tests/testthat/test-madlib_lda.r

In pivotalsoftware/PivotalR: A Fast, Easy-to-Use Tool for Manipulating Tables in Databases and a Wrapper of MADlib

context("Test cases for madlib.lda and its helper functions")

## ------------------------------------------------------------
## Test preparations

env <- new.env(parent = globalenv())
.dbname = get('pivotalr_dbname', envir=env)
.port = get('pivotalr_port', envir=env)

cid <- db.connect(port = pivotalr_port, dbname = pivotalr_dbname, verbose = FALSE)
db <- .get.dbms.str(cid)

library(topicmodels)

data("AssociatedPress", package = "topicmodels")
temp_dat <- AssociatedPress
dat1 <- cbind(temp_dat$i,temp_dat$j,temp_dat$v)
dat1 <- as.data.frame(dat1)
colnames(dat1) <- c("docid", "wordid", "count")
dat2 <- as.data.frame(temp_dat$dimnames[2]$Terms)
dat2 <- cbind(1:nrow(dat2),dat2)
colnames(dat2) <- c("wordid", "word")

termfreq <- .unique.string()
vocab <- .unique.string()
newdata <- .unique.string()

dat1 <- as.db.data.frame(dat1, conn.id=cid, verbose=FALSE, is.temp=TRUE, table.name=termfreq)
dat2 <- as.db.data.frame(dat2, conn.id=cid, verbose=FALSE, is.temp=TRUE, table.name=vocab)

sql <- paste("DROP TABLE IF EXISTS ", newdata, "; CREATE TEMP TABLE ", newdata,
             " AS SELECT docid, array_agg(word) AS words ",
             "FROM (SELECT *, generate_series(1,count::int) FROM ",
             termfreq,  " JOIN ", vocab,
             " USING (wordid)) subq GROUP BY docid;", sep="")

db.q(sql, verbose=FALSE)

dat.r <- AssociatedPress

dat <- db.data.frame(newdata, conn.id=cid, verbose=FALSE, is.temp=TRUE)

test_that("Test perplexity", {
    testthat::skip_on_cran()
	output.db <- madlib.lda(dat, 10,0.1,0.1, 50)
	output.r <- LDA(AssociatedPress, k=10,
		control=list(iter=50, alpha=0.1, delta=0.1),method="Gibbs")

	perplexity.db <- perplexity.lda.madlib(output.db)
	perplexity.r <- perplexity(output.r, newdata=AssociatedPress)


	expect_equal(perplexity.db, perplexity.r, tolerance=1e-1,
		check.attributes=FALSE)

})

test_that("Test perplexity multiple starts with best output", {
    testthat::skip_on_cran()
	output.db <- madlib.lda(dat, 10,0.1,0.1, 50, nstart=3)
	output.r <- LDA(AssociatedPress, k=10,
		control=list(iter=50, alpha=0.1, delta=0.1, nstart=3,
		seed=c(NA,NA,NA)),method="Gibbs")

	perplexity.db <- perplexity.lda.madlib(output.db)
	perplexity.r <- perplexity(output.r, newdata=AssociatedPress,
		control=list(seed=NA))

	expect_equal(perplexity.db, perplexity.r, tolerance=1e-1,
		check.attributes=FALSE)

})

test_that("Test perplexity multiple starts with multiple output", {
    testthat::skip_on_cran()
	output.db <- madlib.lda(dat, 10,0.1,0.1, 50, nstart=3, best=FALSE)
	output.r <- LDA(AssociatedPress, k=10,
		control=list(iter=50, alpha=0.1, delta=0.1, nstart=3, best=FALSE,
		seed=c(NA,NA,NA)),method="Gibbs")

	perplexity.db <- perplexity.lda.madlib(output.db[[2]])
	perplexity.r <- perplexity(output.r[[2]], newdata=AssociatedPress,
		control=list(seed=c(NA)))


	expect_equal(perplexity.db, perplexity.r, tolerance=1e-1,
		check.attributes=FALSE)

})