Copy disabled (too large)
Download .txt
Showing preview only (30,274K chars total). Download the full file to get everything.
Repository: CCChenhao997/DualGCN-ABSA
Branch: main
Commit: e8975264b080
Files: 133
Total size: 62.8 MB
Directory structure:
gitextract_bclkhkuw/
├── DualGCN/
│ ├── build_vocab.sh
│ ├── data_utils.py
│ ├── dataset/
│ │ ├── Laptops_allennlp/
│ │ │ ├── test.json
│ │ │ └── train.json
│ │ ├── Laptops_corenlp/
│ │ │ ├── test.json
│ │ │ └── train.json
│ │ ├── Laptops_stanza/
│ │ │ ├── test.json
│ │ │ └── train.json
│ │ ├── Restaurants_allennlp/
│ │ │ ├── test.json
│ │ │ └── train.json
│ │ ├── Restaurants_corenlp/
│ │ │ ├── test.json
│ │ │ └── train.json
│ │ ├── Restaurants_stanza/
│ │ │ ├── test.json
│ │ │ └── train.json
│ │ ├── Tweets_allennlp/
│ │ │ ├── test.json
│ │ │ └── train.json
│ │ ├── Tweets_corenlp/
│ │ │ ├── test.json
│ │ │ └── train.json
│ │ └── Tweets_stanza/
│ │ ├── test.json
│ │ └── train.json
│ ├── glove/
│ │ └── readme.md
│ ├── layers.py
│ ├── models/
│ │ ├── atae_lstm.py
│ │ ├── dualgcn.py
│ │ ├── dualgcn_bert.py
│ │ ├── ian.py
│ │ ├── semgcn.py
│ │ └── syngcn.py
│ ├── prepare_vocab.py
│ ├── run.sh
│ ├── train.py
│ └── tree.py
├── LAL-Parser/
│ ├── EVALB/
│ │ ├── COLLINS.prm
│ │ ├── LICENSE
│ │ ├── Makefile
│ │ ├── README
│ │ ├── bug/
│ │ │ ├── bug.gld
│ │ │ ├── bug.rsl-new
│ │ │ ├── bug.rsl-old
│ │ │ └── bug.tst
│ │ ├── evalb
│ │ ├── evalb.c
│ │ ├── evalb.dSYM/
│ │ │ └── Contents/
│ │ │ ├── Info.plist
│ │ │ └── Resources/
│ │ │ └── DWARF/
│ │ │ └── evalb
│ │ ├── new.prm
│ │ ├── nk.prm
│ │ ├── sample/
│ │ │ ├── sample.gld
│ │ │ ├── sample.prm
│ │ │ ├── sample.rsl
│ │ │ └── sample.tst
│ │ └── tgrep_proc.prl
│ ├── best_model/
│ │ └── readme.md
│ └── src_joint/
│ ├── KM_parser.py
│ ├── absa_parser.py
│ ├── const_decoder.pyx
│ ├── dep_eval.py
│ ├── dep_reader.py
│ ├── evaluate.py
│ ├── hpsg_decoder.pyx
│ ├── main.py
│ ├── makehp.py
│ ├── pretrained_bert/
│ │ ├── __init__.py
│ │ ├── __main__.py
│ │ ├── convert_tf_checkpoint_to_pytorch.py
│ │ ├── file_utils.py
│ │ ├── modeling.py
│ │ ├── optimization.py
│ │ └── tokenization.py
│ ├── pytorch_transformers/
│ │ ├── __init__.py
│ │ ├── __main__.py
│ │ ├── convert_gpt2_checkpoint_to_pytorch.py
│ │ ├── convert_openai_checkpoint_to_pytorch.py
│ │ ├── convert_pytorch_checkpoint_to_tf.py
│ │ ├── convert_roberta_checkpoint_to_pytorch.py
│ │ ├── convert_tf_checkpoint_to_pytorch.py
│ │ ├── convert_transfo_xl_checkpoint_to_pytorch.py
│ │ ├── convert_xlm_checkpoint_to_pytorch.py
│ │ ├── convert_xlnet_checkpoint_to_pytorch.py
│ │ ├── file_utils.py
│ │ ├── modeling_auto.py
│ │ ├── modeling_bert.py
│ │ ├── modeling_gpt2.py
│ │ ├── modeling_openai.py
│ │ ├── modeling_roberta.py
│ │ ├── modeling_transfo_xl.py
│ │ ├── modeling_transfo_xl_utilities.py
│ │ ├── modeling_utils.py
│ │ ├── modeling_xlm.py
│ │ ├── modeling_xlnet.py
│ │ ├── optimization.py
│ │ ├── tests/
│ │ │ ├── __init__.py
│ │ │ ├── conftest.py
│ │ │ ├── fixtures/
│ │ │ │ ├── input.txt
│ │ │ │ ├── sample_text.txt
│ │ │ │ └── test_sentencepiece.model
│ │ │ ├── modeling_auto_test.py
│ │ │ ├── modeling_bert_test.py
│ │ │ ├── modeling_common_test.py
│ │ │ ├── modeling_gpt2_test.py
│ │ │ ├── modeling_openai_test.py
│ │ │ ├── modeling_roberta_test.py
│ │ │ ├── modeling_transfo_xl_test.py
│ │ │ ├── modeling_xlm_test.py
│ │ │ ├── modeling_xlnet_test.py
│ │ │ ├── optimization_test.py
│ │ │ ├── tokenization_auto_test.py
│ │ │ ├── tokenization_bert_test.py
│ │ │ ├── tokenization_gpt2_test.py
│ │ │ ├── tokenization_openai_test.py
│ │ │ ├── tokenization_roberta_test.py
│ │ │ ├── tokenization_tests_commons.py
│ │ │ ├── tokenization_transfo_xl_test.py
│ │ │ ├── tokenization_utils_test.py
│ │ │ ├── tokenization_xlm_test.py
│ │ │ └── tokenization_xlnet_test.py
│ │ ├── tokenization_auto.py
│ │ ├── tokenization_bert.py
│ │ ├── tokenization_gpt2.py
│ │ ├── tokenization_openai.py
│ │ ├── tokenization_roberta.py
│ │ ├── tokenization_transfo_xl.py
│ │ ├── tokenization_utils.py
│ │ ├── tokenization_xlm.py
│ │ └── tokenization_xlnet.py
│ ├── test.py
│ ├── transliterate.py
│ ├── trees.py
│ ├── utils.py
│ ├── utils_io.py
│ └── vocabulary.py
├── LICENSE
├── README.md
└── requirements.txt
================================================
FILE CONTENTS
================================================
================================================
FILE: DualGCN/build_vocab.sh
================================================
#!/bin/bash
# build vocab for different datasets
python ./DualGCN/prepare_vocab.py --data_dir DualGCN/dataset/Restaurants_corenlp --vocab_dir DualGCN/dataset/Restaurants_corenlp
python ./DualGCN/prepare_vocab.py --data_dir DualGCN/dataset/Laptops_corenlp --vocab_dir DualGCN/dataset/Laptops_corenlp
python ./DualGCN/prepare_vocab.py --data_dir DualGCN/dataset/Tweets_corenlp --vocab_dir DualGCN/dataset/Tweets_corenlp
python ./DualGCN/prepare_vocab.py --data_dir DualGCN/dataset/Restaurants_allennlp --vocab_dir DualGCN/dataset/Restaurants_allennlp
python ./DualGCN/prepare_vocab.py --data_dir DualGCN/dataset/Laptops_allennlp --vocab_dir DualGCN/dataset/Laptops_allennlp
python ./DualGCN/prepare_vocab.py --data_dir DualGCN/dataset/Tweets_allennlp --vocab_dir DualGCN/dataset/Tweets_allennlp
python ./DualGCN/prepare_vocab.py --data_dir DualGCN/dataset/Restaurants_stanza --vocab_dir DualGCN/dataset/Restaurants_stanza
python ./DualGCN/prepare_vocab.py --data_dir DualGCN/dataset/Laptops_stanza --vocab_dir DualGCN/dataset/Laptops_stanza
python ./DualGCN/prepare_vocab.py --data_dir DualGCN/dataset/Tweets_stanza --vocab_dir DualGCN/dataset/Tweets_stanza
================================================
FILE: DualGCN/data_utils.py
================================================
'''
Description:
version:
Author: chenhao
Date: 2021-06-09 14:17:37
'''
import os
import sys
sys.path.append(r'./LAL-Parser/src_joint')
import re
import json
import pickle
import numpy as np
from tqdm import tqdm
from transformers import BertTokenizer
from torch.utils.data import Dataset
def ParseData(data_path):
with open(data_path) as infile:
all_data = []
data = json.load(infile)
for d in data:
for aspect in d['aspects']:
text_list = list(d['token'])
tok = list(d['token']) # word token
length = len(tok) # real length
# if args.lower == True:
tok = [t.lower() for t in tok]
tok = ' '.join(tok)
asp = list(aspect['term']) # aspect
asp = [a.lower() for a in asp]
asp = ' '.join(asp)
label = aspect['polarity'] # label
pos = list(d['pos']) # pos_tag
head = list(d['head']) # head
deprel = list(d['deprel']) # deprel
# position
aspect_post = [aspect['from'], aspect['to']]
post = [i-aspect['from'] for i in range(aspect['from'])] \
+[0 for _ in range(aspect['from'], aspect['to'])] \
+[i-aspect['to']+1 for i in range(aspect['to'], length)]
# aspect mask
if len(asp) == 0:
mask = [1 for _ in range(length)] # for rest16
else:
mask = [0 for _ in range(aspect['from'])] \
+[1 for _ in range(aspect['from'], aspect['to'])] \
+[0 for _ in range(aspect['to'], length)]
sample = {'text': tok, 'aspect': asp, 'pos': pos, 'post': post, 'head': head,\
'deprel': deprel, 'length': length, 'label': label, 'mask': mask, \
'aspect_post': aspect_post, 'text_list': text_list}
all_data.append(sample)
return all_data
def build_tokenizer(fnames, max_length, data_file):
parse = ParseData
if os.path.exists(data_file):
print('loading tokenizer:', data_file)
tokenizer = pickle.load(open(data_file, 'rb'))
else:
tokenizer = Tokenizer.from_files(fnames=fnames, max_length=max_length, parse=parse)
pickle.dump(tokenizer, open(data_file, 'wb'))
return tokenizer
class Vocab(object):
''' vocabulary of dataset '''
def __init__(self, vocab_list, add_pad, add_unk):
self._vocab_dict = dict()
self._reverse_vocab_dict = dict()
self._length = 0
if add_pad:
self.pad_word = '<pad>'
self.pad_id = self._length
self._length += 1
self._vocab_dict[self.pad_word] = self.pad_id
if add_unk:
self.unk_word = '<unk>'
self.unk_id = self._length
self._length += 1
self._vocab_dict[self.unk_word] = self.unk_id
for w in vocab_list:
self._vocab_dict[w] = self._length
self._length += 1
for w, i in self._vocab_dict.items():
self._reverse_vocab_dict[i] = w
def word_to_id(self, word):
if hasattr(self, 'unk_id'):
return self._vocab_dict.get(word, self.unk_id)
return self._vocab_dict[word]
def id_to_word(self, id_):
if hasattr(self, 'unk_word'):
return self._reverse_vocab_dict.get(id_, self.unk_word)
return self._reverse_vocab_dict[id_]
def has_word(self, word):
return word in self._vocab_dict
def __len__(self):
return self._length
@staticmethod
def load_vocab(vocab_path: str):
with open(vocab_path, "rb") as f:
return pickle.load(f)
def save_vocab(self, vocab_path):
with open(vocab_path, "wb") as f:
pickle.dump(self, f)
class Tokenizer(object):
''' transform text to indices '''
def __init__(self, vocab, max_length, lower, pos_char_to_int, pos_int_to_char):
self.vocab = vocab
self.max_length = max_length
self.lower = lower
self.pos_char_to_int = pos_char_to_int
self.pos_int_to_char = pos_int_to_char
@classmethod
def from_files(cls, fnames, max_length, parse, lower=True):
corpus = set()
pos_char_to_int, pos_int_to_char = {}, {}
for fname in fnames:
for obj in parse(fname):
text_raw = obj['text']
if lower:
text_raw = text_raw.lower()
corpus.update(Tokenizer.split_text(text_raw))
return cls(vocab=Vocab(corpus, add_pad=True, add_unk=True), max_length=max_length, lower=lower, pos_char_to_int=pos_char_to_int, pos_int_to_char=pos_int_to_char)
@staticmethod
def pad_sequence(sequence, pad_id, maxlen, dtype='int64', padding='post', truncating='post'):
x = (np.zeros(maxlen) + pad_id).astype(dtype)
if truncating == 'pre':
trunc = sequence[-maxlen:]
else:
trunc = sequence[:maxlen]
trunc = np.asarray(trunc, dtype=dtype)
if padding == 'post':
x[:len(trunc)] = trunc
else:
x[-len(trunc):] = trunc
return x
def text_to_sequence(self, text, reverse=False, padding='post', truncating='post'):
if self.lower:
text = text.lower()
words = Tokenizer.split_text(text)
sequence = [self.vocab.word_to_id(w) for w in words]
if len(sequence) == 0:
sequence = [0]
if reverse:
sequence.reverse()
return Tokenizer.pad_sequence(sequence, pad_id=self.vocab.pad_id, maxlen=self.max_length,
padding=padding, truncating=truncating)
@staticmethod
def split_text(text):
# for ch in ["\'s", "\'ve", "n\'t", "\'re", "\'m", "\'d", "\'ll", ",", ".", "!", "*", "/", "?", "(", ")", "\"", "-", ":"]:
# text = text.replace(ch, " "+ch+" ")
return text.strip().split()
class SentenceDataset(Dataset):
''' PyTorch standard dataset class '''
def __init__(self, fname, tokenizer, opt, vocab_help):
parse = ParseData
post_vocab, pos_vocab, dep_vocab, pol_vocab = vocab_help
data = list()
polarity_dict = {'positive':0, 'negative':1, 'neutral':2}
for obj in tqdm(parse(fname), total=len(parse(fname)), desc="Training examples"):
text = tokenizer.text_to_sequence(obj['text'])
aspect = tokenizer.text_to_sequence(obj['aspect']) # max_length=10
post = [post_vocab.stoi.get(t, post_vocab.unk_index) for t in obj['post']]
post = tokenizer.pad_sequence(post, pad_id=opt.pad_id, maxlen=opt.max_length, dtype='int64', padding='post', truncating='post')
pos = [pos_vocab.stoi.get(t, pos_vocab.unk_index) for t in obj['pos']]
pos = tokenizer.pad_sequence(pos, pad_id=opt.pad_id, maxlen=opt.max_length, dtype='int64', padding='post', truncating='post')
deprel = [dep_vocab.stoi.get(t, dep_vocab.unk_index) for t in obj['deprel']]
deprel = tokenizer.pad_sequence(deprel, pad_id=opt.pad_id, maxlen=opt.max_length, dtype='int64', padding='post', truncating='post')
mask = tokenizer.pad_sequence(obj['mask'], pad_id=opt.pad_id, maxlen=opt.max_length, dtype='int64', padding='post', truncating='post')
adj = np.ones(opt.max_length) * opt.pad_id
if opt.parseadj:
from absa_parser import headparser
# * adj
headp, syntree = headparser.parse_heads(obj['text'])
adj = softmax(headp[0])
adj = np.delete(adj, 0, axis=0)
adj = np.delete(adj, 0, axis=1)
adj -= np.diag(np.diag(adj))
if not opt.direct:
adj = adj + adj.T
adj = adj + np.eye(adj.shape[0])
adj = np.pad(adj, (0, opt.max_length - adj.shape[0]), 'constant')
if opt.parsehead:
from absa_parser import headparser
headp, syntree = headparser.parse_heads(obj['text'])
syntree2head = [[leaf.father for leaf in tree.leaves()] for tree in syntree]
head = tokenizer.pad_sequence(syntree2head[0], pad_id=opt.pad_id, maxlen=opt.max_length, dtype='int64', padding='post', truncating='post')
else:
head = tokenizer.pad_sequence(obj['head'], pad_id=opt.pad_id, maxlen=opt.max_length, dtype='int64', padding='post', truncating='post')
length = obj['length']
polarity = polarity_dict[obj['label']]
data.append({
'text': text,
'aspect': aspect,
'post': post,
'pos': pos,
'deprel': deprel,
'head': head,
'adj': adj,
'mask': mask,
'length': length,
'polarity': polarity
})
self._data = data
def __getitem__(self, index):
return self._data[index]
def __len__(self):
return len(self._data)
def _load_wordvec(data_path, embed_dim, vocab=None):
with open(data_path, 'r', encoding='utf-8', newline='\n', errors='ignore') as f:
word_vec = dict()
if embed_dim == 200:
for line in f:
tokens = line.rstrip().split()
if tokens[0] == '<pad>' or tokens[0] == '<unk>': # avoid them
continue
if vocab is None or vocab.has_word(tokens[0]):
word_vec[tokens[0]] = np.asarray(tokens[1:], dtype='float32')
elif embed_dim == 300:
for line in f:
tokens = line.rstrip().split()
if tokens[0] == '<pad>': # avoid them
continue
elif tokens[0] == '<unk>':
word_vec['<unk>'] = np.random.uniform(-0.25, 0.25, 300)
word = ''.join((tokens[:-300]))
if vocab is None or vocab.has_word(tokens[0]):
word_vec[word] = np.asarray(tokens[-300:], dtype='float32')
else:
print("embed_dim error!!!")
exit()
return word_vec
def build_embedding_matrix(vocab, embed_dim, data_file):
if os.path.exists(data_file):
print('loading embedding matrix:', data_file)
embedding_matrix = pickle.load(open(data_file, 'rb'))
else:
print('loading word vectors...')
embedding_matrix = np.zeros((len(vocab), embed_dim))
fname = './DualGCN/glove/glove.840B.300d.txt'
word_vec = _load_wordvec(fname, embed_dim, vocab)
for i in range(len(vocab)):
vec = word_vec.get(vocab.id_to_word(i))
if vec is not None:
embedding_matrix[i] = vec
pickle.dump(embedding_matrix, open(data_file, 'wb'))
return embedding_matrix
def softmax(x):
if len(x.shape) > 1:
# matrix
tmp = np.max(x, axis=1)
x -= tmp.reshape((x.shape[0], 1))
x = np.exp(x)
tmp = np.sum(x, axis=1)
x /= tmp.reshape((x.shape[0], 1))
else:
# vector
tmp = np.max(x)
x -= tmp
x = np.exp(x)
tmp = np.sum(x)
x /= tmp
return x
class Tokenizer4BertGCN:
def __init__(self, max_seq_len, pretrained_bert_name):
self.max_seq_len = max_seq_len
self.tokenizer = BertTokenizer.from_pretrained(pretrained_bert_name)
self.cls_token_id = self.tokenizer.cls_token_id
self.sep_token_id = self.tokenizer.sep_token_id
def tokenize(self, s):
return self.tokenizer.tokenize(s)
def convert_tokens_to_ids(self, tokens):
return self.tokenizer.convert_tokens_to_ids(tokens)
class ABSAGCNData(Dataset):
def __init__(self, fname, tokenizer, opt):
self.data = []
parse = ParseData
polarity_dict = {'positive':0, 'negative':1, 'neutral':2}
for obj in tqdm(parse(fname), total=len(parse(fname)), desc="Training examples"):
polarity = polarity_dict[obj['label']]
text = obj['text']
term = obj['aspect']
term_start = obj['aspect_post'][0]
term_end = obj['aspect_post'][1]
text_list = obj['text_list']
left, term, right = text_list[: term_start], text_list[term_start: term_end], text_list[term_end: ]
from absa_parser import headparser
headp, syntree = headparser.parse_heads(text)
ori_adj = softmax(headp[0])
ori_adj = np.delete(ori_adj, 0, axis=0)
ori_adj = np.delete(ori_adj, 0, axis=1)
ori_adj -= np.diag(np.diag(ori_adj))
if not opt.direct:
ori_adj = ori_adj + ori_adj.T
ori_adj = ori_adj + np.eye(ori_adj.shape[0])
assert len(text_list) == ori_adj.shape[0] == ori_adj.shape[1], '{}-{}-{}'.format(len(text_list), text_list, ori_adj.shape)
left_tokens, term_tokens, right_tokens = [], [], []
left_tok2ori_map, term_tok2ori_map, right_tok2ori_map = [], [], []
for ori_i, w in enumerate(left):
for t in tokenizer.tokenize(w):
left_tokens.append(t) # * ['expand', '##able', 'highly', 'like', '##ing']
left_tok2ori_map.append(ori_i) # * [0, 0, 1, 2, 2]
asp_start = len(left_tokens)
offset = len(left)
for ori_i, w in enumerate(term):
for t in tokenizer.tokenize(w):
term_tokens.append(t)
# term_tok2ori_map.append(ori_i)
term_tok2ori_map.append(ori_i + offset)
asp_end = asp_start + len(term_tokens)
offset += len(term)
for ori_i, w in enumerate(right):
for t in tokenizer.tokenize(w):
right_tokens.append(t)
right_tok2ori_map.append(ori_i+offset)
while len(left_tokens) + len(right_tokens) > tokenizer.max_seq_len-2*len(term_tokens) - 3:
if len(left_tokens) > len(right_tokens):
left_tokens.pop(0)
left_tok2ori_map.pop(0)
else:
right_tokens.pop()
right_tok2ori_map.pop()
bert_tokens = left_tokens + term_tokens + right_tokens
tok2ori_map = left_tok2ori_map + term_tok2ori_map + right_tok2ori_map
truncate_tok_len = len(bert_tokens)
tok_adj = np.zeros(
(truncate_tok_len, truncate_tok_len), dtype='float32')
for i in range(truncate_tok_len):
for j in range(truncate_tok_len):
tok_adj[i][j] = ori_adj[tok2ori_map[i]][tok2ori_map[j]]
context_asp_ids = [tokenizer.cls_token_id]+tokenizer.convert_tokens_to_ids(
bert_tokens)+[tokenizer.sep_token_id]+tokenizer.convert_tokens_to_ids(term_tokens)+[tokenizer.sep_token_id]
context_asp_len = len(context_asp_ids)
paddings = [0] * (tokenizer.max_seq_len - context_asp_len)
context_len = len(bert_tokens)
context_asp_seg_ids = [0] * (1 + context_len + 1) + [1] * (len(term_tokens) + 1) + paddings
src_mask = [0] + [1] * context_len + [0] * (opt.max_length - context_len - 1)
aspect_mask = [0] + [0] * asp_start + [1] * (asp_end - asp_start)
aspect_mask = aspect_mask + (opt.max_length - len(aspect_mask)) * [0]
context_asp_attention_mask = [1] * context_asp_len + paddings
context_asp_ids += paddings
context_asp_ids = np.asarray(context_asp_ids, dtype='int64')
context_asp_seg_ids = np.asarray(context_asp_seg_ids, dtype='int64')
context_asp_attention_mask = np.asarray(context_asp_attention_mask, dtype='int64')
src_mask = np.asarray(src_mask, dtype='int64')
aspect_mask = np.asarray(aspect_mask, dtype='int64')
# pad adj
context_asp_adj_matrix = np.zeros(
(tokenizer.max_seq_len, tokenizer.max_seq_len)).astype('float32')
context_asp_adj_matrix[1:context_len + 1, 1:context_len + 1] = tok_adj
data = {
'text_bert_indices': context_asp_ids,
'bert_segments_ids': context_asp_seg_ids,
'attention_mask': context_asp_attention_mask,
'asp_start': asp_start,
'asp_end': asp_end,
'adj_matrix': context_asp_adj_matrix,
'src_mask': src_mask,
'aspect_mask': aspect_mask,
'polarity': polarity,
}
self.data.append(data)
def __len__(self):
return len(self.data)
def __getitem__(self, idx):
return self.data[idx]
================================================
FILE: DualGCN/dataset/Laptops_allennlp/test.json
================================================
[
{
"token": [
"Boot",
"time",
"is",
"super",
"fast",
",",
"around",
"anywhere",
"from",
"35",
"seconds",
"to",
"1",
"minute",
"."
],
"pos": [
"NOUN",
"NOUN",
"AUX",
"ADV",
"ADJ",
"PUNCT",
"ADV",
"ADV",
"ADP",
"NUM",
"NOUN",
"ADP",
"NUM",
"NOUN",
"PUNCT"
],
"head": [
3,
3,
0,
3,
4,
5,
8,
6,
8,
9,
9,
9,
14,
12,
3
],
"deprel": [
"nsubj",
"nsubj",
"root",
"dep",
"dep",
"mark",
"nsubj",
"dep",
"dep",
"pobj",
"dep",
"dep",
"num",
"dep",
"dep"
],
"aspects": [
{
"term": [
"Boot",
"time"
],
"from": 0,
"to": 2,
"polarity": "positive"
}
]
},
{
"token": [
"tech",
"support",
"would",
"not",
"fix",
"the",
"problem",
"unless",
"I",
"bought",
"your",
"plan",
"for",
"$",
"150",
"plus",
"."
],
"pos": [
"NOUN",
"NOUN",
"VERB",
"PART",
"VERB",
"DET",
"NOUN",
"SCONJ",
"PRON",
"VERB",
"DET",
"NOUN",
"ADP",
"SYM",
"NUM",
"CCONJ",
"PUNCT"
],
"head": [
5,
1,
5,
5,
0,
7,
5,
10,
10,
5,
10,
10,
10,
13,
14,
14,
5
],
"deprel": [
"advmod",
"dep",
"aux",
"neg",
"root",
"dep",
"dobj",
"advmod",
"nsubj",
"dep",
"dep",
"dobj",
"prep",
"pobj",
"num",
"dep",
"punct"
],
"aspects": [
{
"term": [
"tech",
"support"
],
"from": 0,
"to": 2,
"polarity": "negative"
}
]
},
{
"token": [
"Set",
"up",
"was",
"easy",
"."
],
"pos": [
"VERB",
"ADP",
"AUX",
"ADJ",
"PUNCT"
],
"head": [
4,
1,
4,
0,
4
],
"deprel": [
"nsubj",
"advmod",
"cop",
"root",
"dep"
],
"aspects": [
{
"term": [
"Set",
"up"
],
"from": 0,
"to": 2,
"polarity": "positive"
}
]
},
{
"token": [
"Did",
"not",
"enjoy",
"the",
"new",
"Windows",
"8",
"and",
"touchscreen",
"functions",
"."
],
"pos": [
"AUX",
"PART",
"VERB",
"DET",
"ADJ",
"PROPN",
"NUM",
"CCONJ",
"NOUN",
"NOUN",
"PUNCT"
],
"head": [
3,
3,
0,
10,
10,
10,
10,
10,
10,
3,
3
],
"deprel": [
"nsubj",
"neg",
"root",
"dep",
"amod",
"amod",
"dep",
"cc",
"dep",
"dep",
"dep"
],
"aspects": [
{
"term": [
"Windows",
"8"
],
"from": 5,
"to": 7,
"polarity": "negative"
},
{
"term": [
"touchscreen",
"functions"
],
"from": 8,
"to": 10,
"polarity": "negative"
}
]
},
{
"token": [
"Other",
"than",
"not",
"being",
"a",
"fan",
"of",
"click",
"pads",
"(",
"industry",
"standard",
"these",
"days",
")",
"and",
"the",
"lousy",
"internal",
"speakers",
",",
"it",
"'s",
"hard",
"for",
"me",
"to",
"find",
"things",
"about",
"this",
"notebook",
"I",
"do",
"n't",
"like",
",",
"especially",
"considering",
"the",
"$",
"350",
"price",
"tag",
"."
