# Tests buildVptree().
# library(BiocNeighbors); library(testthat); source("test-build-vptree.R")
set.seed(20000)
test_that("buildVptree() works as expected", {
for (ndim in c(1, 5, 10, 20)) {
for (nobs in c(500, 1000, 2000)) {
X <- matrix(runif(nobs * ndim), nrow=nobs)
out <- buildVptree(X)
expect_identical(dim(out), dim(X))
expect_identical(rev(dim(bndata(out))), dim(X))
expect_identical(sort(bnorder(out)), seq_len(nobs))
expect_identical(bndata(out), t(X[bnorder(out),]))
expect_identical(bndistance(out), "Euclidean")
node.data <- VptreeIndex_nodes(out)
expect_identical(sort(node.data[[1]]), seq_len(nobs)-1L)
expect_true(!anyDuplicated(setdiff(node.data[[2]], -1L)))
expect_true(!anyDuplicated(setdiff(node.data[[3]], -1L)))
expect_true(all(node.data[[4]] >= 0))
}
}
})
set.seed(200001)
test_that("buildVptree() preserves dimension names", {
nobs <- 1011
ndim <- 23
X <- matrix(runif(nobs * ndim), nrow=nobs)
rownames(X) <- paste0("POINT", seq_len(nobs))
colnames(X) <- paste0("DIM", seq_len(ndim))
out <- buildVptree(X)
expect_identical(rownames(out), rownames(X))
expect_identical(rownames(bndata(out)), colnames(X))
expect_identical(colnames(bndata(out)), rownames(X)[bnorder(out)])
# Still true if there are no cells.
out <- buildVptree(X[0,,drop=FALSE])
expect_identical(rownames(bndata(out)), colnames(X))
expect_identical(colnames(bndata(out)), NULL)
expect_identical(rownames(out), NULL)
})
set.seed(200002)
test_that("buildVptree() responds to transposition", {
nobs <- 1011
ndim <- 10
X <- matrix(runif(nobs * ndim), nrow=nobs)
rownames(X) <- paste0("POINT", seq_len(nobs))
colnames(X) <- paste0("DIM", seq_len(ndim))
set.seed(101)
ref <- buildVptree(X)
set.seed(101)
out <- buildVptree(t(X), transposed=TRUE)
expect_identical(ref, out)
# Testing use in a function.
ref <- findVptree(X, k=5)
out <- findVptree(t(X), k=5, transposed=TRUE)
expect_identical(ref, out)
})
set.seed(200003)
test_that("buildVptree() works with the Manhattan distance", {
nobs <- 1011
ndim <- 10
X <- matrix(runif(nobs * ndim), nrow=nobs)
set.seed(102)
ref <- buildVptree(X, distance="Manhattan")
expect_identical(bndistance(ref), "Manhattan")
res <- findVptree(precomputed=ref, k=5)
val <- findVptree(X, k=5, distance="Manhattan")
expect_identical(res, val)
})
set.seed(200004)
test_that("buildVptree() works with the Cosine distance", {
nobs <- 1011
ndim <- 10
X <- matrix(runif(nobs * ndim), nrow=nobs)
set.seed(102)
ref <- buildVptree(X, distance="Cosine")
expect_identical(bndistance(ref), "Cosine")
res <- findVptree(precomputed=ref, k=5)
val <- findVptree(X, k=5, distance="Cosine")
expect_identical(res, val)
})
set.seed(20001)
test_that("buildVptree() behaves sensibly with silly inputs", {
nobs <- 100L
ndim <- 10L
X <- matrix(runif(nobs * ndim), nrow=nobs)
# What happens when there are no cells.
out <- buildVptree(X[0,,drop=FALSE])
expect_identical(dim(bndata(out)), c(ndim, 0L))
expect_identical(length(bnorder(out)), 0L)
expect_identical(lengths(VptreeIndex_nodes(out)), integer(4))
# What happens when there are no dimensions.
out <- buildVptree(X[,0,drop=FALSE])
expect_identical(dim(bndata(out)), c(0L, nobs))
expect_identical(sort(bnorder(out)), seq_len(nobs)) # arbitrary ordering.
expect_identical(lengths(VptreeIndex_nodes(out)), rep(nobs, 4))
# Checking that it behaves without distinct data points.
expect_error(prec <- buildVptree(matrix(0, 10,10)), NA)
# We get the same result when 'X' is not, strictly, a matrix.
set.seed(1999)
ref <- buildVptree(X)
set.seed(1999)
Y <- data.frame(X, check.names=FALSE, fix.empty.names=FALSE)
colnames(Y) <- NULL
out <- buildVptree(Y)
dimnames(out@data) <- NULL # fix not-quite-empty dimnames.
expect_equal(ref, out)
})
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.