tests/testthat/test_da_multinom.R
In schiffner/locClass: Collection of Local Classification Methods
context("damultinom")
test_that("damultinom: misspecified arguments", {
data(iris)
# wrong variable names
expect_error(damultinom(formula = Species ~ V1, data = iris, wf = "gaussian", bw = 10, trace = FALSE))
# wrong class
# expect_error(damultinom(formula = iris, data = iris, wf = "gaussian", bw = 10, trace = FALSE)) #??? Sepal.Length becomes response
# expect_error(damultinom(iris, data = iris, wf = "gaussian", bw = 10, trace = FALSE)) #??? Sepal.Length becomes response
# target variable also in x
expect_error(damultinom(iris$Species ~ iris, wf = "gaussian", bw = 10, trace = FALSE)) #
expect_warning(damultinom(Species ~ Species + Petal.Width, data = iris, wf = "gaussian", bw = 10, trace = FALSE)) ## warning, Species on RHS removed
## itr
expect_that(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, itr = -5, trace = FALSE), throws_error("'itr' must be >= 1"))
expect_that(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, itr = 0, trace = FALSE), throws_error("'itr' must be >= 1"))
})
test_that("damultinom throws a warning if grouping variable is numeric", {
data(iris)
# formula, data
expect_that(damultinom(formula = Sepal.Length ~ ., data = iris, wf = "gaussian", bw = 10, trace = FALSE), gives_warning("response variable was coerced to a factor"))
# grouping, x
expect_that(damultinom(Petal.Width ~ Petal.Length + Sepal.Width + Sepal.Length, data = iris, wf = "gaussian", bw = 10, trace = FALSE), gives_warning("response variable was coerced to a factor"))
})
test_that("dalda: training data from only one class", {
data(iris)
expect_that(damultinom(Species ~ ., data = iris, bw = 2, subset = 1:50, trace = FALSE), throws_error("need two or more classes to fit a damultinom model"))
})
test_that("damultinom: one training observation", {
data(iris)
# one training observation
expect_that(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = 1, trace = FALSE), throws_error("need two or more classes to fit a damultinom model"))
# one training observation in one predictor variable
expect_that(damultinom(Species ~ Petal.Width, data = iris, wf = "gaussian", bw = 1, subset = 1, trace = FALSE), throws_error("need two or more classes to fit a damultinom model"))
})
test_that("damultinom: initial weighting works correctly", {
data(iris)
## check if weighted solution with initial weights = 1 equals unweighted solution
fit1 <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 2, trace = FALSE)
fit2 <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 2, weights = rep(1,150), trace = FALSE)
expect_equal(fit1[-16],fit2[-16])
## returned weights
a <- rep(1,150)
names(a) <- 1:150
expect_equal(fit1$weights[[1]], a)
expect_equal(fit1$weights, fit2$weights)
## weights and subsetting
# formula, data
expect_that(fit <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 2, subset = 11:60, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
a <- rep(1,50)
names(a) <- 11:60
expect_equal(fit$weights[[1]], a)
# formula, data, weights
a <- rep(1:3,50)[11:60]
a <- a/sum(a) * length(a)
names(a) <- 11:60
expect_that(fit <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 2, weights = rep(1:3, 50), subset = 11:60, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(fit$weights[[1]], a)
## wrong specification of weights argument
# weights in a matrix
weight <- matrix(seq(1:150), nrow = 50)
expect_error(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 2, weights = weight, trace = FALSE))
# weights < 0
expect_error(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 2, weights = rep(-5, 150), trace = FALSE))
# weights true/false
expect_error(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 2, weights = TRUE, trace = FALSE))
})
test_that("damultinom breaks out of for-loop if only one class is left", {
expect_that(fit <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 0.5, k = 30, trace = FALSE), gives_warning("for at least one class all weights are zero"))
expect_equal(fit$itr, 3)
expect_equal(length(fit$weights), 4)
expect_that(fit <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 0.5, k = 2, trace = FALSE), gives_warning("for at least one class all weights are zero"))
expect_that(fit <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 0.5, k = 2, trace = FALSE), gives_warning("training data from only one class, breaking out of iterative procedure")) ###
expect_equal(fit$itr, 1)
expect_equal(length(fit$weights), 2)
})
#sapply(fit$weights, function(x) return(list(sum(x[1:50]), sum(x[51:100]), sum(x[101:150]))))
test_that("damultinom: subsetting works", {
data(iris)
# formula, data
expect_that(fit1 <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 2, subset = 1:80, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_that(fit2 <- damultinom(Species ~ ., data = iris[1:80,], wf = "gaussian", bw = 2, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(fit1[-16],fit2[-16])
a <- rep(1,80)
names(a) <- 1:80
expect_equal(fit1$weights[[1]], a)
# formula, data, weights
expect_that(fit1 <- damultinom(Species ~ ., data = iris, weights = rep(1:3, each = 50), wf = "gaussian", bw = 2, subset = 1:80, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_that(fit2 <- damultinom(Species ~ ., data = iris[1:80,], weights = rep(1:3, each = 50)[1:80], wf = "gaussian", bw = 2, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(fit1[-16],fit2[-16])
a <- rep(80, 4)
names(a) <- 0:3
expect_equal(sapply(fit1$weights, length), a)
b <- rep(1:3, each = 50)[1:80]
b <- b/sum(b) * length(b)
names(b) <- 1:80
expect_equal(fit1$weights[[1]], b)
# wrong specification of subset argument
expect_error(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = iris[1:10,], trace = FALSE))
expect_error(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = FALSE, trace = FALSE))
expect_error(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = 0, trace = FALSE))
expect_error(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = -10:50, trace = FALSE))
})
test_that("damultinom: NA handling works correctly", {