],
"pos": [
"ADJ",
"SCONJ",
"PART",
"AUX",
"DET",
"NOUN",
"ADP",
"NOUN",
"NOUN",
"PUNCT",
"NOUN",
"NOUN",
"DET",
"NOUN",
"PUNCT",
"CCONJ",
"DET",
"ADJ",
"ADJ",
"NOUN",
"PUNCT",
"PRON",
"AUX",
"ADJ",
"ADP",
"PRON",
"PART",
"VERB",
"NOUN",
"ADP",
"DET",
"NOUN",
"PRON",
"AUX",
"PART",
"VERB",
"PUNCT",
"ADV",
"VERB",
"DET",
"SYM",
"NUM",
"NOUN",
"NOUN",
"PUNCT"
],
"head": [
6,
0,
6,
6,
6,
2,
6,
9,
7,
11,
7,
7,
14,
12,
14,
15,
20,
20,
20,
15,
24,
24,
24,
6,
24,
28,
28,
24,
28,
29,
32,
36,
36,
36,
36,
28,
36,
39,
28,
41,
39,
41,
44,
41,
44
],
"deprel": [
"nsubj",
"root",
"aux",
"cop",
"det",
"dep",
"prep",
"dep",
"pobj",
"dep",
"pobj",
"pobj",
"dep",
"dep",
"dep",
"prep",
"det",
"amod",
"amod",
"dep",
"dep",
"nsubj",
"cop",
"dep",
"prep",
"dep",
"dep",
"dep",
"dobj",
"prep",
"dep",
"dep",
"nsubj",
"aux",
"dep",
"ccomp",
"dep",
"advmod",
"dep",
"dep",
"dep",
"dep",
"nn",
"dep",
"amod"
],
"aspects": [
{
"term": [
"internal",
"speakers"
],
"from": 18,
"to": 20,
"polarity": "negative"
},
{
"term": [
"price",
"tag"
],
"from": 42,
"to": 44,
"polarity": "positive"
},
{
"term": [
"click",
"pads"
],
"from": 7,
"to": 9,
"polarity": "negative"
}
]
},
{
"token": [
"No",
"installation",
"disk",
"(",
"DVD",
")",
"is",
"included",
"."
],
"pos": [
"DET",
"NOUN",
"NOUN",
"PUNCT",
"NOUN",
"PUNCT",
"AUX",
"VERB",
"PUNCT"
],
"head": [
3,
3,
6,
6,
6,
8,
8,
0,
8
],
"deprel": [
"amod",
"dep",
"dep",
"dep",
"dep",
"nsubjpass",
"cop",
"root",
"advmod"
],
"aspects": [
{
"term": [
"installation",
"disk",
"(",
"DVD"
],
"from": 1,
"to": 5,
"polarity": "neutral"
}
]
},
{
"token": [
"It",
"'s",
"fast",
",",
"light",
",",
"and",
"simple",
"to",
"use",
"."
],
"pos": [
"PRON",
"AUX",
"ADJ",
"PUNCT",
"NOUN",
"PUNCT",
"CCONJ",
"ADJ",
"PART",
"VERB",
"PUNCT"
],
"head": [
3,
3,
0,
3,
3,
3,
3,
3,
10,
8,
3
],
"deprel": [
"nsubj",
"dep",
"root",
"advmod",
"dep",
"advmod",
"advmod",
"dep",
"aux",
"dep",
"dep"
],
"aspects": [
{
"term": [
"use"
],
"from": 9,
"to": 10,
"polarity": "positive"
}
]
},
{
"token": [
"Works",
"well",
",",
"and",
"I",
"am",
"extremely",
"happy",
"to",
"be",
"back",
"to",
"an",
"apple",
"OS",
"."
],
"pos": [
"PROPN",
"ADV",
"PUNCT",
"CCONJ",
"PRON",
"AUX",
"ADV",
"ADJ",
"PART",
"AUX",
"ADV",
"ADP",
"DET",
"NOUN",
"NOUN",
"PUNCT"
],
"head": [
8,
1,
1,
1,
3,
7,
8,
0,
10,
8,
10,
11,
14,
15,
12,
8
],
"deprel": [
"nsubj",
"dep",
"prep",
"cc",
"dep",
"cop",
"dep",
"root",
"aux",
"xcomp",
"advmod",
"prep",
"det",
"dep",
"pobj",
"punct"
],
"aspects": [
{
"term": [
"Works"
],
"from": 0,
"to": 1,
"polarity": "positive"
},
{
"term": [
"apple",
"OS"
],
"from": 13,
"to": 15,
"polarity": "positive"
}
]
},
{
"token": [
"Sure",
"it",
"'s",
"not",
"light",
"and",
"slim",
"but",
"the",
"features",
"make",
"up",
"for",
"it",
"100",
"%",
"."
],
"pos": [
"INTJ",
"PRON",
"AUX",
"PART",
"ADJ",
"CCONJ",
"ADJ",
"CCONJ",
"DET",
"NOUN",
"VERB",
"ADP",
"ADP",
"PRON",
"NUM",
"NOUN",
"PUNCT"
],
"head": [
5,
5,
5,
5,
0,
5,
5,
7,
10,
11,
7,
11,
11,
13,
16,
13,
5
],
"deprel": [
"nsubj",
"nsubj",
"cop",
"neg",
"root",
"dep",
"dep",
"mark",
"det",
"nsubj",
"ccomp",
"advmod",
"xcomp",
"pobj",
"amod",
"pobj",
"dep"
],
"aspects": [
{
"term": [
"features"
],
"from": 9,
"to": 10,
"polarity": "positive"
}
]
},
{
"token": [
"I",
"am",
"pleased",
"with",
"the",
"fast",
"log",
"on",
",",
"speedy",
"WiFi",
"connection",
"and",
"the",
"long",
"battery",
"life",
"(",
">",
"6",
"hrs",
")",
"."
],
"pos": [
"PRON",
"AUX",
"ADJ",
"ADP",
"DET",
"ADJ",
"NOUN",
"ADP",
"PUNCT",
"ADJ",
"PROPN",
"NOUN",
"CCONJ",
"DET",
"ADJ",
"NOUN",
"NOUN",
"PUNCT",
"PROPN",
"NUM",
"PROPN",
"PUNCT",
"PUNCT"
],
"head": [
3,
3,
0,
3,
6,
7,
4,
7,
8,
3,
12,
10,
12,
17,
17,
17,
12,
17,
18,
19,
22,
20,
20
],
"deprel": [
"nsubj",
"cop",
"root",
"prep",
"det",
"dep",
"pobj",
"advmod",
"pobj",
"dep",
"dep",
"dep",
"dep",
"dep",
"amod",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"punct"
],
"aspects": [
{
"term": [
"log",
"on"
],
"from": 6,
"to": 8,
"polarity": "positive"
},
{
"term": [
"WiFi",
"connection"
],
"from": 10,
"to": 12,
"polarity": "positive"
},
{
"term": [
"battery",
"life"
],
"from": 15,
"to": 17,
"polarity": "positive"
}
]
},
{
"token": [
"The",
"Apple",
"engineers",
"have",
"not",
"yet",
"discovered",
"the",
"delete",
"key",
"."
],
"pos": [
"DET",
"PROPN",
"NOUN",
"AUX",
"PART",
"ADV",
"VERB",
"DET",
"ADJ",
"NOUN",
"PUNCT"
],
"head": [
3,
3,
7,
7,
7,
7,
0,
10,
10,
7,
7
],
"deprel": [
"amod",
"dep",
"nsubj",
"aux",
"aux",
"advmod",
"root",
"dep",
"dep",
"dep",
"punct"
],
"aspects": [
{
"term": [
"delete",
"key"
],
"from": 8,
"to": 10,
"polarity": "negative"
}
]
},
{
"token": [
"Made",
"interneting",
"(",
"part",
"of",
"my",
"business",
")",
"very",
"difficult",
"to",
"maintain",
"."
],
"pos": [
"VERB",
"VERB",
"PUNCT",
"NOUN",
"ADP",
"DET",
"NOUN",
"PUNCT",
"ADV",
"ADJ",
"PART",
"VERB",
"PUNCT"
],
"head": [
3,
3,
10,
3,
4,
7,
5,
5,
10,
0,
12,
10,
10
],
"deprel": [
"amod",
"nn",
"nsubj",
"amod",
"prep",
"poss",
"pobj",
"pobj",
"advmod",
"root",
"aux",
"ccomp",
"punct"
],
"aspects": [
{
"term": [
"interneting"
],
"from": 1,
"to": 2,
"polarity": "negative"
}
]
},
{
"token": [
"Luckily",
",",
"for",
"all",
"of",
"us",
"contemplating",
"the",
"decision",
",",
"the",
"Mac",
"Mini",
"is",
"priced",
"just",
"right",
"."
],
"pos": [
"ADV",
"PUNCT",
"ADP",
"DET",
"ADP",
"PRON",
"VERB",
"DET",
"NOUN",
"PUNCT",
"DET",
"PROPN",
"PROPN",
"AUX",
"VERB",
"ADV",
"ADV",
"PUNCT"
],
"head": [
15,
1,
1,
3,
4,
5,
3,
9,
7,
15,
13,
13,
10,
15,
0,
17,
15,
17
],
"deprel": [
"nsubj",
"prep",
"prep",
"pobj",
"prep",
"pobj",
"pcomp",
"det",
"dobj",
"punct",
"amod",
"amod",
"dep",
"cop",
"root",
"dep",
"dep",
"punct"
],
"aspects": [
{
"term": [
"priced"
],
"from": 14,
"to": 15,
"polarity": "positive"
}
]
},
{
"token": [
"Super",
"light",
",",
"super",
"sexy",
"and",
"everything",
"just",
"works",
"."
],
"pos": [
"ADV",
"ADJ",
"PUNCT",
"ADV",
"ADJ",
"CCONJ",
"PRON",
"ADV",
"VERB",
"PUNCT"
],
"head": [
4,
1,
4,
5,
0,
5,
5,
9,
7,
5
],
"deprel": [
"amod",
"dep",
"nsubj",
"dep",
"root",
"cc",
"dep",
"amod",
"dep",
"punct"
],
"aspects": [
{
"term": [
"works"
],
"from": 8,
"to": 9,
"polarity": "positive"
}
]
},
{
"token": [
"Only",
"problem",
"that",
"I",
"had",
"was",
"that",
"the",
"track",
"pad",
"was",
"not",
"very",
"good",
"for",
"me",
",",
"I",
"only",
"had",
"a",
"problem",
"once",
"or",
"twice",
"with",
"it",
",",
"But",
"probably",
"my",
"computer",
"was",
"a",
"bit",
"defective",
"."
],
"pos": [
"ADV",
"NOUN",
"DET",
"PRON",
"AUX",
"AUX",
"SCONJ",
"DET",
"NOUN",
"NOUN",
"AUX",
"PART",
"ADV",
"ADJ",
"ADP",
"PRON",
"PUNCT",
"PRON",
"ADV",
"AUX",
"DET",
"NOUN",
"ADV",
"CCONJ",
"ADV",
"ADP",
"PRON",
"PUNCT",
"CCONJ",
"ADV",
"DET",
"NOUN",
"AUX",
"DET",
"NOUN",
"ADJ",
"PUNCT"
],
"head": [
2,
6,
6,
6,
2,
0,
14,
10,
10,
14,
14,
14,
14,
6,
14,
15,
14,
20,
20,
14,
22,
20,
20,
23,
23,
25,
26,
29,
26,
29,
32,
30,
36,
35,
36,
26,
36
],
"deprel": [
"advmod",
"nsubj",
"nsubj",
"nsubj",
"dep",
"root",
"mark",
"amod",
"amod",
"nsubj",
"cop",
"neg",
"advmod",
"dep",
"prep",
"pobj",
"dep",
"nsubj",
"nsubj",
"dep",
"dep",
"dobj",
"dep",
"dep",
"dep",
"dep",
"pobj",
"dep",
"pobj",
"dep",
"amod",
"dep",
"cop",
"dep",
"npadvmod",
"pcomp",
"ccomp"
],
"aspects": [
{
"term": [
"track",
"pad"
],
"from": 8,
"to": 10,
"polarity": "negative"
}
]
},
{
"token": [
"It",
"is",
"super",
"fast",
"and",
"has",
"outstanding",
"graphics",
"."
],
"pos": [
"PRON",
"AUX",
"ADV",
"ADJ",
"CCONJ",
"AUX",
"ADJ",
"NOUN",
"PUNCT"
],
"head": [
3,
4,
4,
0,
3,
4,
8,
3,
3
],
"deprel": [
"nsubj",
"cop",
"dep",
"root",
"advmod",
"dep",
"amod",
"dep",
"dep"
],
"aspects": [
{
"term": [
"graphics"
],
"from": 7,
"to": 8,
"polarity": "positive"
}
]
},
{
"token": [
"But",
"the",
"mountain",
"lion",
"is",
"just",
"too",
"slow",
"."
],
"pos": [
"CCONJ",
"DET",
"NOUN",
"NOUN",
"AUX",
"ADV",
"ADV",
"ADJ",
"PUNCT"
],
"head": [
8,
4,
4,
1,
8,
8,
8,
0,
8
],
"deprel": [
"nsubj",
"det",
"dep",
"dep",
"cop",
"dep",
"amod",
"root",
"punct"
],
"aspects": [
{
"term": [
"mountain",
"lion"
],
"from": 2,
"to": 4,
"polarity": "negative"
}
]
},
{
"token": [
"Strong",
"build",
"though",
"which",
"really",
"adds",
"to",
"its",
"durability",
"."
],
"pos": [
"ADJ",
"NOUN",
"SCONJ",
"DET",
"ADV",
"VERB",
"ADP",
"DET",
"NOUN",
"PUNCT"
],
"head": [
6,
6,
6,
6,
6,
0,
6,
9,
7,
6
],
"deprel": [
"nsubj",
"dep",
"nsubj",
"nsubj",
"nsubj",
"root",
"prep",
"poss",
"pobj",
"punct"
],
"aspects": [
{
"term": [
"durability"
],
"from": 8,
"to": 9,
"polarity": "positive"
},
{
"term": [
"build"
],
"from": 1,
"to": 2,
"polarity": "positive"
}
]
},
{
"token": [
"The",
"battery",
"life",
"is",
"excellent",
"-",
"6",
"-",
"7",
"hours",
"without",
"charging",
"."
],
"pos": [
"DET",
"NOUN",
"NOUN",
"AUX",
"ADJ",
"PUNCT",
"NUM",
"SYM",
"NUM",
"NOUN",
"ADP",
"VERB",
"PUNCT"
],
"head": [
2,
5,
2,
5,
0,
5,
5,
7,
10,
7,
5,
11,
5
],
"deprel": [
"det",
"nsubj",
"dep",
"cop",
"root",
"dep",
"dep",
"dep",
"num",
"dep",
"prep",
"pcomp",
"dep"
],
"aspects": [
{
"term": [
"battery",
"life"
],
"from": 1,
"to": 3,
"polarity": "positive"
}
]
},
{
"token": [
"I",
"'",
"ve",
"had",
"my",
"computer",
"for",
"2",
"weeks",
"already",
"and",
"it",
"works",
"perfectly",
"."
],
"pos": [
"PRON",
"PUNCT",
"PROPN",
"AUX",
"DET",
"NOUN",
"ADP",
"NUM",
"NOUN",
"ADV",
"CCONJ",
"PRON",
"VERB",
"ADV",
"PUNCT"
],
"head": [
3,
3,
0,
3,
6,
4,
6,
9,
7,
7,
10,
14,
14,
10,
14
],
"deprel": [
"nsubj",
"punct",
"root",
"dep",
"dep",
"dobj",
"prep",
"num",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep"
],
"aspects": [
{
"term": [
"works"
],
"from": 11,
"to": 12,
"polarity": "positive"
}
]
},
{
"token": [
"And",
"I",
"may",
"be",
"the",
"only",
"one",
"but",
"I",
"am",
"really",
"liking",
"Windows",
"8",
"."
],
"pos": [
"CCONJ",
"PRON",
"VERB",
"AUX",
"DET",
"ADJ",
"NUM",
"CCONJ",
"PRON",
"AUX",
"ADV",
"VERB",
"PROPN",
"NUM",
"PUNCT"
],
"head": [
5,
5,
5,
5,
0,
7,
5,
7,
12,
12,
12,
7,
12,
13,
14
],
"deprel": [
"cc",
"nsubj",
"aux",
"cop",
"root",
"advmod",
"dep",
"prep",
"nsubj",
"aux",
"advmod",
"dep",
"dep",
"amod",
"punct"
],
"aspects": [
{
"term": [
"Windows",
"8"
],
"from": 12,
"to": 14,
"polarity": "positive"
}
]
},
{
"token": [
"The",
"baterry",
"is",
"very",
"longer",
"."
],
"pos": [
"DET",
"NOUN",
"AUX",
"ADV",
"ADJ",
"PUNCT"
],
"head": [
2,
5,
5,
5,
0,
5
],
"deprel": [
"det",
"nsubj",
"cop",
"advmod",
"root",
"dep"
],
"aspects": [
{
"term": [
"baterry"
],
"from": 1,
"to": 2,
"polarity": "positive"
}
]
},
{
"token": [
"Its",
"size",
"is",
"ideal",
"and",
"the",
"weight",
"is",
"acceptable",
"."
],
"pos": [
"DET",
"NOUN",
"AUX",
"ADJ",
"CCONJ",
"DET",
"NOUN",
"AUX",
"ADJ",
"PUNCT"
],
"head": [
2,
4,
4,
0,
4,
7,
9,
9,
4,
4
],
"deprel": [
"amod",
"nsubj",
"cop",
"root",
"dep",
"det",
"nsubj",
"cop",
"ccomp",
"punct"
],
"aspects": [
{
"term": [
"size"
],
"from": 1,
"to": 2,
"polarity": "positive"
},
{
"term": [
"weight"
],
"from": 6,
"to": 7,
"polarity": "positive"
}
]
},
{
"token": [
"I",
"can",
"say",
"that",
"I",
"am",
"fully",
"satisfied",
"with",
"the",
"performance",
"that",
"the",
"computer",
"has",
"supplied",
"."
],
"pos": [
"PRON",
"VERB",
"VERB",
"SCONJ",
"PRON",
"AUX",
"ADV",
"ADJ",
"ADP",
"DET",
"NOUN",
"SCONJ",
"DET",
"NOUN",
"AUX",
"VERB",
"PUNCT"
],
"head": [
3,
3,
0,
8,
8,
7,
8,
3,
8,
11,
9,
16,
14,
16,
16,
11,
3
],
"deprel": [
"nsubj",
"aux",
"root",
"mark",
"nsubj",
"auxpass",
"dep",
"ccomp",
"prep",
"det",
"pobj",
"dobj",
"nn",
"nsubj",
"aux",
"rcmod",
"punct"
],
"aspects": [
{
"term": [
"performance"
],
"from": 10,
"to": 11,
"polarity": "positive"
}
]
},
{
"token": [
"This",
"laptop",
"has",
"only",
"2",
"USB",
"ports",
",",
"and",
"they",
"are",
"both",
"on",
"the",
"same",
"side",
"."
],
"pos": [
"DET",
"NOUN",
"AUX",
"ADV",
"NUM",
"NOUN",
"NOUN",
"PUNCT",
"CCONJ",
"PRON",
"AUX",
"DET",
"ADP",
"DET",
"ADJ",
"NOUN",
"PUNCT"
],
"head": [
3,
3,
0,
5,
7,
7,
3,
3,
11,
11,
5,
13,
11,
16,
16,
13,
3
],
"deprel": [
"nsubj",
"nsubj",
"root",
"dep",
"dep",
"dep",
"dobj",
"advmod",
"dep",
"nsubj",
"dep",
"dep",
"prep",
"det",
"amod",
"pobj",
"punct"
],
"aspects": [
{
"term": [
"USB",
"ports"
],
"from": 5,
"to": 7,
"polarity": "negative"
}
]
},
{
"token": [
"It",
"has",
"so",
"much",
"more",
"speed",
"and",
"the",
"screen",
"is",
"very",
"sharp",
"."
],
"pos": [
"PRON",
"AUX",
"ADV",
"ADV",
"ADJ",
"NOUN",
"CCONJ",
"DET",
"NOUN",
"AUX",
"ADV",
"ADJ",
"PUNCT"
],
"head": [
12,
1,
5,
5,
6,
2,
6,
9,
12,
12,
12,
0,
12
],
"deprel": [
"nsubj",
"dep",
"dep",
"dep",
"dep",
"dep",
"cc",
"det",
"nsubj",
"cop",
"advmod",
"root",
"punct"
],
"aspects": [
{
"term": [
"speed"
],
"from": 5,
"to": 6,
"polarity": "positive"
},
{
"term": [
"screen"
],
"from": 8,
"to": 9,
"polarity": "positive"
}
]
},
{
"token": [
"Everything",
"I",
"wanted",
"and",
"everything",
"I",
"needed",
"and",
"the",
"price",
"was",
"great",
"!"
],
"pos": [
"PRON",
"PRON",
"VERB",
"CCONJ",
"PRON",
"PRON",
"VERB",
"CCONJ",
"DET",
"NOUN",
"AUX",
"ADJ",
"PUNCT"
],
"head": [
3,
3,
0,
3,
3,
7,
5,
7,
10,
12,
12,
3,
12
],
"deprel": [
"nsubj",
"nsubj",
"root",
"prep",
"dobj",
"dep",
"dep",
"dep",
"det",
"nsubj",
"cop",
"dep",
"dep"
],
"aspects": [
{
"term": [
"price"
],
"from": 9,
"to": 10,
"polarity": "positive"
}
]
},
{
"token": [
"It",
"'s",
"not",
"inexpensive",
"but",
"the",
"Hardware",
"performance",
"is",
"impressive",
"for",
"a",
"computer",
"this",
"small",
"."
],
"pos": [
"PRON",
"AUX",
"PART",
"ADJ",
"CCONJ",
"DET",
"PROPN",
"NOUN",
"AUX",
"ADJ",
"ADP",
"DET",
"NOUN",
"DET",
"ADJ",
"PUNCT"
],
"head": [
4,
4,
4,
0,
4,
8,
8,
10,
10,
4,
10,
13,
11,
15,
13,
4
],
"deprel": [
"nsubj",
"cop",
"neg",
"root",
"mark",
"amod",
"amod",
"nsubj",
"cop",
"dep",
"prep",
"det",
"pobj",
"dep",
"amod",
"dep"
],
"aspects": [
{
"term": [
"Hardware",
"performance"
],
"from": 6,
"to": 8,
"polarity": "positive"
}
]
},
{
"token": [
"This",
"thing",
"is",
"awesome",
",",
"everything",
"always",
"works",
",",
"everything",
"is",
"always",
"easy",
"to",
"set",
"up",
",",
"everything",
"is",
"compatible",
",",
"its",
"literally",
"everything",
"I",
"could",
"ask",
"for",
"."
],
"pos": [
"DET",
"NOUN",
"AUX",
"ADJ",
"PUNCT",
"PRON",
"ADV",
"VERB",
"PUNCT",
"PRON",
"AUX",
"ADV",
"ADJ",
"PART",
"VERB",
"ADP",
"PUNCT",
"PRON",
"AUX",
"ADJ",
"PUNCT",
"DET",
"ADJ",
"PRON",
"PRON",
"VERB",
"VERB",
"ADP",
"PUNCT"
],
"head": [
4,
4,
4,
0,
4,
4,
6,
6,
4,
13,
13,
13,
4,
15,
13,
15,
15,
20,
20,
13,
20,
23,
20,
23,
27,
27,
24,
27,
28
],
"deprel": [
"nsubj",
"nsubj",
"cop",
"root",
"punct",
"nsubj",
"amod",
"dep",
"punct",
"nsubj",
"cop",
"advmod",
"ccomp",
"dep",
"dep",
"dep",
"npadvmod",
"nsubj",
"cop",
"ccomp",
"dep",
"poss",
"dep",
"dep",
"nsubj",
"aux",
"rcmod",
"prep",
"pobj"
],
"aspects": [
{
"term": [
"works"
],
"from": 7,
"to": 8,
"polarity": "positive"
},
{
"term": [
"set",
"up"
],
"from": 14,
"to": 16,
"polarity": "positive"
}
]
},
{
"token": [
"Keyboard",
"responds",
"well",
"to",
"presses",
"."
],
"pos": [
"NOUN",
"VERB",
"ADV",
"ADP",
"NOUN",
"PUNCT"
],
"head": [
2,
0,
4,
2,
4,
2
],
"deprel": [
"nsubj",
"root",
"advmod",
"prep",
"pobj",
"punct"
],
"aspects": [
{
"term": [
"Keyboard"
],
"from": 0,
"to": 1,
"polarity": "positive"
}
]
},
{
"token": [
"Lastly",
",",
"Windows",
"8",
"is",
"annoying",
"."
],
"pos": [
"ADV",
"PUNCT",
"PROPN",
"NUM",
"AUX",
"ADJ",
"PUNCT"
],
"head": [
6,
1,
2,
2,
6,
0,
6
],
"deprel": [
"nsubj",
"prep",
"dep",
"dep",
"cop",
"root",
"advmod"
],
"aspects": [
{
"term": [
",",
"Windows",
"8"
],
"from": 1,
"to": 4,
"polarity": "negative"
}
]
},
{
"token": [
"Everything",
"is",
"so",
"easy",
"and",
"intuitive",
"to",
"setup",
"or",
"configure",
"."
],
"pos": [
"PRON",
"AUX",
"ADV",
"ADJ",
"CCONJ",
"ADJ",
"PART",
"NOUN",
"CCONJ",
"NOUN",
"PUNCT"
],
"head": [
4,
4,
4,
0,
4,
4,
8,
4,
8,
8,
4
],
"deprel": [
"nsubj",
"cop",
"advmod",
"root",
"advmod",
"dep",
"dep",
"dep",
"cc",
"dep",
"punct"
],
"aspects": [
{
"term": [
"setup"
],
"from": 7,
"to": 8,
"polarity": "positive"
},
{
"term": [
"configure"
],
"from": 9,
"to": 10,
"polarity": "positive"
}
]
},
{
"token": [
"Biggest",
"complaint",
"is",
"Windows",
"8",
"."
],
"pos": [
"ADJ",
"NOUN",
"AUX",
"PROPN",
"NUM",
"PUNCT"
],
"head": [
4,
4,
4,
0,
4,
4
],
"deprel": [
"nsubj",
"amod",
"cop",
"root",
"dep",
"advmod"
],
"aspects": [
{
"term": [
"Windows",
"8"
],
"from": 3,
"to": 5,
"polarity": "negative"
}
]
},
{
"token": [
"Only",
"2",
"usb",
"ports",
"...",
"seems",
"kind",
"of",
"...",
"limited",
"."
],
"pos": [
"ADV",
"NUM",
"NOUN",
"NOUN",
"PUNCT",
"VERB",
"ADV",
"ADV",
"PUNCT",
"ADJ",
"PUNCT"
],
"head": [
4,
4,
4,
7,
4,
7,
0,
7,
8,
8,
7
],
"deprel": [
"quantmod",
"dep",
"dep",
"nsubj",
"dep",
"cop",
"root",
"prep",
"punct",
"pobj",
"punct"
],
"aspects": [
{
"term": [
"usb",
"ports"
],
"from": 2,
"to": 4,
"polarity": "negative"
}
]
},
{
"token": [
"It",
"has",
"all",
"the",
"expected",
"features",
"and",
"more",
"+",
"plus",
"a",
"wide",
"screen",
"and",
"more",
"than",
"roomy",
"keyboard",
"."
],
"pos": [
"PRON",
"AUX",
"DET",
"DET",
"VERB",
"NOUN",
"CCONJ",
"ADJ",
"CCONJ",
"CCONJ",
"DET",
"ADJ",
"NOUN",
"CCONJ",
"ADJ",
"SCONJ",
"PROPN",
"NOUN",
"PUNCT"
],
"head": [
6,
6,
5,
5,
6,
0,
6,
9,
7,
9,
13,
13,
10,
9,
14,
15,
18,
16,
6
],
"deprel": [
"nsubj",
"aux",
"dep",
"dep",
"dobj",
"root",
"iobj",
"dep",
"dep",
"dep",
"dep",
"amod",
"dep",
"dep",
"dep",
"prep",
"nn",
"pobj",
"dep"
],
"aspects": [
{
"term": [
"features"
],
"from": 5,
"to": 6,
"polarity": "positive"
},
{
"term": [
"screen"
],
"from": 12,
"to": 13,
"polarity": "positive"
},
{
"term": [
"keyboard"
],
"from": 17,
"to": 18,
"polarity": "positive"
}
]
},
{
"token": [
"Amazing",
"Performance",
"for",
"anything",
"I",
"throw",
"at",
"it",
"."
