~ ., data = GBSG2, control = cforest_unbiased(ntree = 30)) |
||
| 50 | > treeresponse(rfS, newdata = GBSG2[1:2,]) |
||
| 51 | $`1` |
||
| 52 | Call: survfit(formula = y ~ 1, weights = weights) |
||
| 53 | |||
| 54 | records n.max n.start events median 0.95LCL 0.95UCL |
||
| 55 | 143 394 394 237 1528 1306 1675 |
||
| 56 | |||
| 57 | $`2` |
||
| 58 | Call: survfit(formula = y ~ 1, weights = weights) |
||
| 59 | |||
| 60 | records n.max n.start events median 0.95LCL 0.95UCL |
||
| 61 | 145 380 380 160 2015 1807 2018 |
||
| 62 | |||
| 63 | > |
||
| 64 | > ### give it a try, at least |
||
| 65 | > varimp(rf, pre1.0_0 = TRUE) |
||
| 66 | ag at as an ai |
||
| 67 | 0.0000000000 0.0004629630 0.0027777778 0.0013888889 0.0032407407 |
||
| 68 | eag eat eas ean eai |
||
| 69 | 0.0000000000 0.0000000000 0.0000000000 0.0027777778 -0.0004629630 |
||
| 70 | abrg abrt abrs abrn abri |
||
| 71 | 0.0000000000 0.0023148148 0.0013888889 0.0018518519 0.0046296296 |
||
| 72 | hic mhcg mhct mhcs mhcn |
||
| 73 | 0.0069444444 0.0000000000 0.0009259259 0.0000000000 0.0009259259 |
||
| 74 | mhci phcg phct phcs phcn |
||
| 75 | 0.0078703704 0.0097222222 0.0000000000 0.0000000000 -0.0004629630 |
||
| 76 | phci hvc vbsg vbst vbss |
||
| 77 | 0.0171296296 0.0018518519 0.0013888889 -0.0004629630 0.0018518519 |
||
| 78 | vbsn vbsi vasg vast vass |
||
| 79 | 0.0000000000 0.0000000000 -0.0023148148 0.0000000000 0.0000000000 |
||
| 80 | vasn vasi vbrg vbrt vbrs |
||
| 81 | 0.0000000000 0.0018518519 0.0000000000 0.0013888889 -0.0004629630 |
||
| 82 | vbrn vbri varg vart vars |
||
| 83 | 0.0032407407 0.0004629630 0.0351851852 0.0000000000 0.0254629630 |
||
| 84 | varn vari mdg mdt mds |
||
| 85 | 0.0138888889 0.0425925926 0.0000000000 0.0000000000 -0.0023148148 |
||
| 86 | mdn mdi tmg tmt tms |
||
| 87 | 0.0032407407 0.0004629630 0.0222222222 0.0009259259 0.0069444444 |
||
| 88 | tmn tmi mr rnf mdic |
||
| 89 | -0.0027777778 0.0273148148 0.0000000000 0.0055555556 0.0074074074 |
||
| 90 | emd mv |
||
| 91 | 0.0000000000 -0.0013888889 |
||
| 92 | > |
||
| 93 | > P <- proximity(rf) |
||
| 94 | > stopifnot(max(abs(P - t(P))) == 0) |
||
| 95 | > |
||
| 96 | > P[1:10,1:10] |
||
| 97 | 2 43 25 65 70 16 6 |
||
| 98 | 2 1.00000000 0.15384615 0.7500000 0.0000000 0.07142857 0.13333333 0.7142857 |
||
| 99 | 43 0.15384615 1.00000000 0.1818182 0.0000000 0.11111111 0.45454545 0.1111111 |
||
| 100 | 25 0.75000000 0.18181818 1.0000000 0.1818182 0.11111111 0.14285714 0.8000000 |
||
| 101 | 65 0.00000000 0.00000000 0.1818182 1.0000000 0.00000000 0.00000000 0.1666667 |
||
| 102 | 70 0.07142857 0.11111111 0.1111111 0.0000000 1.00000000 0.00000000 0.1428571 |
||
| 103 | 16 0.13333333 0.45454545 0.1428571 0.0000000 0.00000000 1.00000000 0.0000000 |
||
| 104 | 6 0.71428571 0.11111111 0.8000000 0.1666667 0.14285714 0.00000000 1.0000000 |
||
| 105 | 5 0.58823529 0.09090909 0.7692308 0.5000000 0.09090909 0.08333333 0.5000000 |