### NA in x
data(iris)
irisna <- iris
irisna[1:10, c(1,3)] <- NA
## formula, data
# na.fail
expect_that(damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 6:60, na.action = na.fail, trace = FALSE), throws_error("missing values in object"))
# check if na.omit works correctly
expect_that(fit1 <- damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 6:60, na.action = na.omit, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_that(fit2 <- damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 11:60, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(fit1[-c(16, 31)], fit2[-16])
a <- rep(50, 4)
names(a) <- 0:3
expect_equal(sapply(fit1$weights, length), a)
## formula, data, weights
# na.fail
expect_that(damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 6:60, weights = rep(1:3, 50), na.action = na.fail, trace = FALSE), throws_error("missing values in object"))
# check if na.omit works correctly
expect_that(fit1 <- damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 6:60, weights = rep(1:3, 50), na.action = na.omit, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_that(fit2 <- damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 11:60, weights = rep(1:3, 50), trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(fit1[-c(16, 31)], fit2[-16])
a <- rep(50, 4)
names(a) <- 0:3
expect_equal(sapply(fit1$weights, length), a)
### NA in grouping
irisna <- iris
irisna$Species[1:10] <- NA
## formula, data
# na.fail
expect_that(damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 6:60, na.action = na.fail, trace = FALSE), throws_error("missing values in object"))
# check if na.omit works correctly
expect_that(fit1 <- damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 6:60, na.action = na.omit, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_that(fit2 <- damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 11:60, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(fit1[-c(16, 31)], fit2[-16])
a <- rep(50, 4)
names(a) <- 0:3
expect_equal(sapply(fit1$weights, length), a)
## formula, data, weights
# na.fail
expect_that(damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 6:60, weights = rep(1:3, 50), na.action = na.fail, trace = FALSE), throws_error("missing values in object"))
# check if na.omit works correctly
expect_that(fit1 <- damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 6:60, weights = rep(1:3, 50), na.action = na.omit, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_that(fit2 <- damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 11:60, weights = rep(1:3, 50), trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(fit1[-c(16, 31)], fit2[-16])
a <- rep(50, 4)
names(a) <- 0:3
expect_equal(sapply(fit1$weights, length), a)
### NA in weights
weights <- rep(1:3,50)
weights[1:10] <- NA
## formula, data, weights
# na.fail
expect_that(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = 6:60, weights = weights, na.action = na.fail, trace = FALSE), throws_error("missing values in object"))
# check if na.omit works correctly
expect_that(fit1 <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = 6:60, weights = weights, na.action = na.omit, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_that(fit2 <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = 11:60, weights = weights, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(fit1[-c(16, 31)], fit2[-16])
a <- rep(50, 4)
names(a) <- 0:3
expect_equal(sapply(fit1$weights, length), a)
### NA in subset
subset <- 6:60
subset[1:5] <- NA
## formula, data
# na.fail
expect_that(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = subset, na.action = na.fail, trace = FALSE), throws_error("missing values in object"))
# check if na.omit works correctly
expect_that(fit1 <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = subset, na.action = na.omit, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_that(fit2 <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = 11:60, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(fit1[-c(16, 31)], fit2[-16])
a <- rep(50, 4)
names(a) <- 0:3
expect_equal(sapply(fit1$weights, length), a)
## formula, data, weights
# na.fail
expect_that(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = subset, weights = rep(1:3, 50), na.action = na.fail, trace = FALSE), throws_error("missing values in object"))
# check if na.omit works correctly
expect_that(fit1 <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = subset, weights = rep(1:3, 50), na.action = na.omit, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_that(fit2 <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = 11:60, weights = rep(1:3, 50), trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(fit1[-c(16, 31)], fit2[-16])
a <- rep(50, 4)
names(a) <- 0:3
expect_equal(sapply(fit1$weights, length), a)
})
test_that("damultinom: try all weight functions", {
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 0.5, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = gaussian(0.5), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 0.5, k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = gaussian(bw = 0.5, k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "epanechnikov", bw = 5, k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = epanechnikov(bw = 5, k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "rectangular", bw = 5, k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = rectangular(bw = 5, k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "triangular", bw = 5, k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = triangular(5, k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "biweight", bw = 5, k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = biweight(5, k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "optcosine", bw = 5, k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = optcosine(5, k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "cosine", bw = 5, k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = cosine(5, k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