],
"pos": [
"ADJ",
"NOUN",
"ADP",
"PRON",
"PRON",
"VERB",
"ADP",
"PRON",
"PUNCT"
],
"head": [
2,
0,
1,
3,
6,
4,
6,
7,
3
],
"deprel": [
"nsubj",
"root",
"prep",
"pobj",
"nsubj",
"dep",
"advmod",
"dep",
"dep"
],
"aspects": [
{
"term": [
"Performance"
],
"from": 1,
"to": 2,
"polarity": "positive"
}
]
},
{
"token": [
"The",
"receiver",
"was",
"full",
"of",
"superlatives",
"for",
"the",
"quality",
"and",
"performance",
"."
],
"pos": [
"DET",
"NOUN",
"AUX",
"ADJ",
"ADP",
"NOUN",
"ADP",
"DET",
"NOUN",
"CCONJ",
"NOUN",
"PUNCT"
],
"head": [
2,
4,
4,
0,
4,
5,
5,
9,
7,
11,
7,
4
],
"deprel": [
"dep",
"nsubj",
"cop",
"root",
"prep",
"pobj",
"prep",
"dep",
"pobj",
"amod",
"pobj",
"punct"
],
"aspects": [
{
"term": [
"quality"
],
"from": 8,
"to": 9,
"polarity": "positive"
},
{
"term": [
"performance"
],
"from": 10,
"to": 11,
"polarity": "positive"
}
]
},
{
"token": [
"I",
"was",
"extremely",
"happy",
"with",
"the",
"OS",
"itself",
"."
],
"pos": [
"PRON",
"AUX",
"ADV",
"ADJ",
"ADP",
"DET",
"NOUN",
"PRON",
"PUNCT"
],
"head": [
4,
4,
4,
0,
4,
7,
5,
7,
4
],
"deprel": [
"nsubj",
"cop",
"dep",
"root",
"prep",
"det",
"pobj",
"amod",
"punct"
],
"aspects": [
{
"term": [
"OS"
],
"from": 6,
"to": 7,
"polarity": "positive"
}
]
},
{
"token": [
"The",
"new",
"MBP",
"offers",
"great",
"portability",
"and",
"gives",
"us",
"confidence",
"that",
"we",
"are",
"not",
"going",
"to",
"need",
"to",
"purchase",
"a",
"new",
"laptop",
"in",
"18",
"months",
"."
],
"pos": [
"DET",
"ADJ",
"PROPN",
"VERB",
"ADJ",
"NOUN",
"CCONJ",
"VERB",
"PRON",
"NOUN",
"SCONJ",
"PRON",
"AUX",
"PART",
"VERB",
"PART",
"VERB",
"PART",
"VERB",
"DET",
"ADJ",
"NOUN",
"ADP",
"NUM",
"NOUN",
"PUNCT"
],
"head": [
4,
3,
4,
0,
4,
4,
4,
4,
10,
8,
15,
15,
15,
15,
10,
17,
15,
19,
17,
22,
22,
19,
19,
25,
23,
24
],
"deprel": [
"advmod",
"dep",
"amod",
"root",
"dep",
"dep",
"cc",
"dep",
"nsubj",
"dep",
"mark",
"nsubj",
"cop",
"neg",
"ccomp",
"aux",
"xcomp",
"aux",
"xcomp",
"dep",
"amod",
"dobj",
"prep",
"num",
"pobj",
"dep"
],
"aspects": [
{
"term": [
"portability"
],
"from": 5,
"to": 6,
"polarity": "positive"
}
]
},
{
"token": [
"The",
"criticism",
"has",
"waned",
",",
"and",
"now",
"I",
"'d",
"be",
"the",
"first",
"to",
"recommend",
"an",
"Air",
"for",
"truly",
"portable",
"computing",
"."
],
"pos": [
"DET",
"NOUN",
"AUX",
"VERB",
"PUNCT",
"CCONJ",
"ADV",
"PRON",
"VERB",
"AUX",
"DET",
"ADJ",
"PART",
"VERB",
"DET",
"NOUN",
"ADP",
"ADV",
"ADJ",
"NOUN",
"PUNCT"
],
"head": [
2,
4,
4,
0,
4,
4,
12,
12,
12,
12,
12,
4,
14,
12,
20,
17,
14,
20,
20,
17,
4
],
"deprel": [
"det",
"nsubj",
"aux",
"root",
"punct",
"cc",
"advmod",
"nsubj",
"aux",
"cop",
"amod",
"dep",
"aux",
"dep",
"det",
"pobj",
"prep",
"amod",
"amod",
"pobj",
"punct"
],
"aspects": [
{
"term": [
"portable",
"computing"
],
"from": 18,
"to": 20,
"polarity": "positive"
}
]
},
{
"token": [
"I",
"would",
"have",
"given",
"it",
"5",
"starts",
"was",
"it",
"not",
"for",
"the",
"fact",
"that",
"it",
"had",
"Windows",
"8"
],
"pos": [
"PRON",
"VERB",
"AUX",
"VERB",
"PRON",
"NUM",
"NOUN",
"AUX",
"PRON",
"PART",
"ADP",
"DET",
"NOUN",
"SCONJ",
"PRON",
"AUX",
"NOUN",
"NUM"
],
"head": [
8,
4,
4,
8,
6,
7,
4,
9,
0,
9,
10,
13,
11,
17,
17,
17,
13,
17
],
"deprel": [
"nsubj",
"aux",
"aux",
"csubj",
"dep",
"dep",
"dep",
"dep",
"root",
"neg",
"prep",
"det",
"pobj",
"mark",
"nsubj",
"aux",
"ccomp",
"dep"
],
"aspects": [
{
"term": [
"Windows",
"8"
],
"from": 16,
"to": 18,
"polarity": "negative"
}
]
},
{
"token": [
"MS",
"Office",
"2011",
"for",
"Mac",
"is",
"wonderful",
",",
"well",
"worth",
"it",
"."
],
"pos": [
"PROPN",
"PROPN",
"NUM",
"ADP",
"PROPN",
"AUX",
"ADJ",
"PUNCT",
"INTJ",
"ADJ",
"PRON",
"PUNCT"
],
"head": [
2,
7,
2,
3,
4,
7,
0,
7,
7,
7,
10,
7
],
"deprel": [
"nn",
"nsubj",
"advmod",
"prep",
"pobj",
"cop",
"root",
"advmod",
"dep",
"dep",
"dep",
"dep"
],
"aspects": [
{
"term": [
"MS",
"Office",
"2011",
"for",
"Mac"
],
"from": 0,
"to": 5,
"polarity": "positive"
}
]
},
{
"token": [
"But",
"the",
"performance",
"of",
"Mac",
"Mini",
"is",
"a",
"huge",
"disappointment",
"."
],
"pos": [
"CCONJ",
"DET",
"NOUN",
"ADP",
"PROPN",
"PROPN",
"AUX",
"DET",
"ADJ",
"NOUN",
"PUNCT"
],
"head": [
10,
3,
10,
3,
6,
4,
10,
10,
10,
0,
10
],
"deprel": [
"csubj",
"det",
"nsubj",
"prep",
"nn",
"pobj",
"cop",
"dep",
"amod",
"root",
"punct"
],
"aspects": [
{
"term": [
"performance"
],
"from": 2,
"to": 3,
"polarity": "negative"
}
]
},
{
"token": [
"They",
"do",
"n't",
"just",
"look",
"good",
";",
"they",
"deliver",
"excellent",
"performance",
"."
],
"pos": [
"PRON",
"AUX",
"PART",
"ADV",
"VERB",
"ADJ",
"PUNCT",
"PRON",
"VERB",
"ADJ",
"NOUN",
"PUNCT"
],
"head": [
5,
5,
5,
5,
0,
5,
5,
9,
5,
11,
9,
5
],
"deprel": [
"nsubj",
"aux",
"neg",
"advmod",
"root",
"acomp",
"punct",
"nsubj",
"parataxis",
"amod",
"dobj",
"dep"
],
"aspects": [
{
"term": [
"look"
],
"from": 4,
"to": 5,
"polarity": "positive"
},
{
"term": [
"performance"
],
"from": 10,
"to": 11,
"polarity": "positive"
}
]
},
{
"token": [
"I",
"have",
"had",
"it",
"over",
"a",
"year",
"now",
"with",
"out",
"a",
"Glitch",
"of",
"any",
"kind",
".",
".",
"I",
"love",
"the",
"lit",
"up",
"keys",
"and",
"screen",
"display",
"...",
"this",
"thing",
"is",
"Fast",
"and",
"clear",
"as",
"can",
"be",
"."
],
"pos": [
"PRON",
"AUX",
"VERB",
"PRON",
"ADP",
"DET",
"NOUN",
"ADV",
"ADP",
"ADP",
"DET",
"NOUN",
"ADP",
"DET",
"NOUN",
"PUNCT",
"PUNCT",
"PRON",
"VERB",
"DET",
"NOUN",
"ADP",
"NOUN",
"CCONJ",
"NOUN",
"NOUN",
"PUNCT",
"DET",
"NOUN",
"AUX",
"ADJ",
"CCONJ",
"ADJ",
"SCONJ",
"VERB",
"AUX",
"PUNCT"
],
"head": [
8,
3,
8,
8,
7,
7,
8,
9,
0,
9,
12,
10,
12,
15,
13,
15,
19,
19,
16,
19,
20,
21,
22,
21,
26,
22,
31,
29,
31,
31,
33,
31,
37,
36,
36,
33,
3
],
"deprel": [
"nsubj",
"aux",
"dep",
"dep",
"advmod",
"dep",
"dep",
"dep",
"root",
"dep",
"det",
"dep",
"prep",
"amod",
"pobj",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"nsubj",
"cop",
"ccomp",
"advmod",
"dep",
"mark",
"dep",
"ccomp",
"punct"
],
"aspects": [
{
"term": [
"lit",
"up",
"keys"
],
"from": 20,
"to": 23,
"polarity": "positive"
},
{
"term": [
"screen",
"display"
],
"from": 24,
"to": 26,
"polarity": "positive"
}
]
},
{
"token": [
"The",
"Mountain",
"Lion",
"OS",
"is",
"not",
"hard",
"to",
"figure",
"out",
"if",
"you",
"are",
"familiar",
"with",
"Microsoft",
"Windows",
"."
],
"pos": [
"DET",
"PROPN",
"PROPN",
"PROPN",
"AUX",
"PART",
"ADJ",
"PART",
"VERB",
"ADP",
"SCONJ",
"PRON",
"AUX",
"ADJ",
"ADP",
"PROPN",
"PROPN",
"PUNCT"
],
"head": [
4,
4,
4,
7,
7,
7,
0,
7,
7,
9,
14,
14,
14,
7,
14,
17,
15,
7
],
"deprel": [
"det",
"dep",
"nn",
"nsubj",
"cop",
"neg",
"root",
"dep",
"dep",
"dep",
"mark",
"nsubj",
"cop",
"advcl",
"prep",
"nn",
"pobj",
"punct"
],
"aspects": [
{
"term": [
"Mountain",
"Lion",
"OS"
],
"from": 1,
"to": 4,
"polarity": "positive"
},
{
"term": [
"Microsoft",
"Windows"
],
"from": 15,
"to": 17,
"polarity": "neutral"
}
]
},
{
"token": [
"However",
",",
"I",
"can",
"refute",
"that",
"OSX",
"is",
"`",
"`",
"FAST",
"''",
"."
],
"pos": [
"ADV",
"PUNCT",
"PRON",
"VERB",
"VERB",
"SCONJ",
"PROPN",
"AUX",
"PUNCT",
"PUNCT",
"VERB",
"PUNCT",
"PUNCT"
],
"head": [
5,
5,
5,
5,
0,
9,
9,
9,
5,
11,
9,
11,
5
],
"deprel": [
"advmod",
"prep",
"nsubj",
"aux",
"root",
"mark",
"nsubj",
"aux",
"ccomp",
"punct",
"dep",
"punct",
"punct"
],
"aspects": [
{
"term": [
"OSX"
],
"from": 6,
"to": 7,
"polarity": "negative"
}
]
},
{
"token": [
"Enjoy",
"using",
"Microsoft",
"Office",
"!"
],
"pos": [
"PROPN",
"VERB",
"PROPN",
"PROPN",
"PUNCT"
],
"head": [
0,
1,
4,
2,
4
],
"deprel": [
"root",
"partmod",
"dep",
"dep",
"dep"
],
"aspects": [
{
"term": [
"Microsoft",
"Office"
],
"from": 2,
"to": 4,
"polarity": "positive"
}
]
},
{
"token": [
"Incredible",
"graphics",
"and",
"brilliant",
"colors",
"."
],
"pos": [
"ADJ",
"NOUN",
"CCONJ",
"ADJ",
"NOUN",
"PUNCT"
],
"head": [
2,
0,
2,
5,
2,
5
],
"deprel": [
"dep",
"root",
"dep",
"advmod",
"dep",
"advmod"
],
"aspects": [
{
"term": [
"graphics"
],
"from": 1,
"to": 2,
"polarity": "positive"
},
{
"term": [
"colors"
],
"from": 4,
"to": 5,
"polarity": "positive"
}
]
},
{
"token": [
"Built",
"-",
"in",
"apps",
"are",
"purely",
"amazing",
"."
],
"pos": [
"VERB",
"PUNCT",
"ADP",
"NOUN",
"AUX",
"ADV",
"ADJ",
"PUNCT"
],
"head": [
6,
1,
1,
3,
6,
7,
0,
7
],
"deprel": [
"csubj",
"punct",
"prep",
"dep",
"cop",
"amod",
"root",
"dep"
],
"aspects": [
{
"term": [
"Built",
"-",
"in",
"apps"
],
"from": 0,
"to": 4,
"polarity": "positive"
}
]
},
{
"token": [
"Cons",
":",
"Screen",
"resolution",
"."
],
"pos": [
"NOUN",
"PUNCT",
"NOUN",
"NOUN",
"PUNCT"
],
"head": [
0,
1,
4,
1,
1
],
"deprel": [
"root",
"dep",
"dep",
"dep",
"punct"
],
"aspects": [
{
"term": [
":",
"Screen",
"resolution"
],
"from": 1,
"to": 4,
"polarity": "negative"
}
]
},
{
"token": [
"From",
"the",
"speed",
"to",
"the",
"multi",
"touch",
"gestures",
"this",
"operating",
"system",
"beats",
"Windows",
"easily",
"."
],
"pos": [
"ADP",
"DET",
"NOUN",
"ADP",
"DET",
"ADJ",
"NOUN",
"VERB",
"DET",
"VERB",
"NOUN",
"VERB",
"NOUN",
"ADV",
"PUNCT"
],
"head": [
12,
3,
1,
1,
7,
7,
8,
4,
11,
11,
12,
0,
12,
12,
14
],
"deprel": [
"prep",
"det",
"dep",
"prep",
"det",
"dep",
"amod",
"pobj",
"amod",
"amod",
"nsubj",
"root",
"dobj",
"acomp",
"punct"
],
"aspects": [
{
"term": [
"operating",
"system"
],
"from": 9,
"to": 11,
"polarity": "positive"
},
{
"term": [
"Windows"
],
"from": 12,
"to": 13,
"polarity": "negative"
},
{
"term": [
"speed"
],
"from": 2,
"to": 3,
"polarity": "positive"
},
{
"term": [
"multi",
"touch",
"gestures"
],
"from": 5,
"to": 8,
"polarity": "positive"
}
]
},
{
"token": [
"I",
"really",
"like",
"the",
"size",
"and",
"I",
"'",
"m",
"a",
"fan",
"of",
"the",
"ACERS",
"."
],
"pos": [
"PRON",
"ADV",
"VERB",
"DET",
"NOUN",
"CCONJ",
"PRON",
"PUNCT",
"PROPN",
"DET",
"NOUN",
"ADP",
"DET",
"PROPN",
"PUNCT"
],
"head": [
3,
3,
0,
5,
6,
3,
9,
7,
11,
11,
3,
11,
14,
12,
3
],
"deprel": [
"nsubj",
"dep",
"root",
"dep",
"dep",
"cc",
"dep",
"possessive",
"dep",
"det",
"dep",
"prep",
"det",
"pobj",
"punct"
],
"aspects": [
{
"term": [
"size"
],
"from": 4,
"to": 5,
"polarity": "positive"
}
]
},
{
"token": [
"I",
"opted",
"for",
"the",
"SquareTrade",
"3-Year",
"Computer",
"Accidental",
"Protection",
"Warranty",
"(",
"$",
"1500",
"-",
"2000",
")",
"which",
"also",
"support",
"`",
"`",
"accidents",
"''",
"like",
"drops",
"and",
"spills",
"that",
"are",
"NOT",
"covered",
"by",
"AppleCare",
"."
],
"pos": [
"PRON",
"VERB",
"ADP",
"DET",
"PROPN",
"PROPN",
"PROPN",
"PROPN",
"PROPN",
"PROPN",
"PUNCT",
"SYM",
"NUM",
"SYM",
"NUM",
"PUNCT",
"DET",
"ADV",
"VERB",
"PUNCT",
"PUNCT",
"NOUN",
"PUNCT",
"SCONJ",
"NOUN",
"CCONJ",
"NOUN",
"DET",
"AUX",
"ADV",
"VERB",
"ADP",
"PROPN",
"PUNCT"
],
"head": [
2,
0,
2,
8,
6,
8,
8,
3,
8,
12,
12,
8,
12,
12,
12,
15,
18,
2,
3,
19,
20,
21,
21,
21,
24,
27,
25,
30,
30,
27,
30,
31,
32,
2
],
"deprel": [
"nsubj",
"root",
"prep",
"amod",
"dep",
"dep",
"amod",
"pobj",
"dep",
"dep",
"dep",
"dep",
"dep",
"prep",
"dep",
"dep",
"nsubj",
"advmod",
"dep",
"dep",
"dep",
"dep",
"punct",
"prep",
"dep",
"amod",
"dep",
"nsubj",
"cop",
"rcmod",
"amod",
"prep",
"pobj",
"punct"
],
"aspects": [
{
"term": [
"SquareTrade",
"3-Year",
"Computer",
"Accidental",
"Protection",
"Warranty"
],
"from": 4,
"to": 10,
"polarity": "positive"
},
{
"term": [
"AppleCare"
],
"from": 29,
"to": 30,
"polarity": "negative"
}
]
},
{
"token": [
"It",
"'s",
"light",
"and",
"easy",
"to",
"transport",
"."
],
"pos": [
"PRON",
"AUX",
"ADJ",
"CCONJ",
"ADJ",
"PART",
"VERB",
"PUNCT"
],
"head": [
3,
3,
0,
3,
3,
7,
5,
3
],
"deprel": [
"nsubj",
"cop",
"root",
"advmod",
"dep",
"aux",
"dep",
"punct"
],
"aspects": [
{
"term": [
"transport"
],
"from": 6,
"to": 7,
"polarity": "positive"
}
]
},
{
"token": [
"Once",
"you",
"get",
"past",
"learning",
"how",
"to",
"use",
"the",
"poorly",
"designed",
"Windows",
"8",
"Set",
"-",
"Up",
"you",
"may",
"feel",
"frustrated",
"."
],
"pos": [
"SCONJ",
"PRON",
"VERB",
"ADP",
"VERB",
"ADV",
"PART",
"VERB",
"DET",
"ADV",
"VERB",
"PROPN",
"NUM",
"NOUN",
"PUNCT",
"ADP",
"PRON",
"VERB",
"VERB",
"ADJ",
"PUNCT"
],
"head": [
3,
3,
19,
3,
4,
8,
8,
5,
11,
11,
8,
14,
14,
11,
14,
14,
19,
19,
0,
19,
19
],
"deprel": [
"mark",
"nsubj",
"advcl",
"dep",
"dep",
"advmod",
"aux",
"ccomp",
"dep",
"amod",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"nsubj",
"nsubj",
"root",
"acomp",
"punct"
],
"aspects": [
{
"term": [
"Windows",
"8",
"Set",
"-",
"Up"
],
"from": 11,
"to": 16,
"polarity": "negative"
}
]
},
{
"token": [
"It",
"'s",
"been",
"time",
"for",
"a",
"new",
"laptop",
",",
"and",
"the",
"only",
"debate",
"was",
"which",
"size",
"of",
"the",
"Mac",
"laptops",
",",
"and",
"whether",
"to",
"spring",
"for",
"the",
"retina",
"display",
"."
],
"pos": [
"PRON",
"AUX",
"AUX",
"NOUN",
"ADP",
"DET",
"ADJ",
"NOUN",
"PUNCT",
"CCONJ",
"DET",
"ADJ",
"NOUN",
"AUX",
"DET",
"NOUN",
"ADP",
"DET",
"PROPN",
"NOUN",
"PUNCT",
"CCONJ",
"SCONJ",
"PART",
"VERB",
"ADP",
"DET",
"PROPN",
"NOUN",
"PUNCT"
],
"head": [
4,
4,
4,
0,
4,
8,
8,
5,
4,
4,
13,
13,
16,
16,
16,
4,
16,
20,
17,
17,
16,
16,
24,
16,
24,
24,
29,
29,
26,
3
],
"deprel": [
"nsubj",
"nsubj",
"cop",
"root",
"prep",
"dep",
"amod",
"pobj",
"advmod",
"cc",
"nsubj",
"dep",
"nsubj",
"cop",
"dep",
"dep",
"prep",
"det",
"pobj",
"pobj",
"dep",
"dep",
"dep",
"dep",
"dep",
"prep",
"det",
"dep",
"pobj",
"dep"
],
"aspects": [
{
"term": [
"size"
],
"from": 15,
"to": 16,
"polarity": "neutral"
},
{
"term": [
"retina",
"display"
],
"from": 27,
"to": 29,
"polarity": "neutral"
}
]
},
{
"token": [
"The",
"reason",
"why",
"I",
"choose",
"apple",
"MacBook",
"because",
"of",
"their",
"design",
"and",
"the",
"aluminum",
"casing",
"."
],
"pos": [
"DET",
"NOUN",
"ADV",
"PRON",
"VERB",
"NOUN",
"PROPN",
"SCONJ",
"ADP",
"DET",
"NOUN",
"CCONJ",
"DET",
"NOUN",
"NOUN",
"PUNCT"
],
"head": [
2,
0,
5,
5,
2,
7,
5,
9,
5,
11,
9,
11,
14,
15,
12,
15
],
"deprel": [
"nsubj",
"root",
"nsubj",
"nsubj",
"dep",
"nsubj",
"dobj",
"mwe",
"prep",
"amod",
"pobj",
"prep",
"dep",
"nsubj",
"dep",
"punct"
],
"aspects": [
{
"term": [
"design"
],
"from": 10,
"to": 11,
"polarity": "positive"
},
{
"term": [
"aluminum",
"casing"
],
"from": 13,
"to": 15,
"polarity": "positive"
}
]
},
{
"token": [
"The",
"aluminum",
"body",
"sure",
"makes",
"it",
"stand",
"out",
"."
],
"pos": [
"DET",
"NOUN",
"NOUN",
"ADV",
"VERB",
"PRON",
"VERB",
"ADP",
"PUNCT"
],
"head": [
2,
4,
4,
0,
7,
7,
4,
7,
4
],
"deprel": [
"det",
"nsubj",
"nsubj",
"root",
"dep",
"nsubj",
"dep",
"dep",
"punct"
],
"aspects": [
{
"term": [
"aluminum",
"body"
],
"from": 1,
"to": 3,
"polarity": "positive"
}
]
},
{
"token": [
"It",
"is",
"very",
"easy",
"to",
"integrate",
"bluetooth",
"devices",
",",
"and",
"USB",
"devices",
"are",
"recognized",
"almost",
"instantly",
"."
],
"pos": [
"PRON",
"AUX",
"ADV",
"ADJ",
"PART",
"VERB",
"ADJ",
"NOUN",
"PUNCT",
"CCONJ",
"NOUN",
"NOUN",
"AUX",
"VERB",
"ADV",
"ADV",
"PUNCT"
],
"head": [
4,
4,
4,
0,
6,
4,
8,
6,
12,
12,
12,
14,
14,
4,
16,
14,
4
],
"deprel": [
"nsubj",
"cop",
"advmod",
"root",
"aux",
"xcomp",
"dep",
"dobj",
"nn",
"nn",
"dep",
"nsubjpass",
"auxpass",
"ccomp",
"dep",
"dep",
"dep"
],
"aspects": [
{
"term": [
"integrate",
"bluetooth",
"devices"
],
"from": 5,
"to": 8,
"polarity": "positive"
},
{
"term": [
"USB",
"devices"
],
"from": 10,
"to": 12,
"polarity": "positive"
}
]
},
{
"token": [
"And",
"the",
"fact",
"that",
"Apple",
"is",
"driving",
"the",
"13",
"''",
"RMBP",
"with",
"the",
"Intel4000",
"graphic",
"chip",
"seems",
"underpowered",
"(",
"to",
"me",
"."
],
"pos": [
"CCONJ",
"DET",
"NOUN",
"SCONJ",
"PROPN",
"AUX",
"VERB",
"DET",
"NUM",
"PUNCT",
"VERB",
"ADP",
"DET",
"PROPN",
"ADJ",
"NOUN",
"VERB",
"ADJ",
"PUNCT",
"ADP",
"PRON",
"PUNCT"
],
"head": [
18,
3,
1,
7,
7,
7,
3,
9,
7,
11,
6,
11,
16,
16,
16,
12,
18,
19,
0,
19,
20,
19
],
"deprel": [
"cc",
"det",
"dep",
"mark",
"nsubj",
"aux",
"ccomp",
"det",
"dep",
"punct",
"dep",
"prep",
"det",
"nn",
"amod",
"pobj",
"cop",
"dep",
"root",
"prep",
"pobj",
"advmod"
],
"aspects": [
{
"term": [
"Intel4000",
"graphic",
"chip"
],
"from": 13,
"to": 16,
"polarity": "negative"
}
]
},
{
"token": [
"Apple",
"removed",
"the",
"DVD",
"drive",
"Firewire",
"port",
"(",
"will",
"work",
"with",
"adapter",
")",
"and",
"put",
"the",
"SDXC",
"slot",
"in",
"a",
"silly",
"position",
"on",
"the",
"back",
"."
],
"pos": [
"PROPN",
"VERB",
"DET",
"NOUN",
"NOUN",
"PROPN",
"NOUN",
"PUNCT",
"VERB",
"VERB",
"ADP",
"NOUN",
"PUNCT",
"CCONJ",
"VERB",
"DET",
"PROPN",
"NOUN",
"ADP",
"DET",
"ADJ",
"NOUN",
"ADP",
"DET",
"NOUN",
"PUNCT"
],
"head": [
8,
8,
5,
5,
7,
5,
9,
7,
10,
13,
0,
13,
11,
13,
13,
18,
18,
15,
15,
22,
22,
19,
22,
25,
23,
24
],
"deprel": [
"dep",
"advmod",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"root",
"dep",
"dep",
"dep",
"dep",
"nn",
"nn",
"dep",
"prep",
"det",
"amod",
"pobj",
"prep",
"dep",
"pobj",
"punct"
],
"aspects": [
{
"term": [
"DVD",
"drive",
"Firewire",
"port"
],
"from": 3,
"to": 7,
"polarity": "neutral"
},
{
"term": [
"adapter"
],
"from": 11,
"to": 12,
"polarity": "neutral"
},
{
"term": [
"SDXC",
"slot"
],
"from": 16,
"to": 18,
"polarity": "negative"
}
]
},
{
"token": [
"The",
"durability",
"of",
"the",
"laptop",
"will",
"make",
"it",
"worth",
"the",
"money",
"."
],
"pos": [
"DET",
"NOUN",
"ADP",
"DET",
"NOUN",
"VERB",
"VERB",
"PRON",
"ADJ",
"DET",
"NOUN",
"PUNCT"
],
"head": [
2,
7,
2,
5,
3,
7,
0,
9,
7,
11,
9,
7
],
"deprel": [
"nn",
"nsubj",
"prep",
"det",
"pobj",
"aux",
"root",
"dep",
"xcomp",
"dep",
"dep",
"punct"
],
"aspects": [
{
"term": [
"durability"
],
"from": 1,
"to": 2,
"polarity": "positive"
}
]
},
{
"token": [
"Well",
"designed",
"and",
"fast",
"."
