Carlos-Francisco Méndez-Cruz / nlp-preprocessing-pipeline

Go to a project
Toggle navigation Toggle navigation pinning
Carlos-Francisco Méndez-Cruz
611eb03b 1 parent 9321869a

New terminological tagging for CRFs

Inline Side-by-side
Showing 1 changed file with 2 additions and 5 deletions
...@@ -154,7 +154,7 @@ if __name__ == "__main__": ...@@ -154,7 +154,7 @@ if __name__ == "__main__":
154 line = '' 154 line = ''
155 for w, l in zip(word.split('-'), lemma.split('-')): 155 for w, l in zip(word.split('-'), lemma.split('-')):
156 line += w + '\t' + listLine1[1] + '\t' + l + ' ' + termTag + ' TermTag' + '\n' 156 line += w + '\t' + listLine1[1] + '\t' + l + ' ' + termTag + ' TermTag' + '\n'
157 - line.rstrip('\r\n') 157 + line = line.rstrip('\n')
158 print("Line: {}".format(line)) 158 print("Line: {}".format(line))
159 if not found: 159 if not found:
160 line = listLine1[0] + '\t' + listLine1[1] + '\t' + listLine2[0] + ' ' + termTag + ' TermTag' 160 line = listLine1[0] + '\t' + listLine1[1] + '\t' + listLine2[0] + ' ' + termTag + ' TermTag'
...@@ -164,10 +164,7 @@ if __name__ == "__main__": ...@@ -164,10 +164,7 @@ if __name__ == "__main__":
164 line = listLine1[0] + '\t' + listLine1[1] + '\t' + listLine2[0] + ' ' + 'O' + ' TermTag' 164 line = listLine1[0] + '\t' + listLine1[1] + '\t' + listLine2[0] + ' ' + 'O' + ' TermTag'
165 # line = listLine1[0] + '\t' + termTag + '\t' + listLine2[0] + ' ' + termTag + ' TermTag' 165 # line = listLine1[0] + '\t' + termTag + '\t' + listLine2[0] + ' ' + termTag + ' TermTag'
166 #line = word + '\t' + termTag + '\t' + lemma + ' ' + termTag + ' TermTag' 166 #line = word + '\t' + termTag + '\t' + lemma + ' ' + termTag + ' TermTag'
167 - if found: 167 + oFile.write(line + '\n')
168 - oFile.write(line)
169 - else:
170 - oFile.write(line + '\n')
171 filesPreprocessed += 1 168 filesPreprocessed += 1
172 169
173 # Imprime archivos procesados 170 # Imprime archivos procesados
......