Project

libsvm_preprocessor

0.0
No commit activity in last 3 years
No release in over 3 years
libsvm_preprocessornodo/libsvm_preprocessorHomepageDocumentationSource CodeBug TrackerWiki
It's a text preprocessor that generate a libsvm input file
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
 Popularity
Downloads
6,522
Stars
1
Forks
0
Watchers
3
 Releases
Current version
0.2
Total releases
2
First release
2013-05-31
Latest release
2013-09-13
 Development
Primary Language
Ruby
Licenses
MIT
Average date of last 50 commits
2013-05-20
Reverse Dependencies
0

 Dependencies

Runtime

ruby-stemmer
~> 0.9.3
stopwords-filter
~> 0.2.1
 Project Readme

Libsvm_preprocessor

This project is a simple ruby gem that provide a way to transform a text into sparse features vector using libsvm/liblinear format (http://www.csie.ntu.edu.tw/~cjlin/libsvm).

Since this tool is thought to be used with short-text it provides only binary representations of tokens.

Usage

	% libsvm_pp --help
	libsvm_pp [options] <filename>
  	 -m, --mode [TYPE]                Select unigram (default)/bigram/trigrams
     -s, --stemming                   Use this you want stemming
     -w, --remove-stopwords           Use this if you want remove stopwords
     -t, --testing                    Use this to use testing mode
     -l, --language [TYPE]            Select your language it / en
     -n N                             Numeric type
	 -o [output]                      output file

It is possible to use the library following these steps:

require "libsvm_preprocessor/preprocesso"

[…]

preprocessor = Preprocessor.new(numeric_type: i)
preprocessor.use("TRAIN.csv", "TRAIN.svm")
preprocessor.use("TEST.csv", "TEST.svm", testing: true)

In this case TRAIN.svm will contain your training set and TEST.svm will contain the testing set.

This project is far to be complete, as soon as possible I will provide a better documentation.