def get_ptb_data(w2id):
all_words = []
for item in ptb.fileids():
all_words.extend(list(map(str.lower, ptb.words(item))))
# print(all_words)
all_words_id = []
for w in all_words:
id = w2id.get(w)
if id == None:
id = w2id.get("<unk>")
all_words_id.append(id)
return all_words_id
def parse_wsj():
from nltk.corpus import ptb
import logging
logger = logging.getLogger('root')
all_lines = []
for wsj_folder in range(0, 25):
for file_number in range(0, 100):
wsj_file = 'wsj/%02d/wsj_%02d%02d.mrg' % (wsj_folder, wsj_folder, file_number)
logger.info('opening %s' % wsj_file)
try:
line = [word.lower() for word in ptb.words(wsj_file)]
all_lines.append(line)
except IOError:
logger.info('not found')
return all_lines
def test_category_words(self):
self.assertEqual(
ptb.words(categories=['humor', 'fiction'])[:6],
['Thirty-three', 'Scotty', 'did', 'not', 'go', 'back'],
)
def test_words(self):
self.assertEqual(
ptb.words('WSJ/00/WSJ_0003.MRG')[:7],
['A', 'form', 'of', 'asbestos', 'once', 'used', '*'],
)
def test_category_words(self):
self.assertEqual(
ptb.words(categories=['humor','fiction'])[:6],
['Thirty-three', 'Scotty', 'did', 'not', 'go', 'back']
)
def test_words(self):
self.assertEqual(
ptb.words('WSJ/00/WSJ_0003.MRG')[:7],
['A', 'form', 'of', 'asbestos', 'once', 'used', '*']
)
def test_words(self):
self.assertEqual(
ptb.words("WSJ/00/WSJ_0003.MRG")[:7],
["A", "form", "of", "asbestos", "once", "used", "*"],
)
def test_category_words(self):
self.assertEqual(
ptb.words(categories=["humor", "fiction"])[:6],
["Thirty-three", "Scotty", "did", "not", "go", "back"],
)
def main(test=False):
"""
makes a big dumb PTB CFG, and ShiftReduceParser, and a ViterbiParser, and
serializes them all to disk for future use.
The ViterbiParser runs in cubic time and give the most likely parse.
The ShiftReduceParser runs in linear time and gives a single parse.
https://stackoverflow.com/questions/7056996/how-do-i-get-a-set-of-grammar-rules-from-penn-treebank-using-python-nltk
https://groups.google.com/forum/#!topic/nltk-users/_LXtbIekLvc
https://www.nltk.org/_modules/nltk/grammar.html
"""
vocabulary = chainer.datasets.get_ptb_words_vocabulary()
freq_thresh = 0 ## ARBITRARY
word_freqs = FreqDist(ptb.words())
if not os.path.isfile('parsers/grammar.pkl'):
productions = []
add_dict = {}
# use the entire treebank's parsed sentences to generate the CFG
for i, tree in enumerate(ptb.parsed_sents()):
# is it a good idea to combine this with my preprocessing?
tree.collapse_unary(collapsePOS=False)
tree.chomsky_normal_form(horzMarkov=2)
# preprocess all productions by removing all tags
these_productions = tree.productions()
for production in these_productions:
# remove all tags from the LHS (only keep primary tag)
production._lhs = preprocess_nt(production._lhs)
rhs = []
for item in production._rhs:
# remove all tags from the Nonterminals on the RHS
if type(item) == nltk.grammar.Nonterminal:
rhs.append(preprocess_nt(item))
# replace numbers with N
elif is_number(item):
rhs.append('N')
# items not in dictionary replaced with <unk>
# dictionary requires lower
elif not is_key(vocabulary, item.lower()):
rhs.append('<unk>')
# replace infrequent words with <unk>
elif word_freqs[item] < freq_thresh:
rhs.append('<unk>')
# lowercase all entries in the grammar
else:
rhs.append(item.lower())
production._rhs = tuple(rhs)
if not is_key(add_dict, production.unicode_repr()):
add_dict[production.unicode_repr()] = True
productions.append(production)
print('** {} productions found! **'.format(len(productions)))
grammar = induce_pcfg(Nonterminal('S'), productions)
with open('parsers/grammar.pkl', 'wb') as f:
f.write(pickle.dumps(grammar))
if not os.path.isfile('parsers/viterbi_parser.pkl'):
