R/vw.R

In RVowpalWabbit: R Interface to the Vowpal Wabbit

## # Test 1:
## # {VW} -b 17 -l 20 --initial_t 128000 --power_t 1 -d train-sets/0001.dat -f models/0001.model -c --passes 2 --compressed --ngram 3 --skips 1
## test1 <- c("-b", "17",
##            "-l", "20",
##            "--initial_t", "128000",
##            "--power_t", "1",
##            "-d", "/tmp/vw/test/train-sets/0001.dat",
##            "-f", "/tmp/vw/test/models/0001.model",
##            "-c",
##            "--passes", "2",
##            "--compressed",
##            "--ngram", "3",
##            "--skips", "1")

## # Test 2: checking predictions as well
## # {VW} -t train-sets/0001.dat -i models/0001.model -p 001.predict.tmp
## test2 <- c("-d", "/tmp/vw/test/train-sets/0001.dat",
##            "-i", "/tmp/vw/test/models/0001.model",
##            "-p", "/tmp/vw/0001.predict.tmp")

## # Test 3: without -d, training only
## # {VW} train-sets/0002.dat    -f models/0002.model
## test3 <- c("-t", "/tmp/vw/test/train-sets/0002.dat",
##            "-f", "/tmp/vw/test/models/0002.model")

## # Test 4: same, with -d
## # {VW} -d train-sets/0002.dat    -f models/0002.model
## test4 <- c("-d", "/tmp/vw/test/train-sets/0002.dat",
##            "-f", "/tmp/vw/test/models/0002.model")

## # Test 5: add -q .., adaptive, and more (same input, different outputs)
## # {VW} --initial_t 1 --power_t 0.5 --adaptive -q Tf -q ff -f models/0002a.model train-sets/0002.dat
## test5 <- c("--initial_t", "1",
##            "--power_t", "0.5",
##            "--adaptive",
##            "-q", "Tf",
##            "-q", "ff",
##            "-f", "/tmp/vw/test/models/0002a.model",
##            "/tmp/vw/test/train-sets/0002.dat")

## # Test 6: run predictions on Test 4 model
## # Pretending the labels aren't there
## # {VW} -t -i models/0002.model -d train-sets/0002.dat -p 0002b.predict
## test6 <- c("-t", "-i", "/tmp/vw/test/models/0002.model",
##            "-d", "/tmp/vw/test/train-sets/0002.dat",
##            "-p", "/tmp/vw/test/0002b.predict")

## # Test 7: using -q and multiple threads
## # {VW} --adaptive -q ff -f models/0002c.model train-sets/0002.dat
## test7 <- c("--adaptive", "-q", "ff",
##            "-f", "/tmp/vw/test/models/0002c.model",
##            "/tmp/vw/test/train-sets/0002.dat")


## ## Test 8: predicts on test 7 model
## ## {VW} -t -i models/0002c.model -d train-sets/0002.dat -p 0002c.predict
## test8 <- c("-t",
##            "-i", "/tmp/vw/test/models/0002c.model",
##            "-d", "/tmp/vw/test/train-sets/0002.dat",
##            "-p", "/tmp/vw/test/0002c.predict")

.test3 <- function() {
    c("-t", system.file(package="RVowpalWabbit", "test/train-sets/0002.dat"),
      "-f", system.file(package="RVowpalWabbit", "test/models/0002.model"))
}

vw <- function(args=.test3(), quiet=TRUE) {
    args <- c("Rvw", args)              # provide an argv[0] for program_options
    if (quiet) args <- c(args, "--quiet")
    Rvw(args)
}