],
"pos": [
"INTJ",
"VERB",
"CCONJ",
"ADJ",
"PUNCT"
],
"head": [
2,
4,
4,
0,
4
],
"deprel": [
"amod",
"dep",
"dep",
"root",
"dep"
],
"aspects": [
{
"term": [
"designed"
],
"from": 1,
"to": 2,
"polarity": "positive"
}
]
},
{
"token": [
"But",
"I",
"was",
"completely",
"wrong",
",",
"this",
"computer",
"is",
"UNBELIEVABLE",
"amazing",
"and",
"easy",
"to",
"use",
"."
],
"pos": [
"CCONJ",
"PRON",
"AUX",
"ADV",
"ADJ",
"PUNCT",
"DET",
"NOUN",
"AUX",
"PROPN",
"ADJ",
"CCONJ",
"ADJ",
"PART",
"VERB",
"PUNCT"
],
"head": [
5,
5,
5,
5,
0,
5,
8,
10,
10,
5,
10,
13,
11,
15,
13,
5
],
"deprel": [
"cc",
"nsubj",
"cop",
"dep",
"root",
"punct",
"det",
"nsubj",
"cop",
"dep",
"amod",
"cc",
"amod",
"aux",
"ccomp",
"punct"
],
"aspects": [
{
"term": [
"use"
],
"from": 14,
"to": 15,
"polarity": "positive"
}
]
},
{
"token": [
"Exactly",
"as",
"posted",
"plus",
"a",
"great",
"value",
"."
],
"pos": [
"ADV",
"SCONJ",
"VERB",
"CCONJ",
"DET",
"ADJ",
"NOUN",
"PUNCT"
],
"head": [
3,
3,
0,
7,
7,
7,
3,
3
],
"deprel": [
"dep",
"dep",
"root",
"dep",
"dep",
"nn",
"dep",
"punct"
],
"aspects": [
{
"term": [
"value"
],
"from": 6,
"to": 7,
"polarity": "positive"
}
]
},
{
"token": [
"The",
"specs",
"are",
"pretty",
"good",
"too",
"."
],
"pos": [
"DET",
"NOUN",
"AUX",
"ADV",
"ADJ",
"ADV",
"PUNCT"
],
"head": [
2,
5,
5,
5,
0,
5,
5
],
"deprel": [
"det",
"nsubj",
"cop",
"amod",
"root",
"advmod",
"advmod"
],
"aspects": [
{
"term": [
"specs"
],
"from": 1,
"to": 2,
"polarity": "positive"
}
]
},
{
"token": [
"Apple",
"is",
"unmatched",
"in",
"product",
"quality",
",",
"aesthetics",
",",
"craftmanship",
",",
"and",
"customer",
"service",
"."
],
"pos": [
"PROPN",
"AUX",
"ADJ",
"ADP",
"NOUN",
"NOUN",
"PUNCT",
"NOUN",
"PUNCT",
"NOUN",
"PUNCT",
"CCONJ",
"NOUN",
"NOUN",
"PUNCT"
],
"head": [
2,
0,
2,
3,
4,
4,
6,
6,
8,
8,
10,
14,
14,
10,
2
],
"deprel": [
"nsubj",
"root",
"nsubj",
"prep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"nn",
"dep",
"dep"
],
"aspects": [
{
"term": [
"product",
"quality"
],
"from": 4,
"to": 6,
"polarity": "positive"
},
{
"term": [
",",
"aesthetics"
],
"from": 6,
"to": 8,
"polarity": "positive"
},
{
"term": [
",",
"craftmanship"
],
"from": 8,
"to": 10,
"polarity": "positive"
},
{
"term": [
"customer",
"service"
],
"from": 12,
"to": 14,
"polarity": "positive"
}
]
},
{
"token": [
"It",
"is",
"a",
"great",
"size",
"and",
"amazing",
"windows",
"8",
"included",
"!"
],
"pos": [
"PRON",
"AUX",
"DET",
"ADJ",
"NOUN",
"CCONJ",
"ADJ",
"NOUN",
"NUM",
"VERB",
"PUNCT"
],
"head": [
5,
5,
5,
5,
0,
5,
8,
5,
5,
5,
10
],
"deprel": [
"nsubj",
"cop",
"dep",
"dep",
"root",
"amod",
"amod",
"dep",
"dep",
"dep",
"dep"
],
"aspects": [
{
"term": [
"size"
],
"from": 4,
"to": 5,
"polarity": "positive"
},
{
"term": [
"windows",
"8"
],
"from": 7,
"to": 9,
"polarity": "positive"
}
]
},
{
"token": [
"I",
"do",
"not",
"like",
"too",
"much",
"Windows",
"8",
"."
],
"pos": [
"PRON",
"AUX",
"PART",
"VERB",
"ADV",
"ADJ",
"NOUN",
"NUM",
"PUNCT"
],
"head": [
4,
4,
4,
0,
6,
7,
4,
7,
4
],
"deprel": [
"nsubj",
"aux",
"neg",
"root",
"dep",
"dep",
"dep",
"dep",
"dep"
],
"aspects": [
{
"term": [
"Windows",
"8"
],
"from": 6,
"to": 8,
"polarity": "negative"
}
]
},
{
"token": [
"Startup",
"times",
"are",
"incredibly",
"long",
":",
"over",
"two",
"minutes",
"."
],
"pos": [
"NOUN",
"NOUN",
"AUX",
"ADV",
"ADJ",
"PUNCT",
"ADP",
"NUM",
"NOUN",
"PUNCT"
],
"head": [
2,
3,
4,
5,
0,
5,
9,
9,
6,
5
],
"deprel": [
"dep",
"nsubj",
"cop",
"dep",
"root",
"dep",
"advmod",
"num",
"dep",
"dep"
],
"aspects": [
{
"term": [
"Startup",
"times"
],
"from": 0,
"to": 2,
"polarity": "negative"
}
]
},
{
"token": [
"Also",
"stunning",
"colors",
"and",
"speedy"
],
"pos": [
"ADV",
"ADJ",
"NOUN",
"CCONJ",
"ADJ"
],
"head": [
2,
0,
2,
3,
3
],
"deprel": [
"dep",
"root",
"dep",
"cc",
"dep"
],
"aspects": [
{
"term": [
"colors"
],
"from": 2,
"to": 3,
"polarity": "positive"
}
]
},
{
"token": [
"great",
"price",
"free",
"shipping",
"what",
"else",
"can",
"i",
"ask",
"for",
"!",
"!"
],
"pos": [
"ADJ",
"NOUN",
"ADJ",
"NOUN",
"PRON",
"ADV",
"VERB",
"PRON",
"VERB",
"ADP",
"PUNCT",
"PUNCT"
],
"head": [
4,
4,
5,
3,
10,
5,
9,
9,
0,
9,
10,
10
],
"deprel": [
"amod",
"dep",
"amod",
"dep",
"dep",
"dep",
"dep",
"dep",
"root",
"prep",
"pobj",
"dep"
],
"aspects": [
{
"term": [
"price"
],
"from": 1,
"to": 2,
"polarity": "positive"
},
{
"term": [
"shipping"
],
"from": 3,
"to": 4,
"polarity": "positive"
}
]
},
{
"token": [
"This",
"mouse",
"is",
"terrific",
"."
],
"pos": [
"DET",
"NOUN",
"AUX",
"ADJ",
"PUNCT"
],
"head": [
2,
4,
4,
0,
4
],
"deprel": [
"det",
"nsubj",
"cop",
"root",
"ccomp"
],
"aspects": [
{
"term": [
"mouse"
],
"from": 1,
"to": 2,
"polarity": "positive"
}
]
},
{
"token": [
"It",
"is",
"really",
"thick",
"around",
"the",
"battery",
"."
],
"pos": [
"PRON",
"AUX",
"ADV",
"ADJ",
"ADP",
"DET",
"NOUN",
"PUNCT"
],
"head": [
2,
0,
2,
3,
4,
7,
5,
2
],
"deprel": [
"nsubj",
"root",
"dep",
"dep",
"advmod",
"det",
"pobj",
"punct"
],
"aspects": [
{
"term": [
"battery"
],
"from": 6,
"to": 7,
"polarity": "neutral"
}
]
},
{
"token": [
"And",
"windows",
"7",
"works",
"like",
"a",
"charm",
"."
],
"pos": [
"CCONJ",
"NOUN",
"NUM",
"NOUN",
"SCONJ",
"DET",
"NOUN",
"PUNCT"
],
"head": [
4,
4,
4,
0,
4,
7,
5,
4
],
"deprel": [
"cc",
"nsubj",
"nsubj",
"root",
"prep",
"dep",
"pobj",
"punct"
],
"aspects": [
{
"term": [
"windows",
"7"
],
"from": 1,
"to": 3,
"polarity": "positive"
}
]
},
{
"token": [
":",
"-RRB-",
"Great",
"product",
",",
"great",
"price",
",",
"great",
"delivery",
",",
"and",
"great",
"service",
"."
],
"pos": [
"PUNCT",
"PUNCT",
"ADJ",
"NOUN",
"PUNCT",
"ADJ",
"NOUN",
"PUNCT",
"ADJ",
"NOUN",
"PUNCT",
"CCONJ",
"ADJ",
"NOUN",
"PUNCT"
],
"head": [
4,
4,
4,
0,
4,
7,
4,
7,
10,
7,
10,
14,
14,
10,
4
],
"deprel": [
"dep",
"discourse",
"dep",
"root",
"dep",
"amod",
"dep",
"dep",
"amod",
"dep",
"dep",
"dep",
"dep",
"dep",
"punct"
],
"aspects": [
{
"term": [
"price"
],
"from": 5,
"to": 6,
"polarity": "positive"
},
{
"term": [
"delivery"
],
"from": 8,
"to": 9,
"polarity": "positive"
},
{
"term": [
"service"
],
"from": 12,
"to": 13,
"polarity": "positive"
}
]
},
{
"token": [
":]",
"It",
"arrived",
"so",
"fast",
"and",
"customer",
"service",
"was",
"great",
"."
],
"pos": [
"PUNCT",
"PRON",
"VERB",
"ADV",
"ADJ",
"CCONJ",
"NOUN",
"NOUN",
"AUX",
"ADJ",
"PUNCT"
],
"head": [
0,
1,
1,
5,
3,
8,
8,
10,
10,
1,
10
],
"deprel": [
"root",
"dep",
"dep",
"dep",
"dep",
"advmod",
"dep",
"nsubj",
"cop",
"dep",
"punct"
],
"aspects": [
{
"term": [
"customer",
"service"
],
"from": 6,
"to": 8,
"polarity": "positive"
}
]
},
{
"token": [
"tried",
"windows",
"8",
"and",
"hated",
"it",
"!",
"!",
"!"
],
"pos": [
"VERB",
"NOUN",
"NUM",
"CCONJ",
"VERB",
"PRON",
"PUNCT",
"PUNCT",
"PUNCT"
],
"head": [
2,
3,
0,
3,
4,
5,
5,
7,
8
],
"deprel": [
"dep",
"dep",
"root",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep"
],
"aspects": [
{
"term": [
"windows",
"8"
],
"from": 1,
"to": 3,
"polarity": "negative"
}
]
},
{
"token": [
"Set",
"up",
"was",
"a",
"breeze",
"."
],
"pos": [
"VERB",
"ADP",
"AUX",
"DET",
"NOUN",
"PUNCT"
],
"head": [
3,
1,
0,
5,
3,
3
],
"deprel": [
"nsubj",
"advmod",
"root",
"dep",
"dep",
"dep"
],
"aspects": [
{
"term": [
"Set",
"up"
],
"from": 0,
"to": 2,
"polarity": "positive"
}
]
},
{
"token": [
"But",
"I",
"do",
"NOT",
"like",
"Win8",
"."
],
"pos": [
"CCONJ",
"PRON",
"AUX",
"ADV",
"VERB",
"PROPN",
"PUNCT"
],
"head": [
4,
4,
4,
0,
4,
5,
4
],
"deprel": [
"cc",
"nsubj",
"dep",
"root",
"prep",
"dep",
"punct"
],
"aspects": [
{
"term": [
"Win8"
],
"from": 5,
"to": 6,
"polarity": "negative"
}
]
},
{
"token": [
"I",
"am",
"still",
"in",
"the",
"process",
"of",
"learning",
"about",
"its",
"features",
"."
],
"pos": [
"PRON",
"AUX",
"ADV",
"ADP",
"DET",
"NOUN",
"ADP",
"VERB",
"ADP",
"DET",
"NOUN",
"PUNCT"
],
"head": [
3,
3,
0,
3,
6,
4,
6,
7,
8,
11,
9,
3
],
"deprel": [
"nsubj",
"dep",
"root",
"prep",
"det",
"pobj",
"prep",
"pcomp",
"prep",
"dep",
"pobj",
"dep"
],
"aspects": [
{
"term": [
"features"
],
"from": 10,
"to": 11,
"polarity": "neutral"
}
]
},
{
"token": [
"I",
"had",
"the",
"same",
"reasons",
"as",
"most",
"PC",
"users",
":",
"the",
"price",
",",
"the",
"overbearing",
"restrictions",
"of",
"OSX",
"and",
"lack",
"of",
"support",
"for",
"games",
"."
],
"pos": [
"PRON",
"AUX",
"DET",
"ADJ",
"NOUN",
"SCONJ",
"ADJ",
"NOUN",
"NOUN",
"PUNCT",
"DET",
"NOUN",
"PUNCT",
"DET",
"NOUN",
"NOUN",
"ADP",
"PROPN",
"CCONJ",
"NOUN",
"ADP",
"NOUN",
"ADP",
"NOUN",
"PUNCT"
],
"head": [
2,
0,
5,
5,
2,
5,
8,
9,
6,
5,
12,
10,
12,
16,
12,
12,
16,
17,
16,
16,
20,
21,
22,
23,
2
],
"deprel": [
"nsubj",
"root",
"det",
"amod",
"dobj",
"prep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"amod",
"nn",
"amod",
"dep",
"prep",
"pobj",
"nn",
"amod",
"prep",
"pobj",
"prep",
"pobj",
"punct"
],
"aspects": [
{
"term": [
"price"
],
"from": 11,
"to": 12,
"polarity": "negative"
},
{
"term": [
"OSX"
],
"from": 17,
"to": 18,
"polarity": "negative"
},
{
"term": [
"support",
"for",
"games"
],
"from": 21,
"to": 24,
"polarity": "negative"
}
]
},
{
"token": [
"I",
"wanted",
"it",
"for",
"it",
"'s",
"mobility",
"and",
"man",
",",
"this",
"little",
"bad",
"boy",
"is",
"very",
"nice",
"."
],
"pos": [
"PRON",
"VERB",
"PRON",
"ADP",
"PRON",
"AUX",
"NOUN",
"CCONJ",
"NOUN",
"PUNCT",
"DET",
"ADJ",
"ADJ",
"NOUN",
"AUX",
"ADV",
"ADJ",
"PUNCT"
],
"head": [
2,
0,
2,
3,
7,
7,
4,
7,
4,
17,
13,
17,
12,
13,
17,
17,
2,
17
],
"deprel": [
"nsubj",
"root",
"dep",
"prep",
"nsubj",
"dep",
"pobj",
"cc",
"pcomp",
"punct",
"dep",
"nsubj",
"amod",
"dep",
"cop",
"advmod",
"dep",
"ccomp"
],
"aspects": [
{
"term": [
"mobility"
],
"from": 6,
"to": 7,
"polarity": "positive"
}
]
},
{
"token": [
"I",
"found",
"the",
"mini",
"to",
"be",
":",
"Exceptionally",
"easy",
"to",
"set",
"up"
],
"pos": [
"PRON",
"VERB",
"DET",
"NOUN",
"PART",
"AUX",
"PUNCT",
"ADV",
"ADJ",
"PART",
"VERB",
"ADP"
],
"head": [
2,
0,
4,
2,
2,
2,
6,
9,
7,
11,
9,
11
],
"deprel": [
"nsubj",
"root",
"det",
"dep",
"prep",
"dep",
"dep",
"nsubj",
"dep",
"dep",
"dep",
"prt"
],
"aspects": [
{
"term": [
"set",
"up"
],
"from": 10,
"to": 12,
"polarity": "positive"
}
]
},
{
"token": [
"Having",
"USB3",
"is",
"why",
"I",
"bought",
"this",
"Mini",
"."
],
"pos": [
"VERB",
"PROPN",
"AUX",
"ADV",
"PRON",
"VERB",
"DET",
"PROPN",
"PUNCT"
],
"head": [
2,
3,
0,
3,
6,
4,
8,
6,
3
],
"deprel": [
"dep",
"dep",
"root",
"nsubj",
"nsubj",
"dep",
"dep",
"dep",
"punct"
],
"aspects": [
{
"term": [
"USB3"
],
"from": 1,
"to": 2,
"polarity": "positive"
}
]
},
{
"token": [
"The",
"sound",
"is",
"nice",
"and",
"loud",
";",
"I",
"do",
"n't",
"have",
"any",
"problems",
"with",
"hearing",
"anything",
"."
],
"pos": [
"DET",
"NOUN",
"AUX",
"ADJ",
"CCONJ",
"ADJ",
"PUNCT",
"PRON",
"AUX",
"PART",
"AUX",
"DET",
"NOUN",
"ADP",
"VERB",
"PRON",
"PUNCT"
],
"head": [
2,
4,
4,
0,
4,
4,
4,
11,
11,
11,
4,
13,
11,
13,
14,
15,
15
],
"deprel": [
"det",
"nsubj",
"cop",
"root",
"advmod",
"dep",
"punct",
"nsubj",
"aux",
"neg",
"parataxis",
"dep",
"dobj",
"prep",
"pobj",
"amod",
"dep"
],
"aspects": [
{
"term": [
"sound"
],
"from": 1,
"to": 2,
"polarity": "positive"
}
]
},
{
"token": [
"It",
"is",
"very",
"slim",
",",
"the",
"track",
"pad",
"is",
"very",
"much",
"impressed",
"with",
"me",
"."
],
"pos": [
"PRON",
"AUX",
"ADV",
"ADJ",
"PUNCT",
"DET",
"NOUN",
"NOUN",
"AUX",
"ADV",
"ADV",
"ADJ",
"ADP",
"PRON",
"PUNCT"
],
"head": [
4,
4,
4,
0,
4,
8,
8,
12,
12,
11,
12,
4,
12,
13,
12
],
"deprel": [
"nsubj",
"cop",
"advmod",
"root",
"punct",
"det",
"amod",
"nsubj",
"cop",
"advmod",
"dep",
"ccomp",
"prep",
"pobj",
"advmod"
],
"aspects": [
{
"term": [
"track",
"pad"
],
"from": 6,
"to": 8,
"polarity": "positive"
}
]
},
{
"token": [
"The",
"settings",
"are",
"not",
"user",
"-",
"friendly",
"either",
"."
],
"pos": [
"DET",
"NOUN",
"AUX",
"PART",
"NOUN",
"PUNCT",
"ADJ",
"ADV",
"PUNCT"
],
"head": [
2,
7,
7,
7,
7,
7,
0,
7,
7
],
"deprel": [
"num",
"nsubj",
"cop",
"neg",
"dep",
"punct",
"root",
"advmod",
"dep"
],
"aspects": [
{
"term": [
"settings"
],
"from": 1,
"to": 2,
"polarity": "negative"
}
]
},
{
"token": [
"Thank",
"goodness",
"for",
"OpenOffice",
"!"
],
"pos": [
"VERB",
"NOUN",
"ADP",
"PROPN",
"PUNCT"
],
"head": [
0,
1,
1,
3,
3
],
"deprel": [
"root",
"advmod",
"prep",
"pobj",
"pobj"
],
"aspects": [
{
"term": [
"OpenOffice"
],
"from": 3,
"to": 4,
"polarity": "positive"
}
]
},
{
"token": [
"Awesome",
"form",
"factor",
",",
"great",
"battery",
"life",
",",
"wonderful",
"UX",
"."
],
"pos": [
"ADJ",
"NOUN",
"NOUN",
"PUNCT",
"ADJ",
"NOUN",
"NOUN",
"PUNCT",
"ADJ",
"PROPN",
"PUNCT"
],
"head": [
0,
3,
1,
3,
7,
7,
3,
7,
10,
7,
10
],
"deprel": [
"root",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"punct"
],
"aspects": [
{
"term": [
"form",
"factor"
],
"from": 1,
"to": 3,
"polarity": "positive"
},
{
"term": [
"battery",
"life"
],
"from": 5,
"to": 7,
"polarity": "positive"
},
{
"term": [
"UX"
],
"from": 9,
"to": 10,
"polarity": "positive"
}
]
},
{
"token": [
"i",
"love",
"the",
"keyboard",
"and",
"the",
"screen",
"."
],
"pos": [
"PRON",
"VERB",
"DET",
"NOUN",
"CCONJ",
"DET",
"NOUN",
"PUNCT"
],
"head": [
2,
0,
4,
2,
2,
7,
2,
2
],
"deprel": [
"dep",
"root",
"det",
"dep",
"dep",
"det",
"dep",
"punct"
],
"aspects": [
{
"term": [
"keyboard"
],
"from": 3,
"to": 4,
"polarity": "positive"
},
{
"term": [
"screen"
],
"from": 6,
"to": 7,
"polarity": "positive"
}
]
},
{
"token": [
"However",
",",
"there",
"are",
"MAJOR",
"issues",
"with",
"the",
"touchpad",
"which",
"render",
"the",
"device",
"nearly",
"useless",
"."
],
"pos": [
"ADV",
"PUNCT",
"PRON",
"AUX",
"PROPN",
"NOUN",
"ADP",
"DET",
"NOUN",
"DET",
"VERB",
"DET",
"NOUN",
"ADV",
"ADJ",
"PUNCT"
],
"head": [
4,
4,
4,
0,
4,
4,
6,
9,
7,
11,
9,
13,
11,
15,
4,
4
],
"deprel": [
"advmod",
"dep",
"expl",
"root",
"nsubj",
"nsubj",
"prep",
"det",
"pobj",
"nsubj",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep"
],
"aspects": [
{
"term": [
"touchpad"
],
"from": 8,
"to": 9,
"polarity": "negative"
}
]
},
{
"token": [
"I",
"'",
"ve",
"already",
"upgraded",
"o",
"Mavericks",
"and",
"I",
"am",
"impressed",
"with",
"everything",
"about",
"this",
"computer",
"."
],
"pos": [
"PRON",
"PUNCT",
"PROPN",
"ADV",
"VERB",
"INTJ",
"PROPN",
"CCONJ",
"PRON",
"AUX",
"ADJ",
"ADP",
"PRON",
"ADP",
"DET",
"NOUN",
"PUNCT"
],
"head": [
3,
3,
5,
5,
0,
7,
5,
5,
11,
11,
5,
11,
12,
13,
16,
14,
5
],
"deprel": [
"nsubj",
"punct",
"nsubj",
"advmod",
"root",
"dep",
"dep",
"cc",
"nsubj",
"cop",
"dep",
"prep",
"pobj",
"prep",
"amod",
"pobj",
"advmod"
],
"aspects": [
{
"term": [
"Mavericks"
],
"from": 5,
"to": 6,
"polarity": "positive"
}
]
},
{
"token": [
"Not",
"as",
"fast",
"as",
"I",
"would",
"have",
"expect",
"for",
"an",
"i5",
"."
],
"pos": [
"PART",
"ADV",
"ADV",
"SCONJ",
"PRON",
"VERB",
"AUX",
"VERB",
"ADP",
"DET",
"NOUN",
"PUNCT"
],
"head": [
8,
3,
1,
3,
8,
8,
8,
0,
8,
11,
9,
8
],
"deprel": [
"dep",
"dep",
"dep",
"prep",
"nsubj",
"aux",
"aux",
"root",
"prep",
"det",
"pobj",
"punct"
],
"aspects": [
{
"term": [
"i5"
],
"from": 10,
"to": 11,
"polarity": "negative"
}
]
},
{
"token": [
"thanks",
"for",
"great",
"service",
"and",
"shipping",
"!"
],
"pos": [
"NOUN",
"ADP",
"ADJ",
"NOUN",
"CCONJ",
"NOUN",
"PUNCT"
],
"head": [
0,
1,
4,
2,
4,
4,
2
],
"deprel": [
"root",
"prep",
"dep",
"pobj",
"prep",
"conj",
"pobj"
],
"aspects": [
{
"term": [
"service"
],
"from": 3,
"to": 4,
"polarity": "positive"
},
{
"term": [
"shipping"
],
"from": 5,
"to": 6,
"polarity": "positive"
}
]
},
{
"token": [
"The",
"performance",
"seems",
"quite",
"good",
",",
"and",
"built",
"-",
"in",
"applications",
"like",
"iPhoto",
"work",
"great",
"with",
"my",
"phone",
"and",
"camera",
"."
],
"pos": [
"DET",
"NOUN",
"VERB",
"ADV",
"ADJ",
"PUNCT",
"CCONJ",
"VERB",
"PUNCT",
"ADP",
"NOUN",
"SCONJ",
"PROPN",
"NOUN",
"ADJ",
"ADP",
"DET",
"NOUN",
"CCONJ",
"NOUN",
"PUNCT"
],
"head": [
2,
5,
5,
5,
0,
5,
5,
6,
10,
8,
10,
11,
12,
12,
12,
12,
18,
16,
18,
18,
5
],
"deprel": [
"det",
"nsubj",
"cop",
"dep",
"root",
"dep",
"dep",
"dep",
"dep",
"dep",
"pobj",
"prep",
"dep",
"dep",
"npadvmod",
"prep",
"dep",
"pobj",
"amod",
"dep",
"dep"
],
"aspects": [
{
"term": [
"performance"
],
"from": 1,
"to": 2,
"polarity": "positive"
},
{
"term": [
"built",
"-",
"in",
"applications"
],
"from": 7,
"to": 11,
"polarity": "positive"
},
{
"term": [
"iPhoto"
],
"from": 10,
"to": 11,
"polarity": "positive"
}
]
},
{
"token": [
"I",
"did",
"swap",
"out",
"the",
"hard",
"drive",
"for",
"a",
"Samsung",
"830",
"SSD",
"which",
"I",
"highly",
"recommend",
"."
],
"pos": [
"PRON",
"AUX",
"VERB",
"ADP",
"DET",
"ADJ",
"NOUN",
"ADP",
"DET",
"PROPN",
"NUM",
"NOUN",
"DET",
"PRON",
"ADV",
"VERB",
"PUNCT"
],
"head": [
3,
3,
0,
3,
7,
7,
4,
3,
11,
11,
8,
11,
16,
16,
16,
12,
16
],
"deprel": [
"nsubj",
"aux",
"root",
"advmod",
"dep",
"amod",
"dep",
"prep",
"dep",
"dep",
"pobj",
"dep",
"dep",
"nsubj",
"dep",
"rcmod",
"dep"
],
"aspects": [
{
"term": [
"hard",
"drive"
],
"from": 5,
"to": 7,
"polarity": "neutral"
},
{
"term": [
"Samsung",
"830",
"SSD"
],
"from": 9,
"to": 12,
"polarity": "positive"
}
]
},
{
"token": [
"Starts",
"up",
"in",
"a",
"hurry",
"and",
"everything",
"is",
"ready",
"to",
"go",
"."
],
"pos": [
"VERB",
"ADP",
"ADP",
"DET",
"NOUN",
"CCONJ",
"PRON",
"AUX",
"ADJ",
"PART",
"VERB",
"PUNCT"
],
"head": [
9,
1,
1,
5,
3,
5,
9,
9,
0,
11,
9,
9
],
"deprel": [
"nsubj",
"advmod",
"prep",
"dep",
"pobj",
"prep",
"nsubj",
"cop",
"root",
"aux",
"xcomp",
"punct"
],
"aspects": [
{
"term": [
"Starts",
"up"
],
"from": 0,
"to": 2,
"polarity": "positive"
}
]
},
{
"token": [
"Yes",
",",
"that",
"'s",
"a",
"good",
"thing",
",",
"but",
"it",
"'s",
"made",
"from",
"aluminum",
"that",
"scratches",
"easily",
"."