})
test_that("damultinom: local solution with rectangular window function and large bw and global solution coincide", {
library(nnet)
fit1 <- multinom(formula = Species ~ ., data = iris, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = rectangular(20), trace = FALSE)
expect_equal(fit1[-c(16,18)], fit2[-c(16:24,26)])
expect_equivalent(as.vector(fit1$weights), fit2$weights[[1]])
})
test_that("damultinom: arguments related to weighting misspecified", {
# bw, k not required
expect_that(fit1 <- damultinom(Species ~ ., data = iris, wf = gaussian(0.5), k = 30, bw = 0.5, trace = FALSE), gives_warning(c("argument 'k' is ignored", "argument 'bw' is ignored")))
fit2 <- damultinom(Species ~ ., data = iris, wf = gaussian(0.5), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
expect_that(fit1 <- damultinom(Species ~ ., data = iris, wf = gaussian(0.5), bw = 0.5, trace = FALSE), gives_warning("argument 'bw' is ignored"))
fit2 <- damultinom(Species ~ ., data = iris, wf = gaussian(0.5), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
expect_equal(fit1$k, NULL)
expect_equal(fit1$nn.only, NULL)
expect_equal(fit1$bw, 0.5)
expect_equal(fit1$adaptive, FALSE)
expect_that(fit1 <- damultinom(Species ~ ., data = iris, wf = function(x) exp(-x), bw = 0.5, k = 30, trace = FALSE), gives_warning(c("argument 'k' is ignored", "argument 'bw' is ignored")))
expect_that(fit2 <- damultinom(Species ~ ., data = iris, wf = function(x) exp(-x), k = 30, trace = FALSE), gives_warning("argument 'k' is ignored"))
expect_equal(fit1[-16], fit2[-16])
expect_equal(fit1$k, NULL)
expect_equal(fit1$nn.only, NULL)
expect_equal(fit1$bw, NULL)
expect_equal(fit1$adaptive, NULL)
expect_that(fit1 <- damultinom(Species ~ ., data = iris, wf = function(x) exp(-x), bw = 0.5, trace = FALSE), gives_warning("argument 'bw' is ignored"))
fit2 <- damultinom(Species ~ ., data = iris, wf = function(x) exp(-x), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
expect_equal(fit1$k, NULL)
expect_equal(fit1$nn.only, NULL)
expect_equal(fit1$bw, NULL)
expect_equal(fit1$adaptive, NULL)
# bw, k missing
expect_that(damultinom(formula = Species ~ ., data = iris, wf = gaussian(), trace = FALSE), throws_error("either 'bw' or 'k' have to be specified"))
expect_that(damultinom(formula = Species ~ ., data = iris, wf = gaussian(), k = 10, trace = FALSE), throws_error("either 'bw' or 'k' have to be specified"))
expect_that(damultinom(Species ~ ., data = iris, trace = FALSE), throws_error("either 'bw' or 'k' have to be specified"))
# bw < 0
expect_that(damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = -5, trace = FALSE), throws_error("'bw' must be positive"))
expect_that(damultinom(formula = Species ~ ., data = iris, wf = "cosine", k = 10, bw = -50, trace = FALSE), throws_error("'bw' must be positive"))
# bw vector
expect_that(damultinom(formula = Species ~., data = iris, wf = "gaussian", bw = rep(1, nrow(iris)), trace = FALSE), gives_warning("only first element of 'bw' used"))
# k < 0
expect_that(damultinom(formula = Species ~ ., data = iris, wf = "gaussian", k =-7, bw = 50, trace = FALSE), throws_error("'k' must be positive"))
# k too small
# expect_error(damultinom(formula = Species ~ ., data = iris, wf = "gaussian", k = 5, bw = 0.005, trace = FALSE))
# k too large
expect_that(damultinom(formula = Species ~ ., data = iris, k = 250, wf = "gaussian", bw = 50, trace = FALSE), throws_error("'k' is larger than 'n'"))
# k vector
expect_that(damultinom(formula = Species ~., data = iris, wf = "gaussian", k = rep(50, nrow(iris)), trace = FALSE), gives_warning("only first element of 'k' used"))
})
test_that("damultinom: weighting schemes work", {
## wf with finite support
# fixed bw
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "rectangular", bw = 5, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = rectangular(bw = 5), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
expect_equal(fit1$bw, 5)
expect_equal(fit1$k, NULL)
expect_equal(fit1$nn.only, NULL)
expect_true(!fit1$adaptive)
# adaptive bw, only knn
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "rectangular", k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = rectangular(k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
is.null(fit1$bw)
expect_equal(fit1$k, 50)
expect_equal(fit1$bw, NULL)
expect_true(fit1$nn.only)
expect_true(fit1$adaptive)
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
# fixed bw, only knn
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "rectangular", bw = 5, k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = rectangular(bw = 5, k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
expect_equal(fit1$bw, 5)
expect_equal(fit1$k, 50)
expect_true(fit1$nn.only)
expect_true(!fit1$adaptive)
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
# nn.only not needed
expect_that(damultinom(formula = Species ~ ., data = iris, wf = "rectangular", bw = 5, nn.only = TRUE, trace = FALSE), gives_warning("argument 'nn.only' is ignored"))
# nn.only has to be TRUE if bw and k are both given
expect_that(damultinom(formula = Species ~ ., data = iris, wf = "rectangular", bw = 5, k = 50, nn.only = FALSE, trace = FALSE), throws_error("if 'bw' and 'k' are given argument 'nn.only' must be TRUE"))
## wf with infinite support
# fixed bw
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 0.5, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = gaussian(bw = 0.5), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
expect_equal(fit1$bw, 0.5)
expect_equal(fit1$k, NULL)
expect_equal(fit1$nn.only, NULL)
expect_true(!fit1$adaptive)
a <- rep(150, 4)
names(a) <- 0:3
expect_equal(sapply(fit1$weights, function(x) sum(x > 0)), a)
# adaptive bw, only knn
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = gaussian(k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
expect_equal(fit1$bw, NULL)
expect_equal(fit1$k, 50)
expect_equal(fit1$nn.only, TRUE)
expect_true(fit1$adaptive)
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