||
| 106 | 12 0.44444444 0.00000000 0.5714286 0.5833333 0.07692308 0.06666667 0.3333333 |
||
| 107 | 63 0.46153846 0.10000000 0.5000000 0.2222222 0.00000000 0.18181818 0.5000000 |
||
| 108 | 5 12 63 |
||
| 109 | 2 0.58823529 0.44444444 0.4615385 |
||
| 110 | 43 0.09090909 0.00000000 0.1000000 |
||
| 111 | 25 0.76923077 0.57142857 0.5000000 |
||
| 112 | 65 0.50000000 0.58333333 0.2222222 |
||
| 113 | 70 0.09090909 0.07692308 0.0000000 |
||
| 114 | 16 0.08333333 0.06666667 0.1818182 |
||
| 115 | 6 0.50000000 0.33333333 0.5000000 |
||
| 116 | 5 1.00000000 0.76923077 0.5454545 |
||
| 117 | 12 0.76923077 1.00000000 0.5714286 |
||
| 118 | 63 0.54545455 0.57142857 1.0000000 |
||
| 119 | > |
||
| 120 | > ### variable importances |
||
| 121 | > a <- cforest(Species ~ ., data = iris, |
||
| 122 | + control = cforest_unbiased(mtry = 2, ntree = 10)) |
||
| 123 | > varimp(a, pre1.0_0 = TRUE) |
||
| 124 | Sepal.Length Sepal.Width Petal.Length Petal.Width |
||
| 125 | 0.036363636 0.007272727 0.312727273 0.276363636 |
||
| 126 | > varimp(a, conditional = TRUE) |
||
| 127 | Sepal.Length Sepal.Width Petal.Length Petal.Width |
||
| 128 | 0.003636364 -0.003636364 0.167272727 0.194545455 |
||
| 129 | > |
||
| 130 | > airq <- subset(airquality, complete.cases(airquality)) |
||
| 131 | > a <- cforest(Ozone ~ ., data = airq, |
||
| 132 | + control = cforest_unbiased(mtry = 2, ntree = 10)) |
||
| 133 | > varimp(a, pre1.0_0 = TRUE) |
||
| 134 | Solar.R Wind Temp Month Day |
||
| 135 | 139.397699 501.974401 500.220403 28.532700 3.806919 |
||
| 136 | > varimp(a, conditional = TRUE) |
||
| 137 | Solar.R Wind Temp Month Day |
||
| 138 | 93.220640 334.737163 212.686904 14.329278 2.061793 |
||
| 139 | > |
||
| 140 | > data("mammoexp", package = "TH.data") |
||
| 141 | > a <- cforest(ME ~ ., data = mammoexp, control = cforest_classical(ntree = 10)) |
||
| 142 | > varimp(a, pre1.0_0 = TRUE) |
||
| 143 | SYMPT PB HIST BSE DECT |
||
| 144 | 0.027998627 0.021174836 0.018630793 0.002646901 0.005578231 |
||
| 145 | > varimp(a, conditional = TRUE) |
||
| 146 | SYMPT PB HIST BSE DECT |
||
| 147 | 0.021408831 0.012420497 0.013407572 0.001282682 0.002857143 |
||
| 148 | > |
||
| 149 | > stopifnot(all.equal(unique(sapply(a@weights, sum)), nrow(mammoexp))) |
||
| 150 | > |
||
| 151 | > ### check user-defined weights |
||
| 152 | > nobs <- nrow(GlaucomaM) |
||
| 153 | > i <- rep(0.0, nobs) |
||
| 154 | > i[1:floor(.632 * nobs)] <- 1 |
||
| 155 | > folds <- replicate(100, sample(i)) |
||
| 156 | > rf2 <- cforest(Class ~ ., data = GlaucomaM, control = cforest_unbiased(ntree = 100), weights = folds) |
||
| 157 | > table(predict(rf), predict(rf2)) |
||
| 158 | |||
| 159 | glaucoma normal |
||
| 160 | glaucoma 89 4 |
||
| 161 | normal 2 101 |
||
| 162 | > |
||
| 163 | > proc.time() |
||
| 164 | user system elapsed |
||
| 165 | 3.132 0.052 3.185 |