],
"pos": [
"INTJ",
"PUNCT",
"DET",
"AUX",
"DET",
"ADJ",
"NOUN",
"PUNCT",
"CCONJ",
"PRON",
"AUX",
"VERB",
"ADP",
"NOUN",
"DET",
"VERB",
"ADV",
"PUNCT"
],
"head": [
7,
1,
7,
7,
7,
7,
0,
7,
7,
12,
12,
7,
12,
13,
17,
17,
14,
1
],
"deprel": [
"discourse",
"dep",
"nsubj",
"cop",
"dep",
"amod",
"root",
"punct",
"cc",
"nsubjpass",
"auxpass",
"ccomp",
"prep",
"pobj",
"nsubj",
"nsubj",
"rcmod",
"punct"
],
"aspects": [
{
"term": [
"aluminum"
],
"from": 13,
"to": 14,
"polarity": "negative"
}
]
},
{
"token": [
"Quick",
"and",
"has",
"built",
"in",
"virus",
"control",
"."
],
"pos": [
"ADJ",
"CCONJ",
"AUX",
"VERB",
"ADP",
"NOUN",
"NOUN",
"PUNCT"
],
"head": [
4,
4,
4,
0,
4,
5,
5,
4
],
"deprel": [
"nsubj",
"nsubj",
"aux",
"root",
"prep",
"dep",
"dep",
"dep"
],
"aspects": [
{
"term": [
"built",
"in",
"virus",
"control"
],
"from": 3,
"to": 7,
"polarity": "positive"
}
]
},
{
"token": [
"Took",
"a",
"long",
"time",
"trying",
"to",
"decide",
"between",
"one",
"with",
"retina",
"display",
"and",
"one",
"without",
"."
],
"pos": [
"VERB",
"DET",
"ADJ",
"NOUN",
"VERB",
"PART",
"VERB",
"ADP",
"NUM",
"ADP",
"NOUN",
"NOUN",
"CCONJ",
"NOUN",
"ADP",
"PUNCT"
],
"head": [
0,
4,
4,
5,
1,
7,
5,
9,
7,
9,
12,
10,
12,
12,
14,
15
],
"deprel": [
"root",
"dep",
"dep",
"dep",
"amod",
"aux",
"xcomp",
"amod",
"nsubj",
"prep",
"dep",
"pobj",
"dep",
"dep",
"dep",
"dep"
],
"aspects": [
{
"term": [
"retina",
"display"
],
"from": 10,
"to": 12,
"polarity": "neutral"
}
]
},
{
"token": [
"I",
"was",
"also",
"informed",
"that",
"the",
"components",
"of",
"the",
"Mac",
"Book",
"were",
"dirty",
"."
],
"pos": [
"PRON",
"AUX",
"ADV",
"VERB",
"SCONJ",
"DET",
"NOUN",
"ADP",
"DET",
"PROPN",
"PROPN",
"AUX",
"ADJ",
"PUNCT"
],
"head": [
3,
4,
4,
0,
13,
7,
13,
7,
11,
11,
8,
13,
4,
4
],
"deprel": [
"nsubjpass",
"auxpass",
"advmod",
"root",
"mark",
"amod",
"nsubj",
"prep",
"det",
"dep",
"pobj",
"cop",
"ccomp",
"advmod"
],
"aspects": [
{
"term": [
"components"
],
"from": 6,
"to": 7,
"polarity": "negative"
}
]
},
{
"token": [
"the",
"hardware",
"problems",
"have",
"been",
"so",
"bad",
",",
"i",
"ca",
"n't",
"wait",
"till",
"it",
"completely",
"dies",
"in",
"3",
"years",
",",
"TOPS",
"!"
],
"pos": [
"DET",
"NOUN",
"NOUN",
"AUX",
"AUX",
"ADV",
"ADJ",
"PUNCT",
"PRON",
"VERB",
"PART",
"VERB",
"SCONJ",
"PRON",
"ADV",
"VERB",
"ADP",
"NUM",
"NOUN",
"PUNCT",
"PROPN",
"PUNCT"
],
"head": [
3,
3,
7,
7,
7,
7,
0,
7,
12,
12,
12,
7,
12,
16,
16,
12,
16,
19,
17,
19,
19,
21
],
"deprel": [
"amod",
"dep",
"nsubj",
"aux",
"cop",
"dep",
"root",
"punct",
"nsubj",
"dep",
"neg",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"amod",
"dep",
"prep",
"dep",
"dep"
],
"aspects": [
{
"term": [
"hardware"
],
"from": 1,
"to": 2,
"polarity": "negative"
}
]
},
{
"token": [
"It",
"'s",
"so",
"nice",
"that",
"the",
"battery",
"last",
"so",
"long",
"and",
"that",
"this",
"machine",
"has",
"the",
"snow",
"lion",
"!"
],
"pos": [
"PRON",
"AUX",
"ADV",
"ADJ",
"SCONJ",
"DET",
"NOUN",
"ADJ",
"ADV",
"ADV",
"CCONJ",
"SCONJ",
"DET",
"NOUN",
"AUX",
"DET",
"NOUN",
"NOUN",
"PUNCT"
],
"head": [
4,
4,
4,
0,
4,
7,
10,
10,
10,
4,
10,
15,
14,
15,
10,
17,
18,
15,
18
],
"deprel": [
"nsubj",
"cop",
"advmod",
"root",
"mark",
"det",
"nsubj",
"dep",
"advmod",
"dep",
"dep",
"mark",
"dep",
"nsubj",
"dep",
"dep",
"dep",
"dep",
"dep"
],
"aspects": [
{
"term": [
"battery"
],
"from": 6,
"to": 7,
"polarity": "positive"
},
{
"term": [
"snow",
"lion"
],
"from": 16,
"to": 18,
"polarity": "positive"
}
]
},
{
"token": [
"HOWEVER",
"I",
"chose",
"two",
"day",
"shipping",
"and",
"it",
"took",
"over",
"a",
"week",
"to",
"arrive",
"."
],
"pos": [
"ADV",
"PRON",
"VERB",
"NUM",
"NOUN",
"NOUN",
"CCONJ",
"PRON",
"VERB",
"ADP",
"DET",
"NOUN",
"PART",
"VERB",
"PUNCT"
],
"head": [
0,
3,
1,
5,
3,
3,
3,
9,
3,
11,
12,
9,
14,
9,
13
],
"deprel": [
"root",
"nsubj",
"dep",
"dep",
"dep",
"dep",
"dep",
"nsubj",
"dep",
"quantmod",
"num",
"tmod",
"dep",
"dep",
"dep"
],
"aspects": [
{
"term": [
"shipping"
],
"from": 5,
"to": 6,
"polarity": "negative"
}
]
},
{
"token": [
"it",
"'s",
"exactly",
"what",
"i",
"wanted",
",",
"and",
"it",
"has",
"all",
"the",
"new",
"features",
"and",
"whatnot",
"."
],
"pos": [
"PRON",
"AUX",
"ADV",
"PRON",
"PRON",
"VERB",
"PUNCT",
"CCONJ",
"PRON",
"AUX",
"DET",
"DET",
"ADJ",
"NOUN",
"CCONJ",
"NOUN",
"PUNCT"
],
"head": [
2,
0,
2,
6,
6,
3,
6,
2,
10,
6,
14,
14,
14,
10,
16,
14,
2
],
"deprel": [
"nsubj",
"root",
"advmod",
"advmod",
"nsubj",
"dep",
"advmod",
"advmod",
"nsubj",
"dep",
"dep",
"dep",
"amod",
"dobj",
"dep",
"dep",
"punct"
],
"aspects": [
{
"term": [
"features"
],
"from": 13,
"to": 14,
"polarity": "positive"
}
]
},
{
"token": [
"Can",
"you",
"buy",
"any",
"laptop",
"that",
"matches",
"the",
"quality",
"of",
"a",
"MacBook",
"?"
],
"pos": [
"VERB",
"PRON",
"VERB",
"DET",
"NOUN",
"DET",
"VERB",
"DET",
"NOUN",
"ADP",
"DET",
"PROPN",
"PUNCT"
],
"head": [
3,
3,
0,
5,
3,
7,
5,
9,
7,
9,
12,
10,
10
],
"deprel": [
"aux",
"nsubj",
"root",
"amod",
"dobj",
"nsubj",
"dep",
"dep",
"dep",
"prep",
"dep",
"pobj",
"pobj"
],
"aspects": [
{
"term": [
"quality"
],
"from": 8,
"to": 9,
"polarity": "positive"
}
]
},
{
"token": [
"It",
"feels",
"cheap",
",",
"the",
"keyboard",
"is",
"not",
"very",
"sensitive",
"."
],
"pos": [
"PRON",
"VERB",
"ADJ",
"PUNCT",
"DET",
"NOUN",
"AUX",
"PART",
"ADV",
"ADJ",
"PUNCT"
],
"head": [
0,
1,
2,
10,
6,
10,
10,
10,
10,
2,
10
],
"deprel": [
"root",
"dep",
"acomp",
"advmod",
"amod",
"nsubj",
"cop",
"neg",
"advmod",
"acomp",
"punct"
],
"aspects": [
{
"term": [
"keyboard"
],
"from": 5,
"to": 6,
"polarity": "negative"
}
]
},
{
"token": [
"Though",
"please",
"note",
"that",
"sometimes",
"it",
"crashes",
",",
"and",
"the",
"sound",
"quality",
"is",
"nt",
"superb",
"."
],
"pos": [
"SCONJ",
"INTJ",
"VERB",
"SCONJ",
"ADV",
"PRON",
"VERB",
"PUNCT",
"CCONJ",
"DET",
"ADJ",
"NOUN",
"AUX",
"PART",
"ADJ",
"PUNCT"
],
"head": [
3,
3,
14,
7,
7,
7,
3,
7,
14,
12,
12,
14,
14,
0,
14,
15
],
"deprel": [
"mark",
"nsubj",
"advcl",
"advmod",
"nsubj",
"nsubj",
"dep",
"punct",
"cc",
"dep",
"nsubj",
"nsubj",
"cop",
"root",
"amod",
"dep"
],
"aspects": [
{
"term": [
"sound",
"quality"
],
"from": 10,
"to": 12,
"polarity": "negative"
}
]
},
{
"token": [
"It",
"is",
"very",
"easy",
"to",
"navigate",
"even",
"for",
"a",
"novice",
"."
],
"pos": [
"PRON",
"AUX",
"ADV",
"ADJ",
"PART",
"VERB",
"ADV",
"ADP",
"DET",
"NOUN",
"PUNCT"
],
"head": [
4,
4,
4,
0,
6,
4,
8,
6,
10,
8,
4
],
"deprel": [
"nsubj",
"cop",
"advmod",
"root",
"aux",
"ccomp",
"dep",
"prep",
"dep",
"pobj",
"punct"
],
"aspects": [
{
"term": [
"navigate"
],
"from": 5,
"to": 6,
"polarity": "positive"
}
]
},
{
"token": [
"Does",
"everything",
"I",
"need",
"it",
"to",
",",
"has",
"a",
"wonderful",
"battery",
"life",
"and",
"I",
"could",
"n't",
"be",
"happier",
"."
],
"pos": [
"AUX",
"PRON",
"PRON",
"VERB",
"PRON",
"PART",
"PUNCT",
"AUX",
"DET",
"ADJ",
"NOUN",
"NOUN",
"CCONJ",
"PRON",
"VERB",
"PART",
"AUX",
"ADJ",
"PUNCT"
],
"head": [
4,
4,
4,
0,
6,
4,
6,
6,
11,
11,
8,
8,
12,
18,
18,
18,
18,
4,
18
],
"deprel": [
"aux",
"nsubj",
"nsubj",
"root",
"dep",
"dep",
"dep",
"dep",
"dep",
"amod",
"dep",
"dep",
"cc",
"nsubj",
"aux",
"aux",
"cop",
"ccomp",
"punct"
],
"aspects": [
{
"term": [
"battery",
"life"
],
"from": 10,
"to": 12,
"polarity": "positive"
}
]
},
{
"token": [
"Great",
"Performance",
"and",
"Quality",
"."
],
"pos": [
"PROPN",
"PROPN",
"CCONJ",
"PROPN",
"PUNCT"
],
"head": [
2,
4,
2,
0,
4
],
"deprel": [
"nn",
"nsubj",
"cc",
"root",
"punct"
],
"aspects": [
{
"term": [
"Performance"
],
"from": 1,
"to": 2,
"polarity": "positive"
},
{
"term": [
"Quality"
],
"from": 3,
"to": 4,
"polarity": "positive"
}
]
},
{
"token": [
"I",
"used",
"windows",
"XP",
",",
"windows",
"Vista",
",",
"and",
"Windows",
"7",
"extensively",
"."
],
"pos": [
"PRON",
"VERB",
"PROPN",
"PROPN",
"PUNCT",
"NOUN",
"PROPN",
"PUNCT",
"CCONJ",
"PROPN",
"NUM",
"ADV",
"PUNCT"
],
"head": [
2,
0,
2,
2,
4,
4,
4,
4,
4,
4,
10,
11,
11
],
"deprel": [
"nsubj",
"root",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"cc",
"dep",
"dep",
"advmod",
"dep"
],
"aspects": [
{
"term": [
"windows",
"XP"
],
"from": 2,
"to": 4,
"polarity": "neutral"
},
{
"term": [
",",
"windows",
"Vista"
],
"from": 4,
"to": 7,
"polarity": "neutral"
},
{
"term": [
"Windows",
"7"
],
"from": 9,
"to": 11,
"polarity": "neutral"
}
]
},
{
"token": [
"I",
"did",
"add",
"a",
"SSD",
"drive",
"and",
"memory"
],
"pos": [
"PRON",
"AUX",
"VERB",
"DET",
"NOUN",
"NOUN",
"CCONJ",
"NOUN"
],
"head": [
3,
3,
0,
6,
6,
3,
6,
6
],
"deprel": [
"nsubj",
"aux",
"root",
"dep",
"dep",
"dobj",
"cc",
"conj"
],
"aspects": [
{
"term": [
"SSD",
"drive"
],
"from": 4,
"to": 6,
"polarity": "neutral"
},
{
"term": [
"memory"
],
"from": 7,
"to": 8,
"polarity": "neutral"
}
]
},
{
"token": [
"On",
"start",
"up",
"it",
"asks",
"endless",
"questions",
"just",
"so",
"itune",
"can",
"sell",
"you",
"more",
"of",
"their",
"products",
"."
],
"pos": [
"ADP",
"VERB",
"ADP",
"PRON",
"VERB",
"ADJ",
"NOUN",
"ADV",
"ADV",
"PROPN",
"VERB",
"VERB",
"PRON",
"ADJ",
"ADP",
"DET",
"NOUN",
"PUNCT"
],
"head": [
5,
5,
2,
2,
0,
7,
5,
9,
10,
12,
12,
7,
14,
12,
14,
17,
15,
5
],
"deprel": [
"nsubj",
"dep",
"advmod",
"prep",
"root",
"amod",
"dep",
"advmod",
"amod",
"nsubj",
"dep",
"ccomp",
"dep",
"dobj",
"prep",
"amod",
"pobj",
"punct"
],
"aspects": [
{
"term": [
"start",
"up"
],
"from": 1,
"to": 3,
"polarity": "negative"
},
{
"term": [
"itune"
],
"from": 9,
"to": 10,
"polarity": "negative"
}
]
},
{
"token": [
"I",
"Have",
"been",
"a",
"Pc",
"user",
"for",
"a",
"very",
"long",
"time",
"now",
"but",
"I",
"will",
"get",
"used",
"to",
"this",
"new",
"OS",
"."
],
"pos": [
"PRON",
"AUX",
"AUX",
"DET",
"PROPN",
"NOUN",
"ADP",
"DET",
"ADV",
"ADJ",
"NOUN",
"ADV",
"CCONJ",
"PRON",
"VERB",
"AUX",
"ADJ",
"ADP",
"DET",
"ADJ",
"NOUN",
"PUNCT"
],
"head": [
6,
6,
6,
6,
6,
0,
6,
7,
10,
7,
10,
11,
17,
17,
17,
17,
6,
17,
18,
21,
18,
6
],
"deprel": [
"nsubj",
"nsubj",
"cop",
"det",
"nn",
"root",
"prep",
"pobj",
"advmod",
"pobj",
"dep",
"dep",
"mark",
"nsubj",
"aux",
"dep",
"dep",
"dep",
"dep",
"amod",
"pobj",
"punct"
],
"aspects": [
{
"term": [
"OS"
],
"from": 20,
"to": 21,
"polarity": "neutral"
}
]
},
{
"token": [
"One",
"more",
"thing",
",",
"this",
"mac",
"does",
"NOT",
"come",
"with",
"restore",
"disks",
"and",
"I",
"am",
"not",
"sure",
"if",
"you",
"can",
"make",
"them",
"direct",
"from",
"the",
"mac",
"like",
"you",
"can",
"with",
"newer",
"PC",
"'s",
",",
"also",
"the",
"charging",
"cables",
"are",
"made",
"of",
"the",
"same",
"cheap",
"material",
"as",
"the",
"iPhone",
"/",
"iPod",
"touch",
"cables",
"."
],
"pos": [
"NUM",
"ADJ",
"NOUN",
"PUNCT",
"DET",
"PROPN",
"AUX",
"ADV",
"VERB",
"ADP",
"NOUN",
"NOUN",
"CCONJ",
"PRON",
"AUX",
"PART",
"ADJ",
"SCONJ",
"PRON",
"VERB",
"VERB",
"PRON",
"ADJ",
"ADP",
"DET",
"PROPN",
"SCONJ",
"PRON",
"VERB",
"ADP",
"ADJ",
"PROPN",
"PART",
"PUNCT",
"ADV",
"DET",
"VERB",
"NOUN",
"AUX",
"VERB",
"ADP",
"DET",
"ADJ",
"ADJ",
"NOUN",
"SCONJ",
"DET",
"PROPN",
"SYM",
"PROPN",
"NOUN",
"NOUN",
"PUNCT"
],
"head": [
3,
3,
0,
3,
6,
3,
8,
3,
8,
9,
10,
10,
8,
17,
17,
17,
8,
21,
21,
21,
17,
21,
21,
23,
27,
27,
31,
31,
31,
23,
30,
31,
31,
31,
40,
38,
38,
40,
40,
17,
40,
45,
44,
45,
41,
45,
48,
49,
50,
52,
52,
46,
50
],
"deprel": [
"advmod",
"advmod",
"root",
"advmod",
"dep",
"dep",
"dep",
"dep",
"advmod",
"prep",
"pobj",
"pobj",
"prep",
"dep",
"dep",
"dep",
"dep",
"mark",
"nsubj",
"dep",
"ccomp",
"dobj",
"xcomp",
"prep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"dep",
"possessive",
"amod",
"dep",
"nn",
"amod",
"nsubjpass",
"auxpass",
"ccomp",
"prep",
"det",
"dep",
"amod",
"pobj",
"prep",
"amod",
"dep",
"dep",
"dep",
"amod",
"pobj",
"amod"
],
"aspects": [
{
"term": [
"restore",
"disks"
],
"from": 10,
"to": 12,
"polarity": "negative"
},
{
"term": [
"charging",
"cables"
],
"from": 36,
"to": 38,
"polarity": "negative"
},
{
"term": [
"material"
],
"from": 44,
"to": 45,
"polarity": "negative"
}
]
},
{
"token": [
"I",
"bought",
"it",
"to",
"my",
"son",
"who",
"uses",
"it",
"for",
"graphic",
"design",
"."
],
"pos": [
"PRON",
"VERB",
"PRON",
"ADP",
"DET",
"NOUN",
"PRON",
"VERB",
gitextract_bclkhkuw/ ├── DualGCN/ │ ├── build_vocab.sh │ ├── data_utils.py │ ├── dataset/ │ │ ├── Laptops_allennlp/ │ │ │ ├── test.json │ │ │ └── train.json │ │ ├── Laptops_corenlp/ │ │ │ ├── test.json │ │ │ └── train.json │ │ ├── Laptops_stanza/ │ │ │ ├── test.json │ │ │ └── train.json │ │ ├── Restaurants_allennlp/ │ │ │ ├── test.json │ │ │ └── train.json │ │ ├── Restaurants_corenlp/ │ │ │ ├── test.json │ │ │ └── train.json │ │ ├── Restaurants_stanza/ │ │ │ ├── test.json │ │ │ └── train.json │ │ ├── Tweets_allennlp/ │ │ │ ├── test.json │ │ │ └── train.json │ │ ├── Tweets_corenlp/ │ │ │ ├── test.json │ │ │ └── train.json │ │ └── Tweets_stanza/ │ │ ├── test.json │ │ └── train.json │ ├── glove/ │ │ └── readme.md │ ├── layers.py │ ├── models/ │ │ ├── atae_lstm.py │ │ ├── dualgcn.py │ │ ├── dualgcn_bert.py │ │ ├── ian.py │ │ ├── semgcn.py │ │ └── syngcn.py │ ├── prepare_vocab.py │ ├── run.sh │ ├── train.py │ └── tree.py ├── LAL-Parser/ │ ├── EVALB/ │ │ ├── COLLINS.prm │ │ ├── LICENSE │ │ ├── Makefile │ │ ├── README │ │ ├── bug/ │ │ │ ├── bug.gld │ │ │ ├── bug.rsl-new │ │ │ ├── bug.rsl-old │ │ │ └── bug.tst │ │ ├── evalb │ │ ├── evalb.c │ │ ├── evalb.dSYM/ │ │ │ └── Contents/ │ │ │ ├── Info.plist │ │ │ └── Resources/ │ │ │ └── DWARF/ │ │ │ └── evalb │ │ ├── new.prm │ │ ├── nk.prm │ │ ├── sample/ │ │ │ ├── sample.gld │ │ │ ├── sample.prm │ │ │ ├── sample.rsl │ │ │ └── sample.tst │ │ └── tgrep_proc.prl │ ├── best_model/ │ │ └── readme.md │ └── src_joint/ │ ├── KM_parser.py │ ├── absa_parser.py │ ├── const_decoder.pyx │ ├── dep_eval.py │ ├── dep_reader.py │ ├── evaluate.py │ ├── hpsg_decoder.pyx │ ├── main.py │ ├── makehp.py │ ├── pretrained_bert/ │ │ ├── __init__.py │ │ ├── __main__.py │ │ ├── convert_tf_checkpoint_to_pytorch.py │ │ ├── file_utils.py │ │ ├── modeling.py │ │ ├── optimization.py │ │ └── tokenization.py │ ├── pytorch_transformers/ │ │ ├── __init__.py │ │ ├── __main__.py │ │ ├── convert_gpt2_checkpoint_to_pytorch.py │ │ ├── convert_openai_checkpoint_to_pytorch.py │ │ ├── convert_pytorch_checkpoint_to_tf.py │ │ ├── convert_roberta_checkpoint_to_pytorch.py │ │ ├── convert_tf_checkpoint_to_pytorch.py │ │ ├── convert_transfo_xl_checkpoint_to_pytorch.py │ │ ├── convert_xlm_checkpoint_to_pytorch.py │ │ ├── convert_xlnet_checkpoint_to_pytorch.py │ │ ├── file_utils.py │ │ ├── modeling_auto.py │ │ ├── modeling_bert.py │ │ ├── modeling_gpt2.py │ │ ├── modeling_openai.py │ │ ├── modeling_roberta.py │ │ ├── modeling_transfo_xl.py │ │ ├── modeling_transfo_xl_utilities.py │ │ ├── modeling_utils.py │ │ ├── modeling_xlm.py │ │ ├── modeling_xlnet.py │ │ ├── optimization.py │ │ ├── tests/ │ │ │ ├── __init__.py │ │ │ ├── conftest.py │ │ │ ├── fixtures/ │ │ │ │ ├── input.txt │ │ │ │ ├── sample_text.txt │ │ │ │ └── test_sentencepiece.model │ │ │ ├── modeling_auto_test.py │ │ │ ├── modeling_bert_test.py │ │ │ ├── modeling_common_test.py │ │ │ ├── modeling_gpt2_test.py │ │ │ ├── modeling_openai_test.py │ │ │ ├── modeling_roberta_test.py │ │ │ ├── modeling_transfo_xl_test.py │ │ │ ├── modeling_xlm_test.py │ │ │ ├── modeling_xlnet_test.py │ │ │ ├── optimization_test.py │ │ │ ├── tokenization_auto_test.py │ │ │ ├── tokenization_bert_test.py │ │ │ ├── tokenization_gpt2_test.py │ │ │ ├── tokenization_openai_test.py │ │ │ ├── tokenization_roberta_test.py │ │ │ ├── tokenization_tests_commons.py │ │ │ ├── tokenization_transfo_xl_test.py │ │ │ ├── tokenization_utils_test.py │ │ │ ├── tokenization_xlm_test.py │ │ │ └── tokenization_xlnet_test.py │ │ ├── tokenization_auto.py │ │ ├── tokenization_bert.py │ │ ├── tokenization_gpt2.py │ │ ├── tokenization_openai.py │ │ ├── tokenization_roberta.py │ │ ├── tokenization_transfo_xl.py │ │ ├── tokenization_utils.py │ │ ├── tokenization_xlm.py │ │ └── tokenization_xlnet.py │ ├── test.py │ ├── transliterate.py │ ├── trees.py │ ├── utils.py │ ├── utils_io.py │ └── vocabulary.py ├── LICENSE ├── README.md └── requirements.txt
SYMBOL INDEX (1257 symbols across 80 files)
FILE: DualGCN/data_utils.py
function ParseData (line 19) | def ParseData(data_path):
function build_tokenizer (line 59) | def build_tokenizer(fnames, max_length, data_file):
class Vocab (line 70) | class Vocab(object):
method __init__ (line 72) | def __init__(self, vocab_list, add_pad, add_unk):
method word_to_id (line 92) | def word_to_id(self, word):
method id_to_word (line 97) | def id_to_word(self, id_):
method has_word (line 102) | def has_word(self, word):
method __len__ (line 105) | def __len__(self):
method load_vocab (line 109) | def load_vocab(vocab_path: str):
method save_vocab (line 113) | def save_vocab(self, vocab_path):
class Tokenizer (line 118) | class Tokenizer(object):
method __init__ (line 120) | def __init__(self, vocab, max_length, lower, pos_char_to_int, pos_int_...
method from_files (line 129) | def from_files(cls, fnames, max_length, parse, lower=True):
method pad_sequence (line 141) | def pad_sequence(sequence, pad_id, maxlen, dtype='int64', padding='pos...
method text_to_sequence (line 154) | def text_to_sequence(self, text, reverse=False, padding='post', trunca...
method split_text (line 167) | def split_text(text):
class SentenceDataset (line 173) | class SentenceDataset(Dataset):
method __init__ (line 175) | def __init__(self, fname, tokenizer, opt, vocab_help):
method __getitem__ (line 230) | def __getitem__(self, index):
method __len__ (line 233) | def __len__(self):
function _load_wordvec (line 236) | def _load_wordvec(data_path, embed_dim, vocab=None):
function build_embedding_matrix (line 262) | def build_embedding_matrix(vocab, embed_dim, data_file):
function softmax (line 279) | def softmax(x):
class Tokenizer4BertGCN (line 297) | class Tokenizer4BertGCN:
method __init__ (line 298) | def __init__(self, max_seq_len, pretrained_bert_name):
method tokenize (line 303) | def tokenize(self, s):
method convert_tokens_to_ids (line 305) | def convert_tokens_to_ids(self, tokens):
class ABSAGCNData (line 309) | class ABSAGCNData(Dataset):
method __init__ (line 310) | def __init__(self, fname, tokenizer, opt):
method __len__ (line 405) | def __len__(self):
method __getitem__ (line 408) | def __getitem__(self, idx):
FILE: DualGCN/layers.py
class DynamicLSTM (line 6) | class DynamicLSTM(nn.Module):
method __init__ (line 10) | def __init__(self, input_size, hidden_size, num_layers=1, bias=True, b...
method forward (line 33) | def forward(self, x, x_len):
class SqueezeEmbedding (line 64) | class SqueezeEmbedding(nn.Module):
method __init__ (line 68) | def __init__(self, batch_first=True):
method forward (line 72) | def forward(self, x, x_len):
class SoftAttention (line 91) | class SoftAttention(nn.Module):
method __init__ (line 95) | def __init__(self, hidden_dim, embed_dim):
method forward (line 105) | def forward(self, h, aspect):
class Attention (line 116) | class Attention(nn.Module):
method __init__ (line 117) | def __init__(self, embed_dim, hidden_dim=None, out_dim=None, n_head=1,...