# adaptive bw, all obs
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", k = 50, nn.only = FALSE, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = gaussian(k = 50, nn.only = FALSE), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
expect_equal(fit1$bw, NULL)
expect_equal(fit1$k, 50)
expect_equal(fit1$nn.only, FALSE)
expect_true(fit1$adaptive)
a <- rep(150, 4)
names(a) <- 0:3
expect_equal(sapply(fit1$weights, function(x) sum(x > 0)), a)
# fixed bw, only knn
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 1, k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = gaussian(bw = 1, k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
expect_equal(fit1$bw, 1)
expect_equal(fit1$k, 50)
expect_equal(fit1$nn.only, TRUE)
expect_true(!fit1$adaptive)
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
# nn.only has to be TRUE if bw and k are both given
expect_that(damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 1, k = 50, nn.only = FALSE, trace = FALSE), throws_error("if 'bw' and 'k' are given argument 'nn.only' must be TRUE"))
})
#=================================================================================================================
context("predict.damultinom")
test_that("predict.damultinom works correctly with formula and data.frame interface and with missing newdata", {
data(iris)
ran <- sample(1:150,100)
## formula, data
fit <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 2, subset = ran, trace = FALSE)
pred <- predict(fit)
expect_equal(rownames(pred$posterior), rownames(iris)[ran])
## formula, data, newdata
fit <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 2, subset = ran, trace = FALSE)
pred <- predict(fit, newdata = iris[-ran,])
expect_equal(rownames(pred$posterior), rownames(iris)[-ran])
})
test_that("predict.dalr: retrieving training data works", {
data(iris)
## no subset
# formula, data
fit <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 2, trace = FALSE)
pred1 <- predict(fit)
pred2 <- predict(fit, newdata = iris)
expect_equal(pred1, pred2)
## subset
ran <- sample(1:150,100)
# formula, data
fit <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 2, subset = ran, trace = FALSE)
pred1 <- predict(fit)
pred2 <- predict(fit, newdata = iris[ran,])
expect_equal(pred1, pred2)
})
test_that("predict.damultinom works with missing classes in the training data", {
data(iris)
ran <- sample(1:150,100)
expect_that(fit <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = 1:100, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(ncol(fit$fitted.values), 1)
expect_equal(ncol(fit$residuals), 1)
expect_true(fit$entropy)
pred <- predict(fit, newdata = iris[-ran,])
expect_equal(nlevels(pred$class), 3)
expect_equal(ncol(pred$posterior), 2)
})
test_that("predict.damultinom works with one single predictor variable", {
data(iris)
ran <- sample(1:150,100)
fit <- damultinom(Species ~ Petal.Width, data = iris, wf = "gaussian", bw = 2, subset = ran, trace = FALSE)
expect_equal(fit$coefnames, c("(Intercept)", "Petal.Width"))
pred <- predict(fit, newdata = iris[-ran,])
})
test_that("predict.damultinom works with one single test observation", {
data(iris)
ran <- sample(1:150,100)
fit <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 2, subset = ran, trace = FALSE)
pred <- predict(fit, newdata = iris[5,])
expect_equal(length(pred$class), 1)
expect_equal(dim(pred$posterior), c(1, 3))
a <- factor("setosa", levels = c("setosa", "versicolor", "virginica"))
names(a) <- "5"
expect_equal(pred$class, a)
pred <- predict(fit, newdata = iris[58,])
expect_equal(length(pred$class), 1)
expect_equal(dim(pred$posterior), c(1, 3))
a <- factor("versicolor", levels = c("setosa", "versicolor", "virginica"))
names(a) = "58"
expect_equal(pred$class, a)
})
test_that("predict.damultinom works with one single predictor variable and one single test observation", {
data(iris)
ran <- sample(1:150,100)
fit <- damultinom(Species ~ Petal.Width, data = iris, wf = "gaussian", bw = 2, subset = ran, trace = FALSE)
pred <- predict(fit, newdata = iris[5,])
expect_equal(length(pred$class), 1)
expect_equal(dim(pred$posterior), c(1, 3))
})
test_that("predict.damultinom: NA handling in newdata works", {
data(iris)
ran <- sample(1:150,100)
irisna <- iris
irisna[1:17,c(1,3)] <- NA
fit <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 50, subset = ran, trace = FALSE)
pred <- predict(fit, newdata = irisna)
expect_equal(all(is.na(pred$class[1:17])), TRUE)
expect_equal(all(is.na(pred$posterior[1:17,])), TRUE)
})
test_that("predict.damultinom: misspecified arguments", {
data(iris)
ran <- sample(1:150,100)
fit <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 2, subset = ran, trace = FALSE)
# errors in newdata
expect_error(predict(fit, newdata = TRUE))
expect_error(predict(fit, newdata = -50:50))
})
#=================================================================================================================
## eqivalence global and local with large bandwidth
# options(contrasts = c("contr.treatment", "contr.poly"))
# library(MASS)
# example(birthwt)
# (bwt.mu <- multinom(low ~ ., bwt))
# # Not run: Call:
# dg <- multinom(formula = low ~ ., data = bwt)
# d <- damultinom(formula = low ~ ., data = bwt, wf = "rectangular", bw = 10)
## plot
# irisscale <- data.frame(scale(iris[,1:4]), Species = iris$Species)
# d <- damultinom(formula = Species ~ Sepal.Length + Sepal.Width, data = irisscale, wf = "gaussian", bw = 0.5)
# plot(irisscale[,1:2], cex = d$weights[[4]], col = iris$Species)
# x <- seq(-3,3.5,0.05)
# y <- seq(-3,3,0.05)
# iris.grid <- expand.grid(Sepal.Length = x, Sepal.Width = y)
# pred <- predict(d, newdata = iris.grid)
# contour(x = x, y = y, z = matrix(pred$posterior[,1], length(x)))
# points(irisscale[,1:2], cex = d$weights[[4]], col = iris$Species)
# contour(x = x, y = y, z = matrix(pred$posterior[,2], length(x)))
# points(irisscale[,1:2], cex = d$weights[[4]], col = iris$Species)
# contour(x = x, y = y, z = matrix(pred$posterior[,3], length(x)))
# points(irisscale[,1:2], cex = d$weights[[4]], col = iris$Species)
schiffner/locClass documentation built on May 29, 2019, 3:39 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
