I am running this piece of code:
rf.mod <- caret::train(as.factor(traincpdf$lab) ~ ., method="rf",
preProcess = "pca", data=traincpdf, verbose = FALSE)
And I obtained the following:
*** caught segfault ***
address 0x110, cause 'memory not mapped'
*** caught segfault ***
address 0x110, cause 'memory not mapped'
*** caught segfault ***
address 0x110, cause 'memory not mapped'
*** caught segfault ***
address 0x110, cause 'memory not mapped'
Traceback:
1: La.svd(x, nu, nv)
2: svd(x, nu = 0)
3: prcomp.default(x[, method$pca, drop = FALSE], scale = TRUE, retx = FALSE)
4: prcomp(x[, method$pca, drop = FALSE], scale = TRUE, retx = FALSE)
5: preProcess.default(thresh = 0.95, k = 5, method = "pca", x = c(0.86275, 0.90196, 0.90196, 0.87059, 1, 0.78431, 0.78431, 0.85359, 0.85359, 1, 1, 1, 0.71503, 1, 1, 1, 1, 1, 0.95948, 0.94379, 0.93464, 0.94248, 0.95425, 0.93072, 0.94641, 0.94641, 0.94641, 0.9634, 0.9634, 1, 1, 1, 1, 0.79281, 0.79281, 1, 1, 1, 0.90327, 1, 1, 1, 1, 1, 1, 1, 0.86667, 0.6549, 0.6902, 0.91242, 0.91242, 0.91765, 1, 1, 1, 1, 1, 1, 1, 0.97386, 0.97386, 0.97386, 1, 1, 0.078431, 0.078431, 1, 0.75033, 0.78039, 1, 1, 0.84837, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0.20131, 1, 1, 1, 1, 0.16601, 0.16601, 0.96993, 1, 0.99216, 0.99216, 0.99216, 0.94379, 0.94379, 0.95686, 0.96732, 1, 1, 1, 1, 1, 0.96863, 0.96863, 0.96863,
Traceback:
0.96863, 0.96863, 1,
Traceback:
Traceback:
1: 1: 1: 0.9281, 0.9281, 0.9281, 0.95425, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, La.svd(x, nu, nv)La.svd(x, nu, nv)La.svd(x, nu, nv)
...
I found this post where they suggest to remove and re-install all packages, which seems a little drastic to me.
All those commands La.svd(x, nu, nv)
, prcomp
run perfectly on their own and the caret command itself runs perfectly without the pca option, i.e. :
rf.mod <- caret::train(as.factor(traincpdf$lab) ~ ., method="rf",
data=traincpdf, verbose = FALSE)
Anyone has any idea on how to solve this
*** caught segfault ***
address 0x110, cause 'memory not mapped'
error
Thanks