method reset_parameters (line 147) | def reset_parameters(self):
method forward (line 152) | def forward(self, k, q):
FILE: DualGCN/models/atae_lstm.py
class ATAE_LSTM (line 13) | class ATAE_LSTM(nn.Module):
method __init__ (line 15) | def __init__(self, embedding_matrix, opt):
method forward (line 23) | def forward(self, inputs):
FILE: DualGCN/models/dualgcn.py
class DualGCNClassifier (line 17) | class DualGCNClassifier(nn.Module):
method __init__ (line 18) | def __init__(self, embedding_matrix, opt):
method forward (line 25) | def forward(self, inputs):
class GCNAbsaModel (line 56) | class GCNAbsaModel(nn.Module):
method __init__ (line 57) | def __init__(self, embedding_matrix, opt):
method forward (line 68) | def forward(self, inputs):
class GCN (line 92) | class GCN(nn.Module):
method __init__ (line 93) | def __init__(self, opt, embeddings, mem_dim, num_layers):
method encode_with_rnn (line 131) | def encode_with_rnn(self, rnn_inputs, seq_lens, batch_size):
method forward (line 138) | def forward(self, adj, inputs):
function rnn_zero_state (line 204) | def rnn_zero_state(batch_size, hidden_dim, num_layers, bidirectional=True):
function attention (line 211) | def attention(query, key, mask=None, dropout=None):
function clones (line 224) | def clones(module, N):
class MultiHeadAttention (line 228) | class MultiHeadAttention(nn.Module):
method __init__ (line 230) | def __init__(self, h, d_model, dropout=0.1):
method forward (line 238) | def forward(self, query, key, mask=None):
FILE: DualGCN/models/dualgcn_bert.py
class LayerNorm (line 15) | class LayerNorm(nn.Module):
method __init__ (line 18) | def __init__(self, features, eps=1e-6):
method forward (line 24) | def forward(self, x):
class DualGCNBertClassifier (line 30) | class DualGCNBertClassifier(nn.Module):
method __init__ (line 31) | def __init__(self, bert, opt):
method forward (line 37) | def forward(self, inputs):
class GCNAbsaModel (line 68) | class GCNAbsaModel(nn.Module):
method __init__ (line 69) | def __init__(self, bert, opt):
method forward (line 74) | def forward(self, inputs):
class GCNBert (line 86) | class GCNBert(nn.Module):
method __init__ (line 87) | def __init__(self, bert, opt, num_layers):
method forward (line 115) | def forward(self, adj, inputs):
function attention (line 171) | def attention(query, key, mask=None, dropout=None):
function clones (line 184) | def clones(module, N):
class MultiHeadAttention (line 188) | class MultiHeadAttention(nn.Module):
method __init__ (line 190) | def __init__(self, h, d_model, dropout=0.1):
method forward (line 198) | def forward(self, query, key, mask=None):
FILE: DualGCN/models/ian.py
class IAN (line 13) | class IAN(nn.Module):
method __init__ (line 14) | def __init__(self, embedding_matrix, opt):
method forward (line 24) | def forward(self, inputs):
FILE: DualGCN/models/semgcn.py
class SemGCNClassifier (line 16) | class SemGCNClassifier(nn.Module):
method __init__ (line 17) | def __init__(self, embedding_matrix, opt):
method forward (line 24) | def forward(self, inputs):
class GCNAbsaModel (line 30) | class GCNAbsaModel(nn.Module):
method __init__ (line 31) | def __init__(self, embedding_matrix, opt):
method forward (line 43) | def forward(self, inputs):
class GCN (line 57) | class GCN(nn.Module):
method __init__ (line 58) | def __init__(self, opt, embeddings, mem_dim, num_layers):
method encode_with_rnn (line 94) | def encode_with_rnn(self, rnn_inputs, seq_lens, batch_size):
method forward (line 101) | def forward(self, inputs):
function rnn_zero_state (line 149) | def rnn_zero_state(batch_size, hidden_dim, num_layers, bidirectional=True):
function attention (line 156) | def attention(query, key, mask=None, dropout=None):
function clones (line 169) | def clones(module, N):
class MultiHeadAttention (line 173) | class MultiHeadAttention(nn.Module):
method __init__ (line 175) | def __init__(self, h, d_model, dropout=0.1):
method forward (line 184) | def forward(self, query, key, mask=None):
FILE: DualGCN/models/syngcn.py
class SynGCNClassifier (line 15) | class SynGCNClassifier(nn.Module):
method __init__ (line 16) | def __init__(self, embedding_matrix, opt):
method forward (line 23) | def forward(self, inputs):
class GCNAbsaModel (line 28) | class GCNAbsaModel(nn.Module):
method __init__ (line 29) | def __init__(self, embedding_matrix, opt):
method forward (line 42) | def forward(self, inputs):
class GCN (line 66) | class GCN(nn.Module):
method __init__ (line 67) | def __init__(self, opt, embeddings, mem_dim, num_layers):
method encode_with_rnn (line 95) | def encode_with_rnn(self, rnn_inputs, seq_lens, batch_size):
method forward (line 102) | def forward(self, adj, inputs):
function rnn_zero_state (line 131) | def rnn_zero_state(batch_size, hidden_dim, num_layers, bidirectional=True):
FILE: DualGCN/prepare_vocab.py
class VocabHelp (line 11) | class VocabHelp(object):
method __init__ (line 12) | def __init__(self, counter, specials=['<pad>', '<unk>']):
method __eq__ (line 31) | def __eq__(self, other):
method __len__ (line 38) | def __len__(self):
method extend (line 41) | def extend(self, v):
method load_vocab (line 50) | def load_vocab(vocab_path: str):
method save_vocab (line 54) | def save_vocab(self, vocab_path):
function parse_args (line 58) | def parse_args():
function main (line 66) | def main():
function load_tokens (line 120) | def load_tokens(filename):
FILE: DualGCN/train.py
function setup_seed (line 34) | def setup_seed(seed):
class Instructor (line 42) | class Instructor:
method __init__ (line 44) | def __init__(self, opt):
method _print_args (line 86) | def _print_args(self):
method _reset_params (line 101) | def _reset_params(self):
method get_bert_optimizer (line 110) | def get_bert_optimizer(self, model):
method _train (line 158) | def _train(self, criterion, optimizer, max_test_acc_overall=0):
method _evaluate (line 201) | def _evaluate(self, show_results=False):
method _test (line 227) | def _test(self):
method run (line 237) | def run(self):
function main (line 260) | def main():
FILE: DualGCN/tree.py
class Tree (line 10) | class Tree(object):
method __init__ (line 14) | def __init__(self):
method add_child (line 19) | def add_child(self, child):
method size (line 24) | def size(self):
method depth (line 33) | def depth(self):
method __iter__ (line 46) | def __iter__(self):
function head_to_tree (line 52) | def head_to_tree(head, tokens, len_):
function tree_to_adj (line 79) | def tree_to_adj(sent_len, tree, directed=False, self_loop=True):
FILE: LAL-Parser/EVALB/evalb.c
type s_terminal (line 80) | typedef struct ss_terminal {
type s_term_ind (line 86) | typedef struct ss_term_ind {
type s_bracket (line 94) | typedef struct ss_bracket {
type s_equiv (line 104) | typedef struct ss_equiv {
function main (line 306) | int
function init_global (line 416) | void
function print_head (line 437) | void
function init (line 449) | void
function read_parameter_file (line 500) | void
function set_param (line 546) | void
function narg (line 620) | int
function read_line (line 646) | int
function pushb (line 767) | void
function popb (line 774) | int
function stackempty (line 787) | int
function calc_result (line 801) | void
function fix_quote (line 977) | void
function reinsert_term (line 1017) | void
function massage_data (line 1071) | void
function modify_label (line 1144) | void
function individual_result (line 1162) | void
function print_total (line 1236) | void
function dsp_info (line 1325) | void
function is_terminator (line 1381) | int
function is_deletelabel (line 1392) | int
function is_deletelabel_for_length (line 1407) | int
function is_quote_term (line 1422) | int
function word_comp (line 1443) | int
function label_comp (line 1468) | int
function Error (line 1494) | void
function Fatal (line 1510) | void
function Usage (line 1522) | void
FILE: LAL-Parser/src_joint/KM_parser.py
function from_numpy (line 15) | def from_numpy(ndarray):
class BatchIndices (line 80) | class BatchIndices:
method __init__ (line 84) | def __init__(self, batch_idxs_np):
class FeatureDropoutFunction (line 97) | class FeatureDropoutFunction(torch.autograd.function.InplaceFunction):
method forward (line 99) | def forward(cls, ctx, input, batch_idxs, p=0.5, train=False, inplace=F...
method backward (line 126) | def backward(ctx, grad_output):
class FeatureDropout (line 133) | class FeatureDropout(nn.Module):
method __init__ (line 139) | def __init__(self, p=0.5, inplace=False):
method forward (line 147) | def forward(self, input, batch_idxs):
class LayerNormalization (line 151) | class LayerNormalization(nn.Module):
method __init__ (line 152) | def __init__(self, d_hid, eps=1e-3, affine=True):
method forward (line 161) | def forward(self, z):
class ScaledAttention (line 174) | class ScaledAttention(nn.Module):
method __init__ (line 175) | def __init__(self, hparams, attention_dropout=0.1):
method forward (line 182) | def forward(self, q, k, v, attn_mask=None):
class ScaledDotProductAttention (line 206) | class ScaledDotProductAttention(nn.Module):
method __init__ (line 207) | def __init__(self, d_model, attention_dropout=0.1):
method forward (line 213) | def forward(self, q, k, v, attn_mask=None):
class MultiHeadAttention (line 244) | class MultiHeadAttention(nn.Module):
method __init__ (line 249) | def __init__(self, hparams, n_head, d_model, d_k, d_v, residual_dropou...
method split_qkv_packed (line 303) | def split_qkv_packed(self, inp, qk_inp=None):
method pad_and_rearrange (line 329) | def pad_and_rearrange(self, q_s, k_s, v_s, batch_idxs):
method combine_v (line 357) | def combine_v(self, outputs):
method forward (line 381) | def forward(self, inp, batch_idxs, qk_inp=None):
class PositionwiseFeedForward (line 407) | class PositionwiseFeedForward(nn.Module):
method __init__ (line 415) | def __init__(self, d_hid, d_ff, relu_dropout=0.1, residual_dropout=0.1):
method forward (line 426) | def forward(self, x, batch_idxs):
class PartitionedPositionwiseFeedForward (line 437) | class PartitionedPositionwiseFeedForward(nn.Module):
method __init__ (line 438) | def __init__(self, d_hid, d_ff, d_positional, relu_dropout=0.1, residu...
method forward (line 450) | def forward(self, x, batch_idxs):
class MultiLevelEmbedding (line 469) | class MultiLevelEmbedding(nn.Module):
method __init__ (line 470) | def __init__(self,
method forward (line 548) | def forward(self, xs, pre_words_idxs, batch_idxs, extra_content_annota...
class CharacterLSTM (line 596) | class CharacterLSTM(nn.Module):
method __init__ (line 597) | def __init__(self, num_embeddings, d_embedding, d_out,
method forward (line 617) | def forward(self, chars_padded_np, word_lens_np, batch_idxs):
function get_elmo_class (line 642) | def get_elmo_class():
function get_xlnet (line 692) | def get_xlnet(xlnet_model, xlnet_do_lower_case):
function get_roberta (line 704) | def get_roberta(roberta_model, roberta_do_lower_case):
function get_bert (line 716) | def get_bert(bert_model, bert_do_lower_case):
class BiLinear (line 737) | class BiLinear(nn.Module):
method __init__ (line 741) | def __init__(self, left_features, right_features, out_features, bias=T...
method reset_parameters (line 767) | def reset_parameters(self):
method forward (line 773) | def forward(self, input_left, input_right):
class BiAAttention (line 795) | class BiAAttention(nn.Module):
method __init__ (line 800) | def __init__(self, hparams):
method forward (line 807) | def forward(self, input_d, input_e, input_s = None):
class Dep_score (line 818) | class Dep_score(nn.Module):
method __init__ (line 819) | def __init__(self, hparams, num_labels):
method forward (line 834) | def forward(self, outputs, outpute):
class LabelAttention (line 868) | class LabelAttention(nn.Module):
method __init__ (line 873) | def __init__(self, hparams, d_model, d_k, d_v, d_l, d_proj, use_resdro...
method split_qkv_packed (line 952) | def split_qkv_packed(self, inp, k_inp=None):
method pad_and_rearrange (line 988) | def pad_and_rearrange(self, q_s, k_s, v_s, batch_idxs):
method combine_v (line 1028) | def combine_v(self, outputs):
method forward (line 1058) | def forward(self, inp, batch_idxs, k_inp=None):
class Encoder (line 1105) | class Encoder(nn.Module):
method __init__ (line 1106) | def __init__(self, hparams, embedding,
method forward (line 1165) | def forward(self, xs, pre_words_idxs, batch_idxs, extra_content_annota...
class ChartParser (line 1176) | class ChartParser(nn.Module):
method __init__ (line 1177) | def __init__(
method model (line 1407) | def model(self):
method from_spec (line 1411) | def from_spec(cls, spec, model):
method split_batch (line 1465) | def split_batch(self, sentences, golds, subbatch_max_tokens=3000):
method parse_batch (line 1482) | def parse_batch(self, sentences, golds=None):
method label_scores_from_annotations (line 1869) | def label_scores_from_annotations(self, fencepost_annotations_start, f...
method parse_from_annotations (line 1896) | def parse_from_annotations(self, fencepost_annotations_start, fencepos...
method decode_from_chart_batch (line 1931) | def decode_from_chart_batch(self, sentences, charts_np, golds=None):
method decode_from_chart (line 1942) | def decode_from_chart(self, sentence, label_scores_chart_np, arc_dc_np...
FILE: LAL-Parser/src_joint/absa_parser.py
function torch_load (line 13) | def torch_load(load_path):
class Config (line 20) | class Config(object):
class ParseHead (line 25) | class ParseHead(object):
method __init__ (line 26) | def __init__(self, config) -> None:
method parse_heads (line 36) | def parse_heads(self, sentence):
FILE: LAL-Parser/src_joint/dep_eval.py
function is_uni_punctuation (line 6) | def is_uni_punctuation(word):
function is_punctuation (line 11) | def is_punctuation(word, pos, punct_set=None):
function eval (line 18) | def eval(batch_size, words, postags, heads_pred, types_pred, heads, type...
FILE: LAL-Parser/src_joint/dep_reader.py
class Sentence (line 6) | class Sentence(object):
method __init__ (line 7) | def __init__(self, words, postags):
method length (line 11) | def length(self):
class DependencyInstance (line 15) | class DependencyInstance(object):
method __init__ (line 16) | def __init__(self, sentence, postags, heads, types):
method length (line 22) | def length(self):
class CoNLLXReader (line 26) | class CoNLLXReader(object):
method __init__ (line 27) | def __init__(self, file_path, type_vocab = None):
method close (line 31) | def close(self):
method getNext (line 34) | def getNext(self):
FILE: LAL-Parser/src_joint/evaluate.py
class FScore (line 9) | class FScore(object):
method __init__ (line 10) | def __init__(self, recall, precision, fscore):
method __str__ (line 15) | def __str__(self):
function evalb (line 19) | def evalb(evalb_dir, gold_trees, predicted_trees, ref_gold_path=None):
FILE: LAL-Parser/src_joint/main.py
function torch_load (line 38) | def torch_load(load_path):
function format_elapsed (line 44) | def format_elapsed(start_time):
function make_hparams (line 54) | def make_hparams():
function count_wh (line 134) | def count_wh(str, data, heads, types):
function run_train (line 146) | def run_train(args, hparams):
function run_test (line 544) | def run_test(args):
function run_parse (line 650) | def run_parse(args):
function run_parse_head (line 736) | def run_parse_head(args):
function main (line 794) | def main():
FILE: LAL-Parser/src_joint/makehp.py
class HParams (line 1) | class HParams():
method __init__ (line 3) | def __init__(self, **kwargs):
method __getitem__ (line 7) | def __getitem__(self, item):
method __setitem__ (line 10) | def __setitem__(self, item, value):
method to_dict (line 15) | def to_dict(self):
method populate_arguments (line 23) | def populate_arguments(self, parser):
method set_from_args (line 37) | def set_from_args(self, args):
method print (line 46) | def print(self):
FILE: LAL-Parser/src_joint/pretrained_bert/__main__.py
function main (line 2) | def main():
FILE: LAL-Parser/src_joint/pretrained_bert/convert_tf_checkpoint_to_pytorch.py
function convert_tf_checkpoint_to_pytorch (line 30) | def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, bert_config_fil...
FILE: LAL-Parser/src_joint/pretrained_bert/file_utils.py
function url_to_filename (line 30) | def url_to_filename(url: str, etag: str = None) -> str:
function filename_to_url (line 48) | def filename_to_url(filename: str, cache_dir: Union[str, Path] = None) -...
function cached_path (line 74) | def cached_path(url_or_filename: Union[str, Path], cache_dir: Union[str,...
function split_s3_path (line 104) | def split_s3_path(url: str) -> Tuple[str, str]:
function s3_request (line 117) | def s3_request(func: Callable):
function s3_etag (line 137) | def s3_etag(url: str) -> Optional[str]:
function s3_get (line 146) | def s3_get(url: str, temp_file: IO) -> None:
function http_get (line 153) | def http_get(url: str, temp_file: IO) -> None:
function get_from_cache (line 165) | def get_from_cache(url: str, cache_dir: Union[str, Path] = None) -> str:
function read_set_from_file (line 224) | def read_set_from_file(filename: str) -> Set[str]:
function get_file_extension (line 236) | def get_file_extension(path: str, dot=True, lower: bool = True):
FILE: LAL-Parser/src_joint/pretrained_bert/modeling.py
function gelu (line 51) | def gelu(x):
function swish (line 59) | def swish(x):
class BertConfig (line 66) | class BertConfig(object):
method __init__ (line 69) | def __init__(self,
method from_dict (line 127) | def from_dict(cls, json_object):
method from_json_file (line 135) | def from_json_file(cls, json_file):
method __repr__ (line 141) | def __repr__(self):
method to_dict (line 144) | def to_dict(self):
method to_json_string (line 149) | def to_json_string(self):
class BertLayerNorm (line 157) | class BertLayerNorm(nn.Module):
method __init__ (line 158) | def __init__(self, hidden_size, eps=1e-12):
method forward (line 166) | def forward(self, x):
class BertEmbeddings (line 172) | class BertEmbeddings(nn.Module):
method __init__ (line 175) | def __init__(self, config):
method forward (line 186) | def forward(self, input_ids, token_type_ids=None):
class BertSelfAttention (line 203) | class BertSelfAttention(nn.Module):
method __init__ (line 204) | def __init__(self, config):
method transpose_for_scores (line 220) | def transpose_for_scores(self, x):
method forward (line 225) | def forward(self, hidden_states, attention_mask):
class BertSelfOutput (line 254) | class BertSelfOutput(nn.Module):
method __init__ (line 255) | def __init__(self, config):
method forward (line 261) | def forward(self, hidden_states, input_tensor):
class BertAttention (line 268) | class BertAttention(nn.Module):
method __init__ (line 269) | def __init__(self, config):
method forward (line 274) | def forward(self, input_tensor, attention_mask):
class BertIntermediate (line 280) | class BertIntermediate(nn.Module):
method __init__ (line 281) | def __init__(self, config):
method forward (line 287) | def forward(self, hidden_states):
class BertOutput (line 293) | class BertOutput(nn.Module):
method __init__ (line 294) | def __init__(self, config):
method forward (line 300) | def forward(self, hidden_states, input_tensor):
class BertLayer (line 307) | class BertLayer(nn.Module):
method __init__ (line 308) | def __init__(self, config):
method forward (line 314) | def forward(self, hidden_states, attention_mask):
class BertEncoder (line 321) | class BertEncoder(nn.Module):
method __init__ (line 322) | def __init__(self, config):
method forward (line 327) | def forward(self, hidden_states, attention_mask, output_all_encoded_la...
class BertPooler (line 338) | class BertPooler(nn.Module):
method __init__ (line 339) | def __init__(self, config):
method forward (line 344) | def forward(self, hidden_states):
class BertPredictionHeadTransform (line 353) | class BertPredictionHeadTransform(nn.Module):
method __init__ (line 354) | def __init__(self, config):
method forward (line 361) | def forward(self, hidden_states):
class BertLMPredictionHead (line 368) | class BertLMPredictionHead(nn.Module):
method __init__ (line 369) | def __init__(self, config, bert_model_embedding_weights):
method forward (line 381) | def forward(self, hidden_states):
class BertOnlyMLMHead (line 387) | class BertOnlyMLMHead(nn.Module):
method __init__ (line 388) | def __init__(self, config, bert_model_embedding_weights):
method forward (line 392) | def forward(self, sequence_output):
class BertOnlyNSPHead (line 397) | class BertOnlyNSPHead(nn.Module):
method __init__ (line 398) | def __init__(self, config):
method forward (line 402) | def forward(self, pooled_output):
class BertPreTrainingHeads (line 407) | class BertPreTrainingHeads(nn.Module):
method __init__ (line 408) | def __init__(self, config, bert_model_embedding_weights):
method forward (line 413) | def forward(self, sequence_output, pooled_output):
class PreTrainedBertModel (line 419) | class PreTrainedBertModel(nn.Module):
method __init__ (line 423) | def __init__(self, config, *inputs, **kwargs):
method init_bert_weights (line 434) | def init_bert_weights(self, module):
method from_pretrained (line 448) | def from_pretrained(cls, pretrained_model_name, state_dict=None, cache...
class BertModel (line 554) | class BertModel(PreTrainedBertModel):
method __init__ (line 598) | def __init__(self, config):
method forward (line 605) | def forward(self, input_ids, token_type_ids=None, attention_mask=None,...
class BertForPreTraining (line 637) | class BertForPreTraining(PreTrainedBertModel):
method __init__ (line 687) | def __init__(self, config):
method forward (line 693) | def forward(self, input_ids, token_type_ids=None, attention_mask=None,...
class BertForMaskedLM (line 708) | class BertForMaskedLM(PreTrainedBertModel):
method __init__ (line 750) | def __init__(self, config):
method forward (line 756) | def forward(self, input_ids, token_type_ids=None, attention_mask=None,...
class BertForNextSentencePrediction (line 769) | class BertForNextSentencePrediction(PreTrainedBertModel):
method __init__ (line 812) | def __init__(self, config):
method forward (line 818) | def forward(self, input_ids, token_type_ids=None, attention_mask=None,...
class BertForSequenceClassification (line 831) | class BertForSequenceClassification(PreTrainedBertModel):
method __init__ (line 876) | def __init__(self, config, num_labels=2):
method forward (line 884) | def forward(self, input_ids, token_type_ids=None, attention_mask=None,...
class BertForMultipleChoice (line 897) | class BertForMultipleChoice(PreTrainedBertModel):
method __init__ (line 941) | def __init__(self, config, num_choices=2):
method forward (line 949) | def forward(self, input_ids, token_type_ids=None, attention_mask=None,...
class BertForTokenClassification (line 966) | class BertForTokenClassification(PreTrainedBertModel):
method __init__ (line 1011) | def __init__(self, config, num_labels=2):
method forward (line 1019) | def forward(self, input_ids, token_type_ids=None, attention_mask=None,...
class BertForQuestionAnswering (line 1032) | class BertForQuestionAnswering(PreTrainedBertModel):
method __init__ (line 1087) | def __init__(self, config):
method forward (line 1095) | def forward(self, input_ids, token_type_ids=None, attention_mask=None,...
FILE: LAL-Parser/src_joint/pretrained_bert/optimization.py
function warmup_cosine (line 23) | def warmup_cosine(x, warmup=0.002):
function warmup_constant (line 28) | def warmup_constant(x, warmup=0.002):
function warmup_linear (line 33) | def warmup_linear(x, warmup=0.002):
class BertAdam (line 45) | class BertAdam(Optimizer):
method __init__ (line 59) | def __init__(self, params, lr=required, warmup=-1, t_total=-1, schedul...
method get_lr (line 79) | def get_lr(self):
method step (line 94) | def step(self, closure=None):
FILE: LAL-Parser/src_joint/pretrained_bert/tokenization.py
function load_vocab (line 51) | def load_vocab(vocab_file):
function whitespace_tokenize (line 66) | def whitespace_tokenize(text):
class BertTokenizer (line 75) | class BertTokenizer(object):
method __init__ (line 78) | def __init__(self, vocab_file, do_lower_case=True, max_len=None):
method tokenize (line 90) | def tokenize(self, text):
method convert_tokens_to_ids (line 97) | def convert_tokens_to_ids(self, tokens):
method convert_ids_to_tokens (line 110) | def convert_ids_to_tokens(self, ids):
method from_pretrained (line 118) | def from_pretrained(cls, pretrained_model_name, cache_dir=None, *input...
class BasicTokenizer (line 156) | class BasicTokenizer(object):
method __init__ (line 159) | def __init__(self, do_lower_case=True):
method tokenize (line 167) | def tokenize(self, text):
method _run_strip_accents (line 188) | def _run_strip_accents(self, text):
method _run_split_on_punc (line 199) | def _run_split_on_punc(self, text):
method _tokenize_chinese_chars (line 219) | def _tokenize_chinese_chars(self, text):
method _is_chinese_char (line 232) | def _is_chinese_char(self, cp):
method _clean_text (line 254) | def _clean_text(self, text):
class WordpieceTokenizer (line 268) | class WordpieceTokenizer(object):
method __init__ (line 271) | def __init__(self, vocab, unk_token="[UNK]", max_input_chars_per_word=...
method tokenize (line 276) | def tokenize(self, text):
function _is_whitespace (line 328) | def _is_whitespace(char):
function _is_control (line 340) | def _is_control(char):
function _is_punctuation (line 352) | def _is_punctuation(char):
FILE: LAL-Parser/src_joint/pytorch_transformers/__main__.py
function main (line 2) | def main():
FILE: LAL-Parser/src_joint/pytorch_transformers/convert_gpt2_checkpoint_to_pytorch.py
function convert_gpt2_checkpoint_to_pytorch (line 33) | def convert_gpt2_checkpoint_to_pytorch(gpt2_checkpoint_path, gpt2_config...
FILE: LAL-Parser/src_joint/pytorch_transformers/convert_openai_checkpoint_to_pytorch.py
function convert_openai_checkpoint_to_pytorch (line 33) | def convert_openai_checkpoint_to_pytorch(openai_checkpoint_folder_path, ...
FILE: LAL-Parser/src_joint/pytorch_transformers/convert_pytorch_checkpoint_to_tf.py
function convert_pytorch_checkpoint_to_tf (line 26) | def convert_pytorch_checkpoint_to_tf(model:BertModel, ckpt_dir:str, mode...
function main (line 95) | def main(raw_args=None):
FILE: LAL-Parser/src_joint/pytorch_transformers/convert_roberta_checkpoint_to_pytorch.py
function convert_roberta_checkpoint_to_pytorch (line 42) | def convert_roberta_checkpoint_to_pytorch(roberta_checkpoint_path, pytor...
FILE: LAL-Parser/src_joint/pytorch_transformers/convert_tf_checkpoint_to_pytorch.py
function convert_tf_checkpoint_to_pytorch (line 29) | def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, bert_config_fil...
FILE: LAL-Parser/src_joint/pytorch_transformers/convert_transfo_xl_checkpoint_to_pytorch.py
function convert_transfo_xl_checkpoint_to_pytorch (line 48) | def convert_transfo_xl_checkpoint_to_pytorch(tf_checkpoint_path,
FILE: LAL-Parser/src_joint/pytorch_transformers/convert_xlm_checkpoint_to_pytorch.py
function convert_xlm_checkpoint_to_pytorch (line 32) | def convert_xlm_checkpoint_to_pytorch(xlm_checkpoint_path, pytorch_dump_...
FILE: LAL-Parser/src_joint/pytorch_transformers/convert_xlnet_checkpoint_to_pytorch.py
function convert_xlnet_checkpoint_to_pytorch (line 46) | def convert_xlnet_checkpoint_to_pytorch(tf_checkpoint_path, bert_config_...