filename = open('parsers/grammar.pkl', 'rb')
grammar = pickle.load(filename)
viterbi_parser = ViterbiParser(grammar, trace=0) # cubic time
with open('parsers/viterbi_parser.pkl', 'wb') as f:
f.write(pickle.dumps(viterbi_parser))
if not os.path.isfile('parsers/shift_reduce_parser.pkl'):
filename = open('parsers/grammar.pkl', 'rb')
grammar = pickle.load(filename)
shift_reduce_parser = ShiftReduceParser(grammar, trace=0) # linear time
with open('parsers/shift_reduce_parser.pkl', 'wb') as f:
f.write(pickle.dumps(shift_reduce_parser))
with open('data/ptb.train.txt', 'r') as f:
data = f.readlines()
if test:
for sample in [1, 23, 20330, 20332, 443]:
t1 = time.time()
viterbi_parser.parse_one(data[sample].split())
t2 = time.time()
print('viterbi = {:.2f} sec for {} words'.format(
t2-t1, len(data[sample].split())))
t1 = time.time()
shift_reduce_parser.parse_one(data[sample].split())
t2 = time.time()
print('shift reduce = {:.2f} sec for {} words'.format(
t2-t1, len(data[sample].split())))
def test_category_words(self):
self.assertEqual(
ptb.words(categories=["humor", "fiction"])[:6], ["Thirty-three", "Scotty", "did", "not", "go", "back"]
)
def test_words(self):
self.assertEqual(ptb.words("WSJ/00/WSJ_0003.MRG")[:7], ["A", "form", "of", "asbestos", "once", "used", "*"])
singleWords=set([","])
punc=set([":","#","-LRB-","CD","TO","``","S","''","$","NP","IN","-RRB-",",","."])
punc2=set([":","#","``","''","$",",","."])
i=0
# I read all files then I will go thorugh all lines in each file and I used stack to extract the grammers.
#for each ')' I pop from stack until I face '(' so I generate rule and push back the last item in the stack for a next rule.
for o in os.listdir('wsj'):
if (os.path.isdir(os.path.join('wsj',o))):
for filename in os.listdir(os.path.join('wsj',o)):
i=i+1
print i
if filename.endswith(".mrg"):
path='wsj/'+o+'/'+filename
singleWords=set(ptb.words(path))| singleWords
pranthesis=Stack()
words=Stack()
f=open(path,'r')
content=f.readlines()
allWords=[]
for line in content:
for l in line.strip().split():
# with regular expression we break each line to: words '(' ')'
result=re.findall("[)]|[(]|[^\)\(]*",l.strip())
for r in result:
if r !="":
allWords.append(r)
buffer=[]
for word in allWords:
print(docs[0].docno)
print(docs[0].doctype)
print(docs[0].date_time)
print(docs[0].headline)
print(docs[0].text) # doctest: +ELLIPSIS
# parsed corpora
print(treebank.fileids()) # doctest: +ELLIPSIS
print(treebank.words('wsj_0003.mrg'))
print(treebank.tagged_words('wsj_0003.mrg'))
print(treebank.parsed_sents('wsj_0003.mrg')
[0]) # doctest: +ELLIPSIS +NORMALIZE_WHITESPACE
# nltk.download('ptb')
print(ptb.fileids()) # doctest: +SKIP
# download the corpus from here: https://raw.githubusercontent.com/nltk/nltk_data/gh-pages/packages/corpora/treebank.zip
# then extract and place to the following location: .../nltk_data/corpora/ptb/
print(ptb.words('treebank/combined/wsj_0003.mrg')) # doctest: +SKIP
print(ptb.tagged_words('treebank/combined/wsj_0003.mrg')) # doctest: +SKIP
# print(ptb.categories()) # doctest: +SKIP
# print(ptb.fileids('news')) # doctest: +SKIP
# print(ptb.words(categories=['humor', 'fiction'])) # doctest: +SKIP
# nltk.download('sinica_treebank')
print(sinica_treebank.sents()) # doctest: +SKIP
print(sinica_treebank.parsed_sents()[25]) # doctest: +SKIP
# nltk.download('conll2007')
print(conll2007.sents('esp.train')[0]) # doctest: +SKIP
print(conll2007.parsed_sents('esp.train')[0]) # doctest: +SKIP
print(conll2007.parsed_sents('esp.train')[0].tree()) # doctest: +SKIP
# for tree in ycoe.parsed_sents('cocuraC')[:4]:
# print(tree) # doctest: +SKIP
# word lists and lexicons
print(words.fileids())