context("damultinom")
test_that("damultinom: misspecified arguments", {
data(iris)
# wrong variable names
expect_error(damultinom(formula = Species ~ V1, data = iris, wf = "gaussian", bw = 10, trace = FALSE))
# wrong class
# expect_error(damultinom(formula = iris, data = iris, wf = "gaussian", bw = 10, trace = FALSE)) #??? Sepal.Length becomes response
# expect_error(damultinom(iris, data = iris, wf = "gaussian", bw = 10, trace = FALSE)) #??? Sepal.Length becomes response
# target variable also in x
expect_error(damultinom(iris$Species ~ iris, wf = "gaussian", bw = 10, trace = FALSE)) #
expect_warning(damultinom(Species ~ Species + Petal.Width, data = iris, wf = "gaussian", bw = 10, trace = FALSE)) ## warning, Species on RHS removed
## itr
expect_that(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, itr = -5, trace = FALSE), throws_error("'itr' must be >= 1"))
expect_that(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, itr = 0, trace = FALSE), throws_error("'itr' must be >= 1"))
})
test_that("damultinom throws a warning if grouping variable is numeric", {
data(iris)
# formula, data
expect_that(damultinom(formula = Sepal.Length ~ ., data = iris, wf = "gaussian", bw = 10, trace = FALSE), gives_warning("response variable was coerced to a factor"))
# grouping, x
expect_that(damultinom(Petal.Width ~ Petal.Length + Sepal.Width + Sepal.Length, data = iris, wf = "gaussian", bw = 10, trace = FALSE), gives_warning("response variable was coerced to a factor"))
})
test_that("dalda: training data from only one class", {
data(iris)
expect_that(damultinom(Species ~ ., data = iris, bw = 2, subset = 1:50, trace = FALSE), throws_error("need two or more classes to fit a damultinom model"))
})
test_that("damultinom: one training observation", {
data(iris)
# one training observation
expect_that(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = 1, trace = FALSE), throws_error("need two or more classes to fit a damultinom model"))
# one training observation in one predictor variable
expect_that(damultinom(Species ~ Petal.Width, data = iris, wf = "gaussian", bw = 1, subset = 1, trace = FALSE), throws_error("need two or more classes to fit a damultinom model"))
})
test_that("damultinom: initial weighting works correctly", {
data(iris)
## check if weighted solution with initial weights = 1 equals unweighted solution
fit1 <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 2, trace = FALSE)
fit2 <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 2, weights = rep(1,150), trace = FALSE)
expect_equal(fit1[-16],fit2[-16])
## returned weights
a <- rep(1,150)
names(a) <- 1:150
expect_equal(fit1$weights[[1]], a)
expect_equal(fit1$weights, fit2$weights)
## weights and subsetting
# formula, data
expect_that(fit <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 2, subset = 11:60, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
a <- rep(1,50)
names(a) <- 11:60
expect_equal(fit$weights[[1]], a)
# formula, data, weights
a <- rep(1:3,50)[11:60]
a <- a/sum(a) * length(a)
names(a) <- 11:60
expect_that(fit <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 2, weights = rep(1:3, 50), subset = 11:60, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(fit$weights[[1]], a)
## wrong specification of weights argument
# weights in a matrix
weight <- matrix(seq(1:150), nrow = 50)
expect_error(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 2, weights = weight, trace = FALSE))
# weights < 0
expect_error(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 2, weights = rep(-5, 150), trace = FALSE))
# weights true/false
expect_error(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 2, weights = TRUE, trace = FALSE))
})
test_that("damultinom breaks out of for-loop if only one class is left", {
expect_that(fit <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 0.5, k = 30, trace = FALSE), gives_warning("for at least one class all weights are zero"))
expect_equal(fit$itr, 3)
expect_equal(length(fit$weights), 4)
expect_that(fit <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 0.5, k = 2, trace = FALSE), gives_warning("for at least one class all weights are zero"))
expect_that(fit <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 0.5, k = 2, trace = FALSE), gives_warning("training data from only one class, breaking out of iterative procedure")) ###
expect_equal(fit$itr, 1)
expect_equal(length(fit$weights), 2)
})
#sapply(fit$weights, function(x) return(list(sum(x[1:50]), sum(x[51:100]), sum(x[101:150]))))
test_that("damultinom: subsetting works", {
data(iris)
# formula, data
expect_that(fit1 <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 2, subset = 1:80, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_that(fit2 <- damultinom(Species ~ ., data = iris[1:80,], wf = "gaussian", bw = 2, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(fit1[-16],fit2[-16])
a <- rep(1,80)
names(a) <- 1:80
expect_equal(fit1$weights[[1]], a)
# formula, data, weights
expect_that(fit1 <- damultinom(Species ~ ., data = iris, weights = rep(1:3, each = 50), wf = "gaussian", bw = 2, subset = 1:80, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_that(fit2 <- damultinom(Species ~ ., data = iris[1:80,], weights = rep(1:3, each = 50)[1:80], wf = "gaussian", bw = 2, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(fit1[-16],fit2[-16])
a <- rep(80, 4)
names(a) <- 0:3
expect_equal(sapply(fit1$weights, length), a)
b <- rep(1:3, each = 50)[1:80]
b <- b/sum(b) * length(b)
names(b) <- 1:80
expect_equal(fit1$weights[[1]], b)
# wrong specification of subset argument
expect_error(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = iris[1:10,], trace = FALSE))
expect_error(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = FALSE, trace = FALSE))
expect_error(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = 0, trace = FALSE))
expect_error(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = -10:50, trace = FALSE))
})
test_that("damultinom: NA handling works correctly", {
### NA in x
data(iris)
irisna <- iris
irisna[1:10, c(1,3)] <- NA