FILE: LAL-Parser/src_joint/pytorch_transformers/file_utils.py
function url_to_filename (line 52) | def url_to_filename(url, etag=None):
function filename_to_url (line 70) | def filename_to_url(filename, cache_dir=None):
function cached_path (line 96) | def cached_path(url_or_filename, cache_dir=None):
function split_s3_path (line 126) | def split_s3_path(url):
function s3_request (line 139) | def s3_request(func):
function s3_etag (line 159) | def s3_etag(url):
function s3_get (line 168) | def s3_get(url, temp_file):
function http_get (line 175) | def http_get(url, temp_file):
function get_from_cache (line 187) | def get_from_cache(url, cache_dir=None):
FILE: LAL-Parser/src_joint/pytorch_transformers/modeling_auto.py
class AutoConfig (line 38) | class AutoConfig(object):
method __init__ (line 59) | def __init__(self):
method from_pretrained (line 64) | def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):
class AutoModel (line 133) | class AutoModel(object):
method __init__ (line 155) | def __init__(self):
method from_pretrained (line 160) | def from_pretrained(cls, pretrained_model_name_or_path, *model_args, *...
FILE: LAL-Parser/src_joint/pytorch_transformers/modeling_bert.py
function load_tf_weights_in_bert (line 69) | def load_tf_weights_in_bert(model, config, tf_checkpoint_path):
function gelu (line 136) | def gelu(x):
function swish (line 145) | def swish(x):
class BertConfig (line 152) | class BertConfig(PretrainedConfig):
method __init__ (line 183) | def __init__(self,
class BertLayerNorm (line 227) | class BertLayerNorm(nn.Module):
method __init__ (line 228) | def __init__(self, hidden_size, eps=1e-12):
method forward (line 236) | def forward(self, x):
class BertEmbeddings (line 242) | class BertEmbeddings(nn.Module):
method __init__ (line 245) | def __init__(self, config):
method forward (line 256) | def forward(self, input_ids, token_type_ids=None, position_ids=None):
class BertSelfAttention (line 274) | class BertSelfAttention(nn.Module):
method __init__ (line 275) | def __init__(self, config):
method transpose_for_scores (line 293) | def transpose_for_scores(self, x):
method forward (line 298) | def forward(self, hidden_states, attention_mask, head_mask=None):
class BertSelfOutput (line 334) | class BertSelfOutput(nn.Module):
method __init__ (line 335) | def __init__(self, config):
method forward (line 341) | def forward(self, hidden_states, input_tensor):
class BertAttention (line 348) | class BertAttention(nn.Module):
method __init__ (line 349) | def __init__(self, config):
method prune_heads (line 354) | def prune_heads(self, heads):
method forward (line 371) | def forward(self, input_tensor, attention_mask, head_mask=None):
class BertIntermediate (line 378) | class BertIntermediate(nn.Module):
method __init__ (line 379) | def __init__(self, config):
method forward (line 387) | def forward(self, hidden_states):
class BertOutput (line 393) | class BertOutput(nn.Module):
method __init__ (line 394) | def __init__(self, config):
method forward (line 400) | def forward(self, hidden_states, input_tensor):
class BertLayer (line 407) | class BertLayer(nn.Module):
method __init__ (line 408) | def __init__(self, config):
method forward (line 414) | def forward(self, hidden_states, attention_mask, head_mask=None):
class BertEncoder (line 423) | class BertEncoder(nn.Module):
method __init__ (line 424) | def __init__(self, config):
method forward (line 430) | def forward(self, hidden_states, attention_mask, head_mask=None):
class BertPooler (line 455) | class BertPooler(nn.Module):
method __init__ (line 456) | def __init__(self, config):
method forward (line 461) | def forward(self, hidden_states):
class BertPredictionHeadTransform (line 470) | class BertPredictionHeadTransform(nn.Module):
method __init__ (line 471) | def __init__(self, config):
method forward (line 480) | def forward(self, hidden_states):
class BertLMPredictionHead (line 487) | class BertLMPredictionHead(nn.Module):
method __init__ (line 488) | def __init__(self, config):
method forward (line 500) | def forward(self, hidden_states):
class BertOnlyMLMHead (line 506) | class BertOnlyMLMHead(nn.Module):
method __init__ (line 507) | def __init__(self, config):
method forward (line 511) | def forward(self, sequence_output):
class BertOnlyNSPHead (line 516) | class BertOnlyNSPHead(nn.Module):
method __init__ (line 517) | def __init__(self, config):
method forward (line 521) | def forward(self, pooled_output):
class BertPreTrainingHeads (line 526) | class BertPreTrainingHeads(nn.Module):
method __init__ (line 527) | def __init__(self, config):
method forward (line 532) | def forward(self, sequence_output, pooled_output):
class BertPreTrainedModel (line 538) | class BertPreTrainedModel(PreTrainedModel):
method __init__ (line 547) | def __init__(self, *inputs, **kwargs):
method init_weights (line 550) | def init_weights(self, module):
class BertModel (line 624) | class BertModel(BertPreTrainedModel):
method __init__ (line 653) | def __init__(self, config):
method _resize_token_embeddings (line 662) | def _resize_token_embeddings(self, new_num_tokens):
method _prune_heads (line 668) | def _prune_heads(self, heads_to_prune):
method forward (line 676) | def forward(self, input_ids, token_type_ids=None, attention_mask=None,...
class BertForPreTraining (line 726) | class BertForPreTraining(BertPreTrainedModel):
method __init__ (line 763) | def __init__(self, config):
method tie_weights (line 772) | def tie_weights(self):
method forward (line 779) | def forward(self, input_ids, token_type_ids=None, attention_mask=None,...
class BertForMaskedLM (line 801) | class BertForMaskedLM(BertPreTrainedModel):
method __init__ (line 831) | def __init__(self, config):
method tie_weights (line 840) | def tie_weights(self):
method forward (line 847) | def forward(self, input_ids, token_type_ids=None, attention_mask=None,...
class BertForNextSentencePrediction (line 866) | class BertForNextSentencePrediction(BertPreTrainedModel):
method __init__ (line 896) | def __init__(self, config):
method forward (line 904) | def forward(self, input_ids, token_type_ids=None, attention_mask=None,...
class BertForSequenceClassification (line 924) | class BertForSequenceClassification(BertPreTrainedModel):
method __init__ (line 955) | def __init__(self, config):
method forward (line 965) | def forward(self, input_ids, token_type_ids=None, attention_mask=None,...
class BertForMultipleChoice (line 992) | class BertForMultipleChoice(BertPreTrainedModel):
method __init__ (line 1060) | def __init__(self, config):
method forward (line 1069) | def forward(self, input_ids, token_type_ids=None, attention_mask=None,...
class BertForTokenClassification (line 1098) | class BertForTokenClassification(BertPreTrainedModel):
method __init__ (line 1127) | def __init__(self, config):
method forward (line 1137) | def forward(self, input_ids, token_type_ids=None, attention_mask=None,...
class BertForQuestionAnswering (line 1165) | class BertForQuestionAnswering(BertPreTrainedModel):
method __init__ (line 1202) | def __init__(self, config):
method forward (line 1211) | def forward(self, input_ids, token_type_ids=None, attention_mask=None,...
FILE: LAL-Parser/src_joint/pytorch_transformers/modeling_gpt2.py
function load_tf_weights_in_gpt2 (line 45) | def load_tf_weights_in_gpt2(model, config, gpt2_checkpoint_path):
function gelu (line 99) | def gelu(x):
class GPT2Config (line 103) | class GPT2Config(PretrainedConfig):
method __init__ (line 125) | def __init__(
method max_position_embeddings (line 200) | def max_position_embeddings(self):
method hidden_size (line 204) | def hidden_size(self):
method num_attention_heads (line 208) | def num_attention_heads(self):
method num_hidden_layers (line 212) | def num_hidden_layers(self):
class Attention (line 217) | class Attention(nn.Module):
method __init__ (line 218) | def __init__(self, nx, n_ctx, config, scale=False):
method prune_heads (line 235) | def prune_heads(self, heads):
method _attn (line 251) | def _attn(self, q, k, v, head_mask=None):
method merge_heads (line 271) | def merge_heads(self, x):
method split_heads (line 276) | def split_heads(self, x, k=False):
method forward (line 284) | def forward(self, x, layer_past=None, head_mask=None):
class MLP (line 307) | class MLP(nn.Module):
method __init__ (line 308) | def __init__(self, n_state, config): # in MLP: n_state=3072 (4 * n_embd)
method forward (line 316) | def forward(self, x):
class Block (line 322) | class Block(nn.Module):
method __init__ (line 323) | def __init__(self, n_ctx, config, scale=False):
method forward (line 331) | def forward(self, x, layer_past=None, head_mask=None):
class GPT2PreTrainedModel (line 343) | class GPT2PreTrainedModel(PreTrainedModel):
method __init__ (line 352) | def __init__(self, *inputs, **kwargs):
method init_weights (line 355) | def init_weights(self, module):
class GPT2Model (line 417) | class GPT2Model(GPT2PreTrainedModel):
method __init__ (line 443) | def __init__(self, config):
method _resize_token_embeddings (line 456) | def _resize_token_embeddings(self, new_num_tokens):
method _prune_heads (line 460) | def _prune_heads(self, heads_to_prune):
method forward (line 467) | def forward(self, input_ids, position_ids=None, token_type_ids=None, p...
class GPT2LMHeadModel (line 541) | class GPT2LMHeadModel(GPT2PreTrainedModel):
method __init__ (line 576) | def __init__(self, config):
method tie_weights (line 584) | def tie_weights(self):
method forward (line 591) | def forward(self, input_ids, position_ids=None, token_type_ids=None, l...
class GPT2DoubleHeadsModel (line 617) | class GPT2DoubleHeadsModel(GPT2PreTrainedModel):
method __init__ (line 694) | def __init__(self, config):
method tie_weights (line 702) | def tie_weights(self):
method forward (line 709) | def forward(self, input_ids, mc_token_ids=None, lm_labels=None, mc_lab...
FILE: LAL-Parser/src_joint/pytorch_transformers/modeling_openai.py
function load_tf_weights_in_openai_gpt (line 44) | def load_tf_weights_in_openai_gpt(model, config, openai_checkpoint_folde...
function gelu (line 119) | def gelu(x):
function swish (line 123) | def swish(x):
class OpenAIGPTConfig (line 130) | class OpenAIGPTConfig(PretrainedConfig):
method __init__ (line 157) | def __init__(
method max_position_embeddings (line 219) | def max_position_embeddings(self):
method hidden_size (line 223) | def hidden_size(self):
method num_attention_heads (line 227) | def num_attention_heads(self):
method num_hidden_layers (line 231) | def num_hidden_layers(self):
class Attention (line 235) | class Attention(nn.Module):
method __init__ (line 236) | def __init__(self, nx, n_ctx, config, scale=False):
method prune_heads (line 253) | def prune_heads(self, heads):
method _attn (line 269) | def _attn(self, q, k, v, head_mask=None):
method merge_heads (line 290) | def merge_heads(self, x):
method split_heads (line 295) | def split_heads(self, x, k=False):
method forward (line 303) | def forward(self, x, head_mask=None):
class MLP (line 321) | class MLP(nn.Module):
method __init__ (line 322) | def __init__(self, n_state, config): # in MLP: n_state=3072 (4 * n_embd)
method forward (line 330) | def forward(self, x):
class Block (line 336) | class Block(nn.Module):
method __init__ (line 337) | def __init__(self, n_ctx, config, scale=False):
method forward (line 345) | def forward(self, x, head_mask=None):
class OpenAIGPTPreTrainedModel (line 357) | class OpenAIGPTPreTrainedModel(PreTrainedModel):
method __init__ (line 366) | def __init__(self, *inputs, **kwargs):
method init_weights (line 369) | def init_weights(self, module):
class OpenAIGPTModel (line 427) | class OpenAIGPTModel(OpenAIGPTPreTrainedModel):
method __init__ (line 449) | def __init__(self, config):
method _resize_token_embeddings (line 461) | def _resize_token_embeddings(self, new_num_tokens):
method _prune_heads (line 465) | def _prune_heads(self, heads_to_prune):
method forward (line 472) | def forward(self, input_ids, position_ids=None, token_type_ids=None, h...
class OpenAIGPTLMHeadModel (line 536) | class OpenAIGPTLMHeadModel(OpenAIGPTPreTrainedModel):
method __init__ (line 567) | def __init__(self, config):
method tie_weights (line 575) | def tie_weights(self):
method forward (line 582) | def forward(self, input_ids, position_ids=None, token_type_ids=None, l...
class OpenAIGPTDoubleHeadsModel (line 607) | class OpenAIGPTDoubleHeadsModel(OpenAIGPTPreTrainedModel):
method __init__ (line 676) | def __init__(self, config):
method tie_weights (line 686) | def tie_weights(self):
method forward (line 693) | def forward(self, input_ids, mc_token_ids=None, lm_labels=None, mc_lab...
FILE: LAL-Parser/src_joint/pytorch_transformers/modeling_roberta.py
class RobertaEmbeddings (line 49) | class RobertaEmbeddings(BertEmbeddings):
method __init__ (line 53) | def __init__(self, config):
method forward (line 57) | def forward(self, input_ids, token_type_ids=None, position_ids=None):
class RobertaConfig (line 67) | class RobertaConfig(BertConfig):
class RobertaModel (line 129) | class RobertaModel(BertModel):
method __init__ (line 162) | def __init__(self, config):
method forward (line 168) | def forward(self, input_ids, token_type_ids=None, attention_mask=None,...
class RobertaForMaskedLM (line 178) | class RobertaForMaskedLM(BertPreTrainedModel):
method __init__ (line 212) | def __init__(self, config):
method tie_weights (line 221) | def tie_weights(self):
method forward (line 227) | def forward(self, input_ids, token_type_ids=None, attention_mask=None,...
class RobertaLMHead (line 244) | class RobertaLMHead(nn.Module):
method __init__ (line 247) | def __init__(self, config):
method forward (line 255) | def forward(self, features, **kwargs):
class RobertaForSequenceClassification (line 269) | class RobertaForSequenceClassification(BertPreTrainedModel):
method __init__ (line 304) | def __init__(self, config):
method forward (line 311) | def forward(self, input_ids, token_type_ids=None, attention_mask=None,...
class RobertaClassificationHead (line 333) | class RobertaClassificationHead(nn.Module):
method __init__ (line 336) | def __init__(self, config):
method forward (line 342) | def forward(self, features, **kwargs):
FILE: LAL-Parser/src_joint/pytorch_transformers/modeling_transfo_xl.py
function build_tf_to_pytorch_map (line 50) | def build_tf_to_pytorch_map(model, config):
function load_tf_weights_in_transfo_xl (line 122) | def load_tf_weights_in_transfo_xl(model, config, tf_path):
class TransfoXLConfig (line 178) | class TransfoXLConfig(PretrainedConfig):
method __init__ (line 215) | def __init__(self,
method max_position_embeddings (line 291) | def max_position_embeddings(self):
method vocab_size (line 295) | def vocab_size(self):
method vocab_size (line 299) | def vocab_size(self, value):
method hidden_size (line 303) | def hidden_size(self):
method num_attention_heads (line 307) | def num_attention_heads(self):
method num_hidden_layers (line 311) | def num_hidden_layers(self):
class PositionalEmbedding (line 315) | class PositionalEmbedding(nn.Module):
method __init__ (line 316) | def __init__(self, demb):
method forward (line 324) | def forward(self, pos_seq, bsz=None):
class PositionwiseFF (line 335) | class PositionwiseFF(nn.Module):
method __init__ (line 336) | def __init__(self, d_model, d_inner, dropout, pre_lnorm=False):
method forward (line 354) | def forward(self, inp):
class MultiHeadAttn (line 372) | class MultiHeadAttn(nn.Module):
method __init__ (line 373) | def __init__(self, n_head, d_model, d_head, dropout, dropatt=0,
method forward (line 403) | def forward(self, h, attn_mask=None, mems=None, head_mask=None):
class RelMultiHeadAttn (line 461) | class RelMultiHeadAttn(nn.Module):
method __init__ (line 462) | def __init__(self, n_head, d_model, d_head, dropout, dropatt=0,
method _parallelogram_mask (line 492) | def _parallelogram_mask(self, h, w, left=False):
method _shift (line 503) | def _shift(self, x, qlen, klen, mask, left=False):
method _rel_shift (line 521) | def _rel_shift(self, x, zero_triu=False):
method forward (line 537) | def forward(self, w, r, attn_mask=None, mems=None):
class RelPartialLearnableMultiHeadAttn (line 540) | class RelPartialLearnableMultiHeadAttn(RelMultiHeadAttn):
method __init__ (line 541) | def __init__(self, *args, **kwargs):
method forward (line 546) | def forward(self, w, r, attn_mask=None, mems=None, head_mask=None):
class RelLearnableMultiHeadAttn (line 628) | class RelLearnableMultiHeadAttn(RelMultiHeadAttn):
method __init__ (line 629) | def __init__(self, *args, **kwargs):
method forward (line 632) | def forward(self, w, r_emb, r_w_bias, r_bias, attn_mask=None, mems=Non...
class DecoderLayer (line 721) | class DecoderLayer(nn.Module):
method __init__ (line 722) | def __init__(self, n_head, d_model, d_head, d_inner, dropout, **kwargs):
method forward (line 729) | def forward(self, dec_inp, dec_attn_mask=None, mems=None, head_mask=No...
class RelLearnableDecoderLayer (line 739) | class RelLearnableDecoderLayer(nn.Module):
method __init__ (line 740) | def __init__(self, n_head, d_model, d_head, d_inner, dropout,
method forward (line 749) | def forward(self, dec_inp, r_emb, r_w_bias, r_bias, dec_attn_mask=None...
class RelPartialLearnableDecoderLayer (line 760) | class RelPartialLearnableDecoderLayer(nn.Module):
method __init__ (line 761) | def __init__(self, n_head, d_model, d_head, d_inner, dropout,
method forward (line 770) | def forward(self, dec_inp, r, dec_attn_mask=None, mems=None, head_mask...
class AdaptiveEmbedding (line 783) | class AdaptiveEmbedding(nn.Module):
method __init__ (line 784) | def __init__(self, n_token, d_embed, d_proj, cutoffs, div_val=1,
method forward (line 814) | def forward(self, inp):
class TransfoXLPreTrainedModel (line 847) | class TransfoXLPreTrainedModel(PreTrainedModel):
method __init__ (line 856) | def __init__(self, *inputs, **kwargs):
method _init_weight (line 859) | def _init_weight(self, weight):
method _init_bias (line 865) | def _init_bias(self, bias):
method init_weights (line 868) | def init_weights(self, m):
method set_num_special_tokens (line 909) | def set_num_special_tokens(self, num_special_tokens):
class TransfoXLModel (line 952) | class TransfoXLModel(TransfoXLPreTrainedModel):
method __init__ (line 978) | def __init__(self, config):
method _resize_token_embeddings (line 1060) | def _resize_token_embeddings(self, new_num_tokens):
method backward_compatible (line 1063) | def backward_compatible(self):
method reset_length (line 1066) | def reset_length(self, tgt_len, ext_len, mem_len):
method _prune_heads (line 1071) | def _prune_heads(self, heads):
method init_mems (line 1075) | def init_mems(self, data):
method _update_mems (line 1088) | def _update_mems(self, hids, mems, qlen, mlen):
method _forward (line 1111) | def _forward(self, dec_inp, mems=None, head_mask=None):
method forward (line 1242) | def forward(self, input_ids, mems=None, head_mask=None):
class TransfoXLLMHeadModel (line 1257) | class TransfoXLLMHeadModel(TransfoXLPreTrainedModel):
method __init__ (line 1293) | def __init__(self, config):
method tie_weights (line 1308) | def tie_weights(self):
method reset_length (line 1335) | def reset_length(self, tgt_len, ext_len, mem_len):
method init_mems (line 1338) | def init_mems(self, data):
method forward (line 1341) | def forward(self, input_ids, labels=None, mems=None, head_mask=None):
FILE: LAL-Parser/src_joint/pytorch_transformers/modeling_transfo_xl_utilities.py
class ProjectedAdaptiveLogSoftmax (line 31) | class ProjectedAdaptiveLogSoftmax(nn.Module):
method __init__ (line 32) | def __init__(self, n_token, d_embed, d_proj, cutoffs, div_val=1,
method _compute_logit (line 78) | def _compute_logit(self, hidden, weight, bias, proj):
method forward (line 92) | def forward(self, hidden, labels=None, keep_order=False):
method log_prob (line 198) | def log_prob(self, hidden):
class LogUniformSampler (line 260) | class LogUniformSampler(object):
method __init__ (line 261) | def __init__(self, range_max, n_sample):
method sample (line 280) | def sample(self, labels):
function sample_logits (line 301) | def sample_logits(embedding, bias, labels, inputs, sampler):
FILE: LAL-Parser/src_joint/pytorch_transformers/modeling_utils.py
class Identity (line 46) | class Identity(nn.Module):
method __init__ (line 49) | def __init__(self, *args, **kwargs):
method forward (line 52) | def forward(self, input):
function add_start_docstrings (line 57) | def add_start_docstrings(*docstr):
function add_start_docstrings (line 64) | def add_start_docstrings(*docstr):
class PretrainedConfig (line 70) | class PretrainedConfig(object):
method __init__ (line 86) | def __init__(self, **kwargs):
method save_pretrained (line 93) | def save_pretrained(self, save_directory):
method from_pretrained (line 105) | def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):
method from_dict (line 195) | def from_dict(cls, json_object):
method from_json_file (line 203) | def from_json_file(cls, json_file):
method __eq__ (line 209) | def __eq__(self, other):
method __repr__ (line 212) | def __repr__(self):
method to_dict (line 215) | def to_dict(self):
method to_json_string (line 220) | def to_json_string(self):
method to_json_file (line 224) | def to_json_file(self, json_file_path):
class PreTrainedModel (line 230) | class PreTrainedModel(nn.Module):
method __init__ (line 252) | def __init__(self, config, *inputs, **kwargs):
method _get_resized_embeddings (line 264) | def _get_resized_embeddings(self, old_embeddings, new_num_tokens=None):
method _tie_or_clone_weights (line 298) | def _tie_or_clone_weights(self, first_module, second_module):
method resize_token_embeddings (line 306) | def resize_token_embeddings(self, new_num_tokens=None):
method prune_heads (line 334) | def prune_heads(self, heads_to_prune):
method save_pretrained (line 344) | def save_pretrained(self, save_directory):
method from_pretrained (line 362) | def from_pretrained(cls, pretrained_model_name_or_path, *model_args, *...
class Conv1D (line 547) | class Conv1D(nn.Module):
method __init__ (line 548) | def __init__(self, nf, nx):
method forward (line 559) | def forward(self, x):
class PoolerStartLogits (line 566) | class PoolerStartLogits(nn.Module):
method __init__ (line 568) | def __init__(self, config):
method forward (line 572) | def forward(self, hidden_states, p_mask=None):
class PoolerEndLogits (line 586) | class PoolerEndLogits(nn.Module):
method __init__ (line 589) | def __init__(self, config):
method forward (line 596) | def forward(self, hidden_states, start_states=None, start_positions=No...
class PoolerAnswerClass (line 627) | class PoolerAnswerClass(nn.Module):
method __init__ (line 629) | def __init__(self, config):
method forward (line 635) | def forward(self, hidden_states, start_states=None, start_positions=No...
class SQuADHead (line 671) | class SQuADHead(nn.Module):
method __init__ (line 711) | def __init__(self, config):
method forward (line 720) | def forward(self, hidden_states, start_positions=None, end_positions=N...
class SequenceSummary (line 780) | class SequenceSummary(nn.Module):
method __init__ (line 795) | def __init__(self, config):
method forward (line 825) | def forward(self, hidden_states, cls_index=None):
function prune_linear_layer (line 857) | def prune_linear_layer(layer, index, dim=0):
function prune_conv1d_layer (line 882) | def prune_conv1d_layer(layer, index, dim=1):
function prune_layer (line 906) | def prune_layer(layer, index, dim=None):
FILE: LAL-Parser/src_joint/pytorch_transformers/modeling_xlm.py
class XLMConfig (line 60) | class XLMConfig(PretrainedConfig):
method __init__ (line 105) | def __init__(self,
method vocab_size (line 184) | def vocab_size(self):
method vocab_size (line 188) | def vocab_size(self, value):
method hidden_size (line 192) | def hidden_size(self):
method num_attention_heads (line 196) | def num_attention_heads(self):
method num_hidden_layers (line 200) | def num_hidden_layers(self):
function create_sinusoidal_embeddings (line 204) | def create_sinusoidal_embeddings(n_pos, dim, out):
function gelu (line 215) | def gelu(x):
function get_masks (line 226) | def get_masks(slen, lengths, causal, padding_mask=None):
class MultiHeadAttention (line 251) | class MultiHeadAttention(nn.Module):
method __init__ (line 255) | def __init__(self, n_heads, dim, config):
method prune_heads (line 269) | def prune_heads(self, heads):
method forward (line 287) | def forward(self, input, mask, kv=None, cache=None, head_mask=None):
class TransformerFFN (line 351) | class TransformerFFN(nn.Module):
method __init__ (line 353) | def __init__(self, in_dim, dim_hidden, out_dim, config):
method forward (line 360) | def forward(self, input):
class XLMPreTrainedModel (line 368) | class XLMPreTrainedModel(PreTrainedModel):
method __init__ (line 377) | def __init__(self, *inputs, **kwargs):
method init_weights (line 380) | def init_weights(self, module):
class XLMModel (line 460) | class XLMModel(XLMPreTrainedModel):
method __init__ (line 487) | def __init__(self, config): #, dico, is_encoder, with_output):
method _resize_token_embeddings (line 549) | def _resize_token_embeddings(self, new_num_tokens):
method _prune_heads (line 553) | def _prune_heads(self, heads_to_prune):
method forward (line 561) | def forward(self, input_ids, lengths=None, position_ids=None, langs=None,
class XLMPredLayer (line 678) | class XLMPredLayer(nn.Module):
method __init__ (line 682) | def __init__(self, config):
method forward (line 700) | def forward(self, x, y=None):
class XLMWithLMHeadModel (line 723) | class XLMWithLMHeadModel(XLMPreTrainedModel):
method __init__ (line 754) | def __init__(self, config):
method tie_weights (line 762) | def tie_weights(self):
method forward (line 767) | def forward(self, input_ids, lengths=None, position_ids=None, langs=No...
class XLMForSequenceClassification (line 783) | class XLMForSequenceClassification(XLMPreTrainedModel):
method __init__ (line 814) | def __init__(self, config):
method forward (line 823) | def forward(self, input_ids, lengths=None, position_ids=None, langs=No...
class XLMForQuestionAnswering (line 850) | class XLMForQuestionAnswering(XLMPreTrainedModel):
method __init__ (line 893) | def __init__(self, config):
method forward (line 901) | def forward(self, input_ids, lengths=None, position_ids=None, langs=No...
