1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
|
test_that("classif_probit", {
# suppressed warnings: "glm.fit: algorithm did not converge"
# "glm.fit: fitted probabilities numerically 0 or 1 occurred"
m = suppressWarnings(
glm(formula = binaryclass.formula, data = binaryclass.train,
family = binomial(link = "probit"))
)
p = predict(m, newdata = binaryclass.test, type = "response")
p.prob = 1 - p
p.class = as.factor(binaryclass.class.levs[ifelse(p > 0.5, 2, 1)])
suppressWarnings(
testSimple("classif.probit", binaryclass.df, binaryclass.target,
binaryclass.train.inds, p.class)
)
suppressWarnings(
testProb("classif.probit", binaryclass.df, binaryclass.target,
binaryclass.train.inds, p.prob)
)
tt = function(formula, data) {
glm(formula, data = data, family = binomial(link = "probit"))
}
tp = function(model, newdata) {
p = predict(model, newdata, type = "response")
as.factor(binaryclass.class.levs[ifelse(p > 0.5, 2, 1)])
}
suppressWarnings(
testCV("classif.probit", binaryclass.df, binaryclass.target, tune.train = tt,
tune.predict = tp)
)
})
|