## formula, data
# na.fail
expect_that(damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 6:60, na.action = na.fail, trace = FALSE), throws_error("missing values in object"))
# check if na.omit works correctly
expect_that(fit1 <- damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 6:60, na.action = na.omit, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_that(fit2 <- damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 11:60, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(fit1[-c(16, 31)], fit2[-16])
a <- rep(50, 4)
names(a) <- 0:3
expect_equal(sapply(fit1$weights, length), a)
## formula, data, weights
# na.fail
expect_that(damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 6:60, weights = rep(1:3, 50), na.action = na.fail, trace = FALSE), throws_error("missing values in object"))
# check if na.omit works correctly
expect_that(fit1 <- damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 6:60, weights = rep(1:3, 50), na.action = na.omit, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_that(fit2 <- damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 11:60, weights = rep(1:3, 50), trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(fit1[-c(16, 31)], fit2[-16])
a <- rep(50, 4)
names(a) <- 0:3
expect_equal(sapply(fit1$weights, length), a)
### NA in grouping
irisna <- iris
irisna$Species[1:10] <- NA
## formula, data
# na.fail
expect_that(damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 6:60, na.action = na.fail, trace = FALSE), throws_error("missing values in object"))
# check if na.omit works correctly
expect_that(fit1 <- damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 6:60, na.action = na.omit, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_that(fit2 <- damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 11:60, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(fit1[-c(16, 31)], fit2[-16])
a <- rep(50, 4)
names(a) <- 0:3
expect_equal(sapply(fit1$weights, length), a)
## formula, data, weights
# na.fail
expect_that(damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 6:60, weights = rep(1:3, 50), na.action = na.fail, trace = FALSE), throws_error("missing values in object"))
# check if na.omit works correctly
expect_that(fit1 <- damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 6:60, weights = rep(1:3, 50), na.action = na.omit, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_that(fit2 <- damultinom(Species ~ ., data = irisna, wf = "gaussian", bw = 10, subset = 11:60, weights = rep(1:3, 50), trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(fit1[-c(16, 31)], fit2[-16])
a <- rep(50, 4)
names(a) <- 0:3
expect_equal(sapply(fit1$weights, length), a)
### NA in weights
weights <- rep(1:3,50)
weights[1:10] <- NA
## formula, data, weights
# na.fail
expect_that(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = 6:60, weights = weights, na.action = na.fail, trace = FALSE), throws_error("missing values in object"))
# check if na.omit works correctly
expect_that(fit1 <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = 6:60, weights = weights, na.action = na.omit, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_that(fit2 <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = 11:60, weights = weights, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(fit1[-c(16, 31)], fit2[-16])
a <- rep(50, 4)
names(a) <- 0:3
expect_equal(sapply(fit1$weights, length), a)
### NA in subset
subset <- 6:60
subset[1:5] <- NA
## formula, data
# na.fail
expect_that(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = subset, na.action = na.fail, trace = FALSE), throws_error("missing values in object"))
# check if na.omit works correctly
expect_that(fit1 <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = subset, na.action = na.omit, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_that(fit2 <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = 11:60, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(fit1[-c(16, 31)], fit2[-16])
a <- rep(50, 4)
names(a) <- 0:3
expect_equal(sapply(fit1$weights, length), a)
## formula, data, weights
# na.fail
expect_that(damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = subset, weights = rep(1:3, 50), na.action = na.fail, trace = FALSE), throws_error("missing values in object"))
# check if na.omit works correctly
expect_that(fit1 <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = subset, weights = rep(1:3, 50), na.action = na.omit, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_that(fit2 <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = 11:60, weights = rep(1:3, 50), trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(fit1[-c(16, 31)], fit2[-16])
a <- rep(50, 4)
names(a) <- 0:3
expect_equal(sapply(fit1$weights, length), a)
})
test_that("damultinom: try all weight functions", {
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 0.5, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = gaussian(0.5), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 0.5, k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = gaussian(bw = 0.5, k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "epanechnikov", bw = 5, k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = epanechnikov(bw = 5, k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "rectangular", bw = 5, k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = rectangular(bw = 5, k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "triangular", bw = 5, k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = triangular(5, k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "biweight", bw = 5, k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = biweight(5, k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "optcosine", bw = 5, k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = optcosine(5, k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "cosine", bw = 5, k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = cosine(5, k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
})
test_that("damultinom: local solution with rectangular window function and large bw and global solution coincide", {
