def __init__(self, directory):
'''
directory: dataset directory as string
stemmer: Potter's Stemmer for stemming and tokenizing from nltk
tf_vectorizer: object to vectorize datas which appears at least 15 documents for featuring
'''
self.directory = directory
self.stemmer = Stemmer()
def test_stemming(test_file, lexicon_file):
stemmer = Stemmer(lexicon_file)
with open(test_file) as f:
for test in yaml.load(f):
lemma = strip_length(test.pop("lemma"))
test_length = test.pop("test_length", True)
location = test.pop("location", "")
for parse, form in test.items():
stemmer.stem(location, lemma, parse, form, test_length)
stemmer.counter.results()
return remove_stopwords(remove_ponctuation(str(s))).upper()
def toString(sentence):
out = ''
if str(sentence) != 'nan':
for word in sentence.split():
if isinstance(word, basestring):
out += (" " + word)
# else:
# out += (" " + str(word))
return out
stemmer = Stemmer()
def DistJaccard(str1, str2):
if str1 != '' and str2 != '':
str1 = set(str1.split())
str2 = set(str2.split())
return 1.0 - float(len(str1 & str2)) / len(str1 | str2)
else:
return numpy.nan
#--------------------------------------------------------------------------------#
# Get products infos from GoldStandard #
#--------------------------------------------------------------------------------#
#!/usr/bin/env python3
from pysblgnt import morphgnt_rows
from stemming import Stemmer
IGNORE_LIST = [
"σαβαχθάνι",
"ἔνι",
"χρή",
]
stemmer = Stemmer("lexicons/morphgnt.yaml")
for book_num in range(1, 28):
for row in morphgnt_rows(book_num):
ccat_pos = row["ccat-pos"]
ccat_parse = row["ccat-parse"]
norm = row["norm"]
lemma = row["lemma"]
if ccat_pos != "V-":
continue
if lemma in IGNORE_LIST:
continue
if ccat_parse[3] == "N":
parse = ccat_parse[1:4]
