Permutation Statter

Purpose

Takes a (set of) permutations, reads it in with a normalisation-reader-lib and calculates the p-value using a measure-lib. Measure-libs can support some ofsix things: One p-value (a compound-value is used for calculating the p-value of the whole), One weight (report the compound-value as it comes out of the measure-lib), One rawnrm (the raw normalisation summed, as it comes out of the data-reader (the measure doesn't touch it)), Partsweights (the weights all parts have in this p-value), Parts-p-values (the separate values are used for calculating the part-p-values), Parts rawnrms (the raw normalisation un-summed)

The -lm, and -ll options are used for listing the measure-, and normalisation- libs that are installed and available (they come with a small description)

The configfile is configpermutationstatter.pl

The following normalisation-readers are available:

The following measure-libraries are available:

Synopsis

./permutationstatter.pl

[-c <corpus>] [-f <fromsubdir>] | [-fd <fromdir>] [-pn <permutation-test-name>] [-tc <tocorpus>] [-t <tosubdir>] | [-td <targetdir>] [-l <readerlibshortname>] | [-lf <readerlibfullname>] [-ll] [-la <extrareaderlibargs>] [-ir <ipnormirnr>] [-e <totempdir>] | [-ed <tempdir>] [-m <permutationmeasure>] | [-ml <permutationmeasurelib>] [-lm] [-ma <extrameasurelibargs>] [-dr] [-? = -h = -help = --help] [-v [verboselvl]]

-c: the corpus of which a permutation should be used

-f: the subdir below corpus in which the permutations can be found

-fd: full path to the dir in which the permutation is stored. The permutation- dir given with -pn should be inside this directory (this option overrides the -c and -f options)

-pn: the name of the permutation-test-output that should be used

-tc: target-corpus. If none is set the source corpus is used

-t: the subdir in which the statresults are stored

-td: full path to the dir in which the statresults should be stored (inside a) sub-dir named <readerlibshortname>."-".<permutationmeasure>) (overrides the -t and -tc options)

-l: the name of the installed library that reads the permutation-data (and that possibly normalizes it along the way). If none is specified the default one will be used. The default one does not normalize, it just reads the raw data (the normalizer libs are subclasses of the default one, using it's reading-functionality and apply normalisation before handing over the data)

-lf: the name of the non-installed permutation-data-reader (conflicts with -ns)

-ll: lists all the normalisation-libs and their descriptions (exits the program immidiately)

-la: extra args to hand over to the reader-lib

-e: the subdir in which the temp-data can be stored. Tempdata is the normalized permutation-data, that can be stored by some normalisation-libs (ignorm & ignormtbpnorm at this moment) to prevent having to do the same normalisat- ions all the time over again when running different measures on them

-ed: full path to the dir in which the temp-data can be stored

-ir: the number of iterations that the ipnorm normalization should run to increase precision

-m: the name of the installed permutationmeasure to use (-lm lists them all). A permutationmeasure is installed if it's present in the lib-dir and if it's set correctly in the measureshash in the configpermutationstatter.pl file

-mf: the (non-installed, but present in the libdir) library that provides the measure applied to the data (conflicts with ms)

-lm: lists all the installed measures and their descriptions (exits the program immidiately)

-ma: extra arguments to the (installed or not) difference-measuring library

-dr: dry-run. Nothing is written or deleted, only reading and reporting is done

-v: the level of verbosity, default verboselevel = 2, available levels: 0,1,2,3

-?: (and equivalents) prints help: the purpose and the synopsys

NOTE:

If no from-dir, division-dir and/or to-file is given the default(s) in the config file are/is used

You can download (or look at the sources of) PermutationStatter [here]. To run it you will also need [the config file] and the [fiauimenre library]. You can also get the entire tool-package (containing the newest version of all fiauimenre tools and the library) [in one download]