library(nnet)
fit1 <- multinom(formula = Species ~ ., data = iris, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = rectangular(20), trace = FALSE)
expect_equal(fit1[-c(16,18)], fit2[-c(16:24,26)])
expect_equivalent(as.vector(fit1$weights), fit2$weights[[1]])
})
test_that("damultinom: arguments related to weighting misspecified", {
# bw, k not required
expect_that(fit1 <- damultinom(Species ~ ., data = iris, wf = gaussian(0.5), k = 30, bw = 0.5, trace = FALSE), gives_warning(c("argument 'k' is ignored", "argument 'bw' is ignored")))
fit2 <- damultinom(Species ~ ., data = iris, wf = gaussian(0.5), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
expect_that(fit1 <- damultinom(Species ~ ., data = iris, wf = gaussian(0.5), bw = 0.5, trace = FALSE), gives_warning("argument 'bw' is ignored"))
fit2 <- damultinom(Species ~ ., data = iris, wf = gaussian(0.5), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
expect_equal(fit1$k, NULL)
expect_equal(fit1$nn.only, NULL)
expect_equal(fit1$bw, 0.5)
expect_equal(fit1$adaptive, FALSE)
expect_that(fit1 <- damultinom(Species ~ ., data = iris, wf = function(x) exp(-x), bw = 0.5, k = 30, trace = FALSE), gives_warning(c("argument 'k' is ignored", "argument 'bw' is ignored")))
expect_that(fit2 <- damultinom(Species ~ ., data = iris, wf = function(x) exp(-x), k = 30, trace = FALSE), gives_warning("argument 'k' is ignored"))
expect_equal(fit1[-16], fit2[-16])
expect_equal(fit1$k, NULL)
expect_equal(fit1$nn.only, NULL)
expect_equal(fit1$bw, NULL)
expect_equal(fit1$adaptive, NULL)
expect_that(fit1 <- damultinom(Species ~ ., data = iris, wf = function(x) exp(-x), bw = 0.5, trace = FALSE), gives_warning("argument 'bw' is ignored"))
fit2 <- damultinom(Species ~ ., data = iris, wf = function(x) exp(-x), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
expect_equal(fit1$k, NULL)
expect_equal(fit1$nn.only, NULL)
expect_equal(fit1$bw, NULL)
expect_equal(fit1$adaptive, NULL)
# bw, k missing
expect_that(damultinom(formula = Species ~ ., data = iris, wf = gaussian(), trace = FALSE), throws_error("either 'bw' or 'k' have to be specified"))
expect_that(damultinom(formula = Species ~ ., data = iris, wf = gaussian(), k = 10, trace = FALSE), throws_error("either 'bw' or 'k' have to be specified"))
expect_that(damultinom(Species ~ ., data = iris, trace = FALSE), throws_error("either 'bw' or 'k' have to be specified"))
# bw < 0
expect_that(damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = -5, trace = FALSE), throws_error("'bw' must be positive"))
expect_that(damultinom(formula = Species ~ ., data = iris, wf = "cosine", k = 10, bw = -50, trace = FALSE), throws_error("'bw' must be positive"))
# bw vector
expect_that(damultinom(formula = Species ~., data = iris, wf = "gaussian", bw = rep(1, nrow(iris)), trace = FALSE), gives_warning("only first element of 'bw' used"))
# k < 0
expect_that(damultinom(formula = Species ~ ., data = iris, wf = "gaussian", k =-7, bw = 50, trace = FALSE), throws_error("'k' must be positive"))
# k too small
# expect_error(damultinom(formula = Species ~ ., data = iris, wf = "gaussian", k = 5, bw = 0.005, trace = FALSE))
# k too large
expect_that(damultinom(formula = Species ~ ., data = iris, k = 250, wf = "gaussian", bw = 50, trace = FALSE), throws_error("'k' is larger than 'n'"))
# k vector
expect_that(damultinom(formula = Species ~., data = iris, wf = "gaussian", k = rep(50, nrow(iris)), trace = FALSE), gives_warning("only first element of 'k' used"))
})
test_that("damultinom: weighting schemes work", {
## wf with finite support
# fixed bw
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "rectangular", bw = 5, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = rectangular(bw = 5), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
expect_equal(fit1$bw, 5)
expect_equal(fit1$k, NULL)
expect_equal(fit1$nn.only, NULL)
expect_true(!fit1$adaptive)
# adaptive bw, only knn
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "rectangular", k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = rectangular(k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
is.null(fit1$bw)
expect_equal(fit1$k, 50)
expect_equal(fit1$bw, NULL)
expect_true(fit1$nn.only)
expect_true(fit1$adaptive)
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
# fixed bw, only knn
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "rectangular", bw = 5, k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = rectangular(bw = 5, k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
expect_equal(fit1$bw, 5)
expect_equal(fit1$k, 50)
expect_true(fit1$nn.only)
expect_true(!fit1$adaptive)
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
# nn.only not needed
expect_that(damultinom(formula = Species ~ ., data = iris, wf = "rectangular", bw = 5, nn.only = TRUE, trace = FALSE), gives_warning("argument 'nn.only' is ignored"))
# nn.only has to be TRUE if bw and k are both given
expect_that(damultinom(formula = Species ~ ., data = iris, wf = "rectangular", bw = 5, k = 50, nn.only = FALSE, trace = FALSE), throws_error("if 'bw' and 'k' are given argument 'nn.only' must be TRUE"))
## wf with infinite support
# fixed bw
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 0.5, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = gaussian(bw = 0.5), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
expect_equal(fit1$bw, 0.5)
expect_equal(fit1$k, NULL)
expect_equal(fit1$nn.only, NULL)
expect_true(!fit1$adaptive)
a <- rep(150, 4)
names(a) <- 0:3
expect_equal(sapply(fit1$weights, function(x) sum(x > 0)), a)
# adaptive bw, only knn
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = gaussian(k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
expect_equal(fit1$bw, NULL)
expect_equal(fit1$k, 50)
expect_equal(fit1$nn.only, TRUE)
expect_true(fit1$adaptive)
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
# adaptive bw, all obs
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", k = 50, nn.only = FALSE, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = gaussian(k = 50, nn.only = FALSE), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
