You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

88 lines
2.8 KiB

import java.util.ArrayList;
import java.util.List;
import data.Sentence;
import data.Word;
public class Common {
public static List<Sentence> corpus;
public static List<Sentence> minCorpus;
public static List<Sentence> midCorpus;
public static List<Sentence> midCorpusSkip;
public static List<Sentence> josTest;
static {
Sentence testSentence;
// full sentence
ArrayList<String> taxonomy = new ArrayList<>();
taxonomy.add("#Ft.Z.N.N");
List<Word> words = new ArrayList<>();
words.add(new Word("ker", "ker", "Vd"));
words.add(new Word("ima", "imeti", "Ggnste-n"));
words.add(new Word("junak", "junak", "Somei"));
words.add(new Word("v", "v", "Dm"));
words.add(new Word("posesti", "posest", "Sozem"));
words.add(new Word("nekaj", "nekaj", "Rsn"));
words.add(new Word("o", "o", "Dm"));
words.add(new Word("čemer", "kar", "Zz-sem"));
words.add(new Word("se", "se", "Zp------k"));
words.add(new Word("mu", "on", "Zotmed--k"));
words.add(new Word("ne", "ne", "L"));
words.add(new Word("sanja", "sanjati", "Ggnste"));
words.add(new Word("a", "a", "Vp"));
words.add(new Word("se", "se", "Zp------k"));
words.add(new Word("onemu", "oni", "Zk-sed"));
words.add(new Word("zdi", "zdeti", "Ggnste"));
words.add(new Word("ključno", "ključen", "Ppnsei"));
words.add(new Word("pri", "pri", "Dm"));
words.add(new Word("operaciji", "operacija", "Sozem"));
words.add(new Word("666", "666", "Kag"));
testSentence = new Sentence(words, taxonomy);
corpus = new ArrayList<>();
corpus.add(testSentence);
// three word sentence
testSentence = new Sentence(corpus.get(0).getSublist(0, 3), taxonomy);
minCorpus = new ArrayList<>();
minCorpus.add(testSentence);
// five word sentence
words = new ArrayList<>();
words.add(new Word("ker", "ker", "Vd"));
words.add(new Word("ima", "imeti", "Ggnste-n"));
words.add(new Word("junak", "junak", "Somei"));
words.add(new Word("ima", "imeti", "Ggnste-n"));
words.add(new Word("posesti", "posest", "Sozem"));
testSentence = new Sentence(words, taxonomy);
midCorpus = new ArrayList<>();
midCorpus.add(testSentence);
// five word sentence - for skipgrams
words = new ArrayList<>();
words.add(new Word("ker", "ker", "Vd"));
words.add(new Word("ima", "imeti", "Ggnste-n"));
words.add(new Word("junak", "junak", "Somei"));
words.add(new Word("v", "v", "Dm"));
words.add(new Word("posesti", "posest", "Sozem"));
testSentence = new Sentence(words, taxonomy);
midCorpusSkip = new ArrayList<>();
midCorpusSkip.add(testSentence);
// JOS test
words = new ArrayList<>();
words.add(new Word("junak", "junak", "Somei"));
words.add(new Word("ima", "imeti", "Ggnste-n"));
words.add(new Word("posesti", "posest", "Sozem"));
testSentence = new Sentence(words, taxonomy);
josTest = new ArrayList<>();
josTest.add(testSentence);
}
}