def train(self, fpath):
"""train pinyin model"""
for fname in self.filelist(fpath):
with io.open(fname, 'r', encoding='utf-8') as f:
for line in f:
line = safe_input(line)
words = line.split()
self.trie.add(words[0], words[1:])
def load_data(self, fpath):
"""load data from file"""
for fname in filelist(fpath):
with io.open(fname, 'r', encoding='utf-8') as f:
for line in f:
line = line.strip()
arr = line.split()
if arr:
yield safe_input(arr[0])
def load_data(self, fpath):
datas = []
for fname in self.filelist(fpath):
with io.open(fname, 'r', encoding='utf-8') as f:
for line in f:
line = safe_input(line)
if len(line) == 0:
continue
self.line_total += 1
datas.append(list(map(lambda x: x.split('/'), line.split())))
return datas
def load_dict(self, fpath, authfreq=False, defaultfreq=5):
counter = {}
total = 0
word_tag = {}
for fname in self.filelist(fpath):
with io.open(fname, 'r', encoding='utf-8') as f:
for idx, line in enumerate(f, 1):
try:
line = safe_input(line)
arr = line.split()
if len(arr) >= 3:
word, freq, tag = arr[:3]
elif len(arr) == 2:
word, x = arr[:2]
if x.isdigit():
freq = x
tag = 'un'
else:
if authfreq:
freq = self.get_freq(word)
else:
freq = defaultfreq
tag = x
elif len(arr) == 1:
word = arr[0]
if authfreq:
freq = self.get_freq(word)
else:
freq = defaultfreq
tag = 'un'
else:
continue
freq = int(freq)
counter[word] = freq
total += freq
word_tag[word] = tag
for ch in range(len(word)):
wfrag = word[:ch + 1]
if wfrag not in counter:
counter[wfrag] = 0
except ValueError:
raise ValueError('''invalid dictionary entry
in %s at Line %s: %s''' %
(fname, idx, line))
return counter, total, word_tag
def load_data(self, posfname, negfname):
"""load dataset from file"""
def get_file(path):
if os.path.isdir(path):
for root, dirs, files in os.walk(path):
if not dirs:
for f in files:
yield os.sep.join([root, f])
else:
yield path
pos_docs, neg_docs = [], []
for fname in get_file(posfname):
with io.open(fname, 'r', encoding='utf-8') as f:
for line in f:
line = safe_input(line)
pos_docs.append(seg(line))
for fname in get_file(negfname):
with io.open(fname, 'r', encoding='utf-8') as f:
for line in f:
line = safe_input(line)
neg_docs.append(seg(line))
return pos_docs, neg_docs