expect_equal(fit1$bw, NULL)
expect_equal(fit1$k, 50)
expect_equal(fit1$nn.only, FALSE)
expect_true(fit1$adaptive)
a <- rep(150, 4)
names(a) <- 0:3
expect_equal(sapply(fit1$weights, function(x) sum(x > 0)), a)
# fixed bw, only knn
fit1 <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 1, k = 50, trace = FALSE)
fit2 <- damultinom(formula = Species ~ ., data = iris, wf = gaussian(bw = 1, k = 50), trace = FALSE)
expect_equal(fit1[-16], fit2[-16])
expect_equal(fit1$bw, 1)
expect_equal(fit1$k, 50)
expect_equal(fit1$nn.only, TRUE)
expect_true(!fit1$adaptive)
a <- rep(50, 3)
names(a) <- 1:3
expect_equal(sapply(fit1$weights[2:4], function(x) sum(x > 0)), a)
# nn.only has to be TRUE if bw and k are both given
expect_that(damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 1, k = 50, nn.only = FALSE, trace = FALSE), throws_error("if 'bw' and 'k' are given argument 'nn.only' must be TRUE"))
})
#=================================================================================================================
context("predict.damultinom")
test_that("predict.damultinom works correctly with formula and data.frame interface and with missing newdata", {
data(iris)
ran <- sample(1:150,100)
## formula, data
fit <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 2, subset = ran, trace = FALSE)
pred <- predict(fit)
expect_equal(rownames(pred$posterior), rownames(iris)[ran])
## formula, data, newdata
fit <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 2, subset = ran, trace = FALSE)
pred <- predict(fit, newdata = iris[-ran,])
expect_equal(rownames(pred$posterior), rownames(iris)[-ran])
})
test_that("predict.dalr: retrieving training data works", {
data(iris)
## no subset
# formula, data
fit <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 2, trace = FALSE)
pred1 <- predict(fit)
pred2 <- predict(fit, newdata = iris)
expect_equal(pred1, pred2)
## subset
ran <- sample(1:150,100)
# formula, data
fit <- damultinom(formula = Species ~ ., data = iris, wf = "gaussian", bw = 2, subset = ran, trace = FALSE)
pred1 <- predict(fit)
pred2 <- predict(fit, newdata = iris[ran,])
expect_equal(pred1, pred2)
})
test_that("predict.damultinom works with missing classes in the training data", {
data(iris)
ran <- sample(1:150,100)
expect_that(fit <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 10, subset = 1:100, trace = FALSE), gives_warning("group ‘virginica’ is empty"))
expect_equal(ncol(fit$fitted.values), 1)
expect_equal(ncol(fit$residuals), 1)
expect_true(fit$entropy)
pred <- predict(fit, newdata = iris[-ran,])
expect_equal(nlevels(pred$class), 3)
expect_equal(ncol(pred$posterior), 2)
})
test_that("predict.damultinom works with one single predictor variable", {
data(iris)
ran <- sample(1:150,100)
fit <- damultinom(Species ~ Petal.Width, data = iris, wf = "gaussian", bw = 2, subset = ran, trace = FALSE)
expect_equal(fit$coefnames, c("(Intercept)", "Petal.Width"))
pred <- predict(fit, newdata = iris[-ran,])
})
test_that("predict.damultinom works with one single test observation", {
data(iris)
ran <- sample(1:150,100)
fit <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 2, subset = ran, trace = FALSE)
pred <- predict(fit, newdata = iris[5,])
expect_equal(length(pred$class), 1)
expect_equal(dim(pred$posterior), c(1, 3))
a <- factor("setosa", levels = c("setosa", "versicolor", "virginica"))
names(a) <- "5"
expect_equal(pred$class, a)
pred <- predict(fit, newdata = iris[58,])
expect_equal(length(pred$class), 1)
expect_equal(dim(pred$posterior), c(1, 3))
a <- factor("versicolor", levels = c("setosa", "versicolor", "virginica"))
names(a) = "58"
expect_equal(pred$class, a)
})
test_that("predict.damultinom works with one single predictor variable and one single test observation", {
data(iris)
ran <- sample(1:150,100)
fit <- damultinom(Species ~ Petal.Width, data = iris, wf = "gaussian", bw = 2, subset = ran, trace = FALSE)
pred <- predict(fit, newdata = iris[5,])
expect_equal(length(pred$class), 1)
expect_equal(dim(pred$posterior), c(1, 3))
})
test_that("predict.damultinom: NA handling in newdata works", {
data(iris)
ran <- sample(1:150,100)
irisna <- iris
irisna[1:17,c(1,3)] <- NA
fit <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 50, subset = ran, trace = FALSE)
pred <- predict(fit, newdata = irisna)
expect_equal(all(is.na(pred$class[1:17])), TRUE)
expect_equal(all(is.na(pred$posterior[1:17,])), TRUE)
})
test_that("predict.damultinom: misspecified arguments", {
data(iris)
ran <- sample(1:150,100)
fit <- damultinom(Species ~ ., data = iris, wf = "gaussian", bw = 2, subset = ran, trace = FALSE)
# errors in newdata
expect_error(predict(fit, newdata = TRUE))
expect_error(predict(fit, newdata = -50:50))
})
#=================================================================================================================
## eqivalence global and local with large bandwidth
# options(contrasts = c("contr.treatment", "contr.poly"))
# library(MASS)
# example(birthwt)
# (bwt.mu <- multinom(low ~ ., bwt))
# # Not run: Call:
# dg <- multinom(formula = low ~ ., data = bwt)
# d <- damultinom(formula = low ~ ., data = bwt, wf = "rectangular", bw = 10)
## plot
# irisscale <- data.frame(scale(iris[,1:4]), Species = iris$Species)
# d <- damultinom(formula = Species ~ Sepal.Length + Sepal.Width, data = irisscale, wf = "gaussian", bw = 0.5)
# plot(irisscale[,1:2], cex = d$weights[[4]], col = iris$Species)
# x <- seq(-3,3.5,0.05)
# y <- seq(-3,3,0.05)
# iris.grid <- expand.grid(Sepal.Length = x, Sepal.Width = y)
# pred <- predict(d, newdata = iris.grid)
# contour(x = x, y = y, z = matrix(pred$posterior[,1], length(x)))
# points(irisscale[,1:2], cex = d$weights[[4]], col = iris$Species)
# contour(x = x, y = y, z = matrix(pred$posterior[,2], length(x)))
# points(irisscale[,1:2], cex = d$weights[[4]], col = iris$Species)
# contour(x = x, y = y, z = matrix(pred$posterior[,3], length(x)))
# points(irisscale[,1:2], cex = d$weights[[4]], col = iris$Species)
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.