FILE: LAL-Parser/src_joint/pytorch_transformers/modeling_xlnet.py
function build_tf_xlnet_to_pytorch_map (line 49) | def build_tf_xlnet_to_pytorch_map(model, config, tf_weights=None):
function load_tf_weights_in_xlnet (line 119) | def load_tf_weights_in_xlnet(model, config, tf_path):
function gelu (line 179) | def gelu(x):
function swish (line 188) | def swish(x):
class XLNetConfig (line 195) | class XLNetConfig(PretrainedConfig):
method __init__ (line 238) | def __init__(self,
method max_position_embeddings (line 312) | def max_position_embeddings(self):
method vocab_size (line 316) | def vocab_size(self):
method vocab_size (line 320) | def vocab_size(self, value):
method hidden_size (line 324) | def hidden_size(self):
method num_attention_heads (line 328) | def num_attention_heads(self):
method num_hidden_layers (line 332) | def num_hidden_layers(self):
class XLNetLayerNorm (line 340) | class XLNetLayerNorm(nn.Module):
method __init__ (line 341) | def __init__(self, d_model, eps=1e-12):
method forward (line 349) | def forward(self, x):
class XLNetRelativeAttention (line 355) | class XLNetRelativeAttention(nn.Module):
method __init__ (line 356) | def __init__(self, config):
method prune_heads (line 384) | def prune_heads(self, heads):
method rel_shift (line 388) | def rel_shift(x, klen=-1):
method rel_attn_core (line 400) | def rel_attn_core(self, q_head, k_head_h, v_head_h, k_head_r, seg_mat=...
method post_attention (line 439) | def post_attention(self, h, attn_vec, residual=True):
method forward (line 451) | def forward(self, h, g,
class XLNetFeedForward (line 544) | class XLNetFeedForward(nn.Module):
method __init__ (line 545) | def __init__(self, config):
method forward (line 557) | def forward(self, inp):
class XLNetLayer (line 567) | class XLNetLayer(nn.Module):
method __init__ (line 568) | def __init__(self, config):
method forward (line 574) | def forward(self, output_h, output_g,
class XLNetPreTrainedModel (line 590) | class XLNetPreTrainedModel(PreTrainedModel):
method __init__ (line 599) | def __init__(self, *inputs, **kwargs):
method init_weights (line 602) | def init_weights(self, module):
class XLNetModel (line 696) | class XLNetModel(XLNetPreTrainedModel):
method __init__ (line 722) | def __init__(self, config):
method _resize_token_embeddings (line 743) | def _resize_token_embeddings(self, new_num_tokens):
method _prune_heads (line 747) | def _prune_heads(self, heads_to_prune):
method create_mask (line 750) | def create_mask(self, qlen, mlen):
method cache_mem (line 780) | def cache_mem(self, curr_out, prev_mem):
method positional_embedding (line 796) | def positional_embedding(pos_seq, inv_freq, bsz=None):
method relative_positional_encoding (line 806) | def relative_positional_encoding(self, qlen, klen, bsz=None):
method forward (line 845) | def forward(self, input_ids, token_type_ids=None, input_mask=None, att...
class XLNetLMHeadModel (line 993) | class XLNetLMHeadModel(XLNetPreTrainedModel):
method __init__ (line 1033) | def __init__(self, config):
method tie_weights (line 1044) | def tie_weights(self):
method forward (line 1049) | def forward(self, input_ids, token_type_ids=None, input_mask=None, att...
class XLNetForSequenceClassification (line 1074) | class XLNetForSequenceClassification(XLNetPreTrainedModel):
method __init__ (line 1109) | def __init__(self, config):
method forward (line 1119) | def forward(self, input_ids, token_type_ids=None, input_mask=None, att...
class XLNetForQuestionAnswering (line 1149) | class XLNetForQuestionAnswering(XLNetPreTrainedModel):
method __init__ (line 1208) | def __init__(self, config):
method forward (line 1220) | def forward(self, input_ids, token_type_ids=None, input_mask=None, att...
FILE: LAL-Parser/src_joint/pytorch_transformers/optimization.py
class ConstantLRSchedule (line 26) | class ConstantLRSchedule(LambdaLR):
method __init__ (line 29) | def __init__(self, optimizer, last_epoch=-1):
class WarmupConstantSchedule (line 33) | class WarmupConstantSchedule(LambdaLR):
method __init__ (line 38) | def __init__(self, optimizer, warmup_steps, last_epoch=-1):
method lr_lambda (line 42) | def lr_lambda(self, step):
class WarmupLinearSchedule (line 48) | class WarmupLinearSchedule(LambdaLR):
method __init__ (line 53) | def __init__(self, optimizer, warmup_steps, t_total, last_epoch=-1):
method lr_lambda (line 58) | def lr_lambda(self, step):
class WarmupCosineSchedule (line 64) | class WarmupCosineSchedule(LambdaLR):
method __init__ (line 70) | def __init__(self, optimizer, warmup_steps, t_total, cycles=.5, last_e...
method lr_lambda (line 76) | def lr_lambda(self, step):
class WarmupCosineWithHardRestartsSchedule (line 84) | class WarmupCosineWithHardRestartsSchedule(LambdaLR):
method __init__ (line 90) | def __init__(self, optimizer, warmup_steps, t_total, cycles=1., last_e...
method lr_lambda (line 96) | def lr_lambda(self, step):
class AdamW (line 107) | class AdamW(Optimizer):
method __init__ (line 117) | def __init__(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-6, weig...
method step (line 130) | def step(self, closure=None):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/conftest.py
function pytest_addoption (line 6) | def pytest_addoption(parser):
function pytest_collection_modifyitems (line 12) | def pytest_collection_modifyitems(config, items):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/modeling_auto_test.py
class AutoModelTest (line 30) | class AutoModelTest(unittest.TestCase):
method test_model_from_pretrained (line 31) | def test_model_from_pretrained(self):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/modeling_bert_test.py
class BertModelTest (line 32) | class BertModelTest(CommonTestCases.CommonModelTester):
class BertModelTester (line 38) | class BertModelTester(object):
method __init__ (line 40) | def __init__(self,
method prepare_config_and_inputs (line 87) | def prepare_config_and_inputs(self):
method check_loss_output (line 121) | def check_loss_output(self, result):
method create_and_check_bert_model (line 126) | def create_and_check_bert_model(self, config, input_ids, token_type_...
method create_and_check_bert_for_masked_lm (line 143) | def create_and_check_bert_for_masked_lm(self, config, input_ids, tok...
method create_and_check_bert_for_next_sequence_prediction (line 156) | def create_and_check_bert_for_next_sequence_prediction(self, config,...
method create_and_check_bert_for_pretraining (line 170) | def create_and_check_bert_for_pretraining(self, config, input_ids, t...
method create_and_check_bert_for_question_answering (line 188) | def create_and_check_bert_for_question_answering(self, config, input...
method create_and_check_bert_for_sequence_classification (line 206) | def create_and_check_bert_for_sequence_classification(self, config, ...
method create_and_check_bert_for_token_classification (line 221) | def create_and_check_bert_for_token_classification(self, config, inp...
method create_and_check_bert_for_multiple_choice (line 236) | def create_and_check_bert_for_multiple_choice(self, config, input_id...
method prepare_config_and_inputs_for_common (line 257) | def prepare_config_and_inputs_for_common(self):
method setUp (line 264) | def setUp(self):
method test_config (line 268) | def test_config(self):
method test_bert_model (line 271) | def test_bert_model(self):
method test_for_masked_lm (line 275) | def test_for_masked_lm(self):
method test_for_multiple_choice (line 279) | def test_for_multiple_choice(self):
method test_for_next_sequence_prediction (line 283) | def test_for_next_sequence_prediction(self):
method test_for_pretraining (line 287) | def test_for_pretraining(self):
method test_for_question_answering (line 291) | def test_for_question_answering(self):
method test_for_sequence_classification (line 295) | def test_for_sequence_classification(self):
method test_for_token_classification (line 299) | def test_for_token_classification(self):
method test_model_from_pretrained (line 304) | def test_model_from_pretrained(self):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/modeling_common_test.py
function _config_zero_init (line 35) | def _config_zero_init(config):
class CommonTestCases (line 42) | class CommonTestCases:
class CommonModelTester (line 44) | class CommonModelTester(unittest.TestCase):
method test_initialization (line 52) | def test_initialization(self):
method test_attention_outputs (line 63) | def test_attention_outputs(self):
method test_torchscript (line 101) | def test_torchscript(self):
method test_torchscript_output_attentions (line 106) | def test_torchscript_output_attentions(self):
method test_torchscript_output_hidden_state (line 112) | def test_torchscript_output_hidden_state(self):
method _create_and_check_torchscript (line 118) | def _create_and_check_torchscript(self, config, inputs_dict):
method test_headmasking (line 160) | def test_headmasking(self):
method test_head_pruning (line 206) | def test_head_pruning(self):
method test_hidden_states_output (line 232) | def test_hidden_states_output(self):
method test_resize_tokens_embeddings (line 249) | def test_resize_tokens_embeddings(self):
method test_tie_model_weights (line 283) | def test_tie_model_weights(self):
class GPTModelTester (line 334) | class GPTModelTester(CommonModelTester):
method __init__ (line 336) | def __init__(self,
method prepare_config_and_inputs (line 382) | def prepare_config_and_inputs(self):
method create_and_check_base_model (line 414) | def create_and_check_base_model(self, config, input_ids, token_type_...
method create_and_check_lm_head (line 429) | def create_and_check_lm_head(self, config, input_ids, token_type_ids...
method create_and_check_presents (line 444) | def create_and_check_presents(self, config, input_ids, token_type_id...
method create_and_check_double_heads (line 457) | def create_and_check_double_heads(self, config, input_ids, token_typ...
method create_and_check_model_from_pretrained (line 477) | def create_and_check_model_from_pretrained(self):
method prepare_config_and_inputs_for_common (line 484) | def prepare_config_and_inputs_for_common(self):
method run_common_tests (line 491) | def run_common_tests(self, test_presents=False):
method run_slow_tests (line 505) | def run_slow_tests(self):
class ConfigTester (line 509) | class ConfigTester(object):
method __init__ (line 510) | def __init__(self, parent, config_class=None, **kwargs):
method create_and_test_config_common_properties (line 515) | def create_and_test_config_common_properties(self):
method create_and_test_config_to_json_string (line 522) | def create_and_test_config_to_json_string(self):
method create_and_test_config_to_json_file (line 528) | def create_and_test_config_to_json_file(self):
method run_common_tests (line 536) | def run_common_tests(self):
function ids_tensor (line 544) | def ids_tensor(shape, vocab_size, rng=None, name=None):
class ModelUtilsTest (line 560) | class ModelUtilsTest(unittest.TestCase):
method test_model_from_pretrained (line 561) | def test_model_from_pretrained(self):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/modeling_gpt2_test.py
class GPT2ModelTest (line 28) | class GPT2ModelTest(unittest.TestCase):
method test_config (line 30) | def test_config(self):
method test_model (line 34) | def test_model(self):
method test_pretrained (line 41) | def test_pretrained(self):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/modeling_openai_test.py
class OpenAIModelTest (line 28) | class OpenAIModelTest(unittest.TestCase):
method test_config (line 30) | def test_config(self):
method test_model (line 34) | def test_model(self):
method test_pretrained (line 41) | def test_pretrained(self):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/modeling_roberta_test.py
class RobertaModelTest (line 30) | class RobertaModelTest(CommonTestCases.CommonModelTester):
class RobertaModelTester (line 34) | class RobertaModelTester(object):
method __init__ (line 36) | def __init__(self,
method prepare_config_and_inputs (line 83) | def prepare_config_and_inputs(self):
method check_loss_output (line 117) | def check_loss_output(self, result):
method create_and_check_roberta_model (line 122) | def create_and_check_roberta_model(self, config, input_ids, token_ty...
method create_and_check_roberta_for_masked_lm (line 139) | def create_and_check_roberta_for_masked_lm(self, config, input_ids, ...
method prepare_config_and_inputs_for_common (line 153) | def prepare_config_and_inputs_for_common(self):
method setUp (line 160) | def setUp(self):
method test_config (line 164) | def test_config(self):
method test_roberta_model (line 167) | def test_roberta_model(self):
method test_for_masked_lm (line 171) | def test_for_masked_lm(self):
method test_model_from_pretrained (line 176) | def test_model_from_pretrained(self):
class RobertaModelIntegrationTest (line 185) | class RobertaModelIntegrationTest(unittest.TestCase):
method test_inference_masked_lm (line 188) | def test_inference_masked_lm(self):
method test_inference_no_head (line 209) | def test_inference_no_head(self):
method test_inference_classification_head (line 225) | def test_inference_classification_head(self):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/modeling_transfo_xl_test.py
class TransfoXLModelTest (line 33) | class TransfoXLModelTest(CommonTestCases.CommonModelTester):
class TransfoXLModelTester (line 40) | class TransfoXLModelTester(object):
method __init__ (line 42) | def __init__(self,
method prepare_config_and_inputs (line 82) | def prepare_config_and_inputs(self):
method set_seed (line 105) | def set_seed(self):
method create_transfo_xl_model (line 109) | def create_transfo_xl_model(self, config, input_ids_1, input_ids_2, ...
method check_transfo_xl_model_output (line 123) | def check_transfo_xl_model_output(self, result):
method create_transfo_xl_lm_head (line 138) | def create_transfo_xl_lm_head(self, config, input_ids_1, input_ids_2...
method check_transfo_xl_lm_head_output (line 157) | def check_transfo_xl_lm_head_output(self, result):
method prepare_config_and_inputs_for_common (line 178) | def prepare_config_and_inputs_for_common(self):
method setUp (line 185) | def setUp(self):
method test_config (line 189) | def test_config(self):
method test_transfo_xl_model (line 192) | def test_transfo_xl_model(self):
method test_transfo_xl_lm_head (line 198) | def test_transfo_xl_lm_head(self):
method test_model_from_pretrained (line 205) | def test_model_from_pretrained(self):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/modeling_xlm_test.py
class XLMModelTest (line 29) | class XLMModelTest(CommonTestCases.CommonModelTester):
class XLMModelTester (line 35) | class XLMModelTester(object):
method __init__ (line 37) | def __init__(self,
method prepare_config_and_inputs (line 97) | def prepare_config_and_inputs(self):
method check_loss_output (line 137) | def check_loss_output(self, result):
method create_and_check_xlm_model (line 142) | def create_and_check_xlm_model(self, config, input_ids, token_type_i...
method create_and_check_xlm_lm_head (line 157) | def create_and_check_xlm_lm_head(self, config, input_ids, token_type...
method create_and_check_xlm_qa (line 176) | def create_and_check_xlm_qa(self, config, input_ids, token_type_ids,...
method create_and_check_xlm_sequence_classif (line 230) | def create_and_check_xlm_sequence_classif(self, config, input_ids, t...
method prepare_config_and_inputs_for_common (line 250) | def prepare_config_and_inputs_for_common(self):
method setUp (line 257) | def setUp(self):
method test_config (line 261) | def test_config(self):
method test_xlm_model (line 264) | def test_xlm_model(self):
method test_model_from_pretrained (line 284) | def test_model_from_pretrained(self):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/modeling_xlnet_test.py
class XLNetModelTest (line 33) | class XLNetModelTest(CommonTestCases.CommonModelTester):
class XLNetModelTester (line 39) | class XLNetModelTester(object):
method __init__ (line 41) | def __init__(self,
method prepare_config_and_inputs (line 89) | def prepare_config_and_inputs(self):
method set_seed (line 128) | def set_seed(self):
method create_and_check_xlnet_base_model (line 132) | def create_and_check_xlnet_base_model(self, config, input_ids_1, inp...
method create_and_check_xlnet_lm_head (line 154) | def create_and_check_xlnet_lm_head(self, config, input_ids_1, input_...
method create_and_check_xlnet_qa (line 194) | def create_and_check_xlnet_qa(self, config, input_ids_1, input_ids_2...
method create_and_check_xlnet_sequence_classif (line 252) | def create_and_check_xlnet_sequence_classif(self, config, input_ids_...
method prepare_config_and_inputs_for_common (line 276) | def prepare_config_and_inputs_for_common(self):
method setUp (line 285) | def setUp(self):
method test_config (line 289) | def test_config(self):
method test_xlnet_base_model (line 292) | def test_xlnet_base_model(self):
method test_xlnet_lm_head (line 297) | def test_xlnet_lm_head(self):
method test_xlnet_sequence_classif (line 302) | def test_xlnet_sequence_classif(self):
method test_xlnet_qa (line 307) | def test_xlnet_qa(self):
method test_model_from_pretrained (line 313) | def test_model_from_pretrained(self):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/optimization_test.py
function unwrap_schedule (line 30) | def unwrap_schedule(scheduler, num_steps=10):
function unwrap_and_save_reload_schedule (line 37) | def unwrap_and_save_reload_schedule(scheduler, num_steps=10):
class OptimizationTest (line 51) | class OptimizationTest(unittest.TestCase):
method assertListAlmostEqual (line 53) | def assertListAlmostEqual(self, list1, list2, tol):
method test_adam_w (line 58) | def test_adam_w(self):
class ScheduleInitTest (line 73) | class ScheduleInitTest(unittest.TestCase):
method assertListAlmostEqual (line 78) | def assertListAlmostEqual(self, list1, list2, tol):
method test_constant_scheduler (line 83) | def test_constant_scheduler(self):
method test_warmup_constant_scheduler (line 94) | def test_warmup_constant_scheduler(self):
method test_warmup_linear_scheduler (line 105) | def test_warmup_linear_scheduler(self):
method test_warmup_cosine_scheduler (line 116) | def test_warmup_cosine_scheduler(self):
method test_warmup_cosine_hard_restart_scheduler (line 127) | def test_warmup_cosine_hard_restart_scheduler(self):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/tokenization_auto_test.py
class AutoTokenizerTest (line 29) | class AutoTokenizerTest(unittest.TestCase):
method test_tokenizer_from_pretrained (line 30) | def test_tokenizer_from_pretrained(self):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/tokenization_bert_test.py
class BertTokenizationTest (line 29) | class BertTokenizationTest(CommonTestCases.CommonTokenizerTester):
method setUp (line 33) | def setUp(self):
method get_tokenizer (line 44) | def get_tokenizer(self):
method get_input_output_texts (line 47) | def get_input_output_texts(self):
method test_full_tokenizer (line 52) | def test_full_tokenizer(self):
method test_chinese (line 59) | def test_chinese(self):
method test_basic_tokenizer_lower (line 66) | def test_basic_tokenizer_lower(self):
method test_basic_tokenizer_no_lower (line 74) | def test_basic_tokenizer_no_lower(self):
method test_wordpiece_tokenizer (line 81) | def test_wordpiece_tokenizer(self):
method test_is_whitespace (line 101) | def test_is_whitespace(self):
method test_is_control (line 111) | def test_is_control(self):
method test_is_punctuation (line 119) | def test_is_punctuation(self):
method test_sequence_builders (line 128) | def test_sequence_builders(self):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/tokenization_gpt2_test.py
class GPT2TokenizationTest (line 25) | class GPT2TokenizationTest(CommonTestCases.CommonTokenizerTester):
method setUp (line 29) | def setUp(self):
method get_tokenizer (line 47) | def get_tokenizer(self):
method get_input_output_texts (line 50) | def get_input_output_texts(self):
method test_full_tokenizer (line 55) | def test_full_tokenizer(self):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/tokenization_openai_test.py
class OpenAIGPTTokenizationTest (line 26) | class OpenAIGPTTokenizationTest(CommonTestCases.CommonTokenizerTester):
method setUp (line 30) | def setUp(self):
method get_tokenizer (line 48) | def get_tokenizer(self):
method get_input_output_texts (line 51) | def get_input_output_texts(self):
method test_full_tokenizer (line 57) | def test_full_tokenizer(self):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/tokenization_roberta_test.py
class RobertaTokenizationTest (line 25) | class RobertaTokenizationTest(CommonTestCases.CommonTokenizerTester):
method setUp (line 28) | def setUp(self):
method get_tokenizer (line 46) | def get_tokenizer(self):
method get_input_output_texts (line 49) | def get_input_output_texts(self):
method test_full_tokenizer (line 54) | def test_full_tokenizer(self):
method roberta_dict_integration_testing (line 66) | def roberta_dict_integration_testing(self):
method test_sequence_builders (line 78) | def test_sequence_builders(self):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/tokenization_tests_commons.py
class TemporaryDirectory (line 27) | class TemporaryDirectory(object):
method __enter__ (line 29) | def __enter__(self):
method __exit__ (line 32) | def __exit__(self, exc_type, exc_value, traceback):
class CommonTestCases (line 40) | class CommonTestCases:
class CommonTokenizerTester (line 42) | class CommonTokenizerTester(unittest.TestCase):
method setUp (line 46) | def setUp(self):
method tearDown (line 49) | def tearDown(self):
method get_tokenizer (line 52) | def get_tokenizer(self):
method get_input_output_texts (line 55) | def get_input_output_texts(self):
method test_save_and_load_tokenizer (line 58) | def test_save_and_load_tokenizer(self):
method test_pickle_tokenizer (line 70) | def test_pickle_tokenizer(self):
method test_add_tokens_tokenizer (line 89) | def test_add_tokens_tokenizer(self):
method test_required_methods_tokenizer (line 135) | def test_required_methods_tokenizer(self):
method test_pretrained_model_lists (line 153) | def test_pretrained_model_lists(self):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/tokenization_transfo_xl_test.py
class TransfoXLTokenizationTest (line 25) | class TransfoXLTokenizationTest(CommonTestCases.CommonTokenizerTester):
method setUp (line 29) | def setUp(self):
method get_tokenizer (line 40) | def get_tokenizer(self):
method get_input_output_texts (line 43) | def get_input_output_texts(self):
method test_full_tokenizer (line 48) | def test_full_tokenizer(self):
method test_full_tokenizer_lower (line 57) | def test_full_tokenizer_lower(self):
method test_full_tokenizer_no_lower (line 64) | def test_full_tokenizer_no_lower(self):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/tokenization_utils_test.py
class TokenizerUtilsTest (line 25) | class TokenizerUtilsTest(unittest.TestCase):
method check_tokenizer_from_pretrained (line 26) | def check_tokenizer_from_pretrained(self, tokenizer_class):
method test_pretrained_tokenizers (line 42) | def test_pretrained_tokenizers(self):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/tokenization_xlm_test.py
class XLMTokenizationTest (line 25) | class XLMTokenizationTest(CommonTestCases.CommonTokenizerTester):
method setUp (line 29) | def setUp(self):
method get_tokenizer (line 47) | def get_tokenizer(self):
method get_input_output_texts (line 50) | def get_input_output_texts(self):
method test_full_tokenizer (line 55) | def test_full_tokenizer(self):
method test_sequence_builders (line 69) | def test_sequence_builders(self):
FILE: LAL-Parser/src_joint/pytorch_transformers/tests/tokenization_xlnet_test.py
class XLNetTokenizationTest (line 27) | class XLNetTokenizationTest(CommonTestCases.CommonTokenizerTester):
method setUp (line 31) | def setUp(self):
method get_tokenizer (line 38) | def get_tokenizer(self):
method get_input_output_texts (line 41) | def get_input_output_texts(self):
method test_full_tokenizer (line 47) | def test_full_tokenizer(self):
method test_tokenizer_lower (line 75) | def test_tokenizer_lower(self):
method test_tokenizer_no_lower (line 84) | def test_tokenizer_no_lower(self):
method test_sequence_builders (line 92) | def test_sequence_builders(self):
FILE: LAL-Parser/src_joint/pytorch_transformers/tokenization_auto.py
class AutoTokenizer (line 31) | class AutoTokenizer(object):
method __init__ (line 52) | def __init__(self):
method from_pretrained (line 57) | def from_pretrained(cls, pretrained_model_name_or_path, *inputs, **kwa...
FILE: LAL-Parser/src_joint/pytorch_transformers/tokenization_bert.py
function load_vocab (line 66) | def load_vocab(vocab_file):
function whitespace_tokenize (line 77) | def whitespace_tokenize(text):
class BertTokenizer (line 86) | class BertTokenizer(PreTrainedTokenizer):
method __init__ (line 105) | def __init__(self, vocab_file, do_lower_case=True, do_basic_tokenize=T...
method vocab_size (line 143) | def vocab_size(self):
method _tokenize (line 146) | def _tokenize(self, text):
method _convert_token_to_id (line 156) | def _convert_token_to_id(self, token):
method _convert_id_to_token (line 160) | def _convert_id_to_token(self, index):
method convert_tokens_to_string (line 164) | def convert_tokens_to_string(self, tokens):
method add_special_tokens_single_sentence (line 169) | def add_special_tokens_single_sentence(self, token_ids):
method add_special_tokens_sentences_pair (line 176) | def add_special_tokens_sentences_pair(self, token_ids_0, token_ids_1):
method save_vocabulary (line 185) | def save_vocabulary(self, vocab_path):
method from_pretrained (line 201) | def from_pretrained(cls, pretrained_model_name_or_path, *inputs, **kwa...
class BasicTokenizer (line 219) | class BasicTokenizer(object):
method __init__ (line 222) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
method tokenize (line 242) | def tokenize(self, text, never_split=None):
method _run_strip_accents (line 273) | def _run_strip_accents(self, text):
method _run_split_on_punc (line 284) | def _run_split_on_punc(self, text, never_split=None):
method _tokenize_chinese_chars (line 306) | def _tokenize_chinese_chars(self, text):
method _is_chinese_char (line 319) | def _is_chinese_char(self, cp):
method _clean_text (line 341) | def _clean_text(self, text):
class WordpieceTokenizer (line 355) | class WordpieceTokenizer(object):
method __init__ (line 358) | def __init__(self, vocab, unk_token, max_input_chars_per_word=100):
method tokenize (line 363) | def tokenize(self, text):
function _is_whitespace (line 415) | def _is_whitespace(char):
function _is_control (line 427) | def _is_control(char):
function _is_punctuation (line 439) | def _is_punctuation(char):
FILE: LAL-Parser/src_joint/pytorch_transformers/tokenization_gpt2.py
function lru_cache (line 31) | def lru_cache():
function bytes_to_unicode (line 62) | def bytes_to_unicode():
function get_pairs (line 84) | def get_pairs(word):
class GPT2Tokenizer (line 96) | class GPT2Tokenizer(PreTrainedTokenizer):
method __init__ (line 105) | def __init__(self, vocab_file, merges_file, errors='replace', unk_toke...
method vocab_size (line 123) | def vocab_size(self):
method bpe (line 126) | def bpe(self, token):
method _tokenize (line 167) | def _tokenize(self, text):
method _convert_token_to_id (line 178) | def _convert_token_to_id(self, token):
method _convert_id_to_token (line 182) | def _convert_id_to_token(self, index):
method convert_tokens_to_string (line 186) | def convert_tokens_to_string(self, tokens):
method save_vocabulary (line 192) | def save_vocabulary(self, save_directory):
FILE: LAL-Parser/src_joint/pytorch_transformers/tokenization_openai.py
function get_pairs (line 50) | def get_pairs(word):
function text_standardize (line 62) | def text_standardize(text):
class OpenAIGPTTokenizer (line 77) | class OpenAIGPTTokenizer(PreTrainedTokenizer):
method __init__ (line 87) | def __init__(self, vocab_file, merges_file, unk_token="<unk>", **kwargs):
method vocab_size (line 108) | def vocab_size(self):
method bpe (line 111) | def bpe(self, token):
method _tokenize (line 154) | def _tokenize(self, text):
method _convert_token_to_id (line 169) | def _convert_token_to_id(self, token):
method _convert_id_to_token (line 173) | def _convert_id_to_token(self, index):
method convert_tokens_to_string (line 177) | def convert_tokens_to_string(self, tokens):
method save_vocabulary (line 182) | def save_vocabulary(self, save_directory):
FILE: LAL-Parser/src_joint/pytorch_transformers/tokenization_roberta.py
function lru_cache (line 34) | def lru_cache():
class RobertaTokenizer (line 66) | class RobertaTokenizer(PreTrainedTokenizer):
method __init__ (line 74) | def __init__(self, vocab_file, merges_file, errors='replace', bos_toke...
method vocab_size (line 94) | def vocab_size(self):
method bpe (line 97) | def bpe(self, token):
method _tokenize (line 138) | def _tokenize(self, text):
method _convert_token_to_id (line 149) | def _convert_token_to_id(self, token):
method _convert_id_to_token (line 153) | def _convert_id_to_token(self, index):
method convert_tokens_to_string (line 157) | def convert_tokens_to_string(self, tokens):
method add_special_tokens_single_sentence (line 163) | def add_special_tokens_single_sentence(self, token_ids):
method add_special_tokens_sentences_pair (line 170) | def add_special_tokens_sentences_pair(self, token_ids_0, token_ids_1):
method save_vocabulary (line 179) | def save_vocabulary(self, save_directory):
FILE: LAL-Parser/src_joint/pytorch_transformers/tokenization_transfo_xl.py
class TransfoXLTokenizer (line 61) | class TransfoXLTokenizer(PreTrainedTokenizer):
method __init__ (line 69) | def __init__(self, special=None, min_freq=0, max_size=None, lower_case...
method count_file (line 99) | def count_file(self, path, verbose=False, add_eos=False):
method count_sents (line 114) | def count_sents(self, sents, verbose=False):
method _build_from_file (line 124) | def _build_from_file(self, vocab_file):
method save_vocabulary (line 139) | def save_vocabulary(self, vocab_path):
method build_vocab (line 146) | def build_vocab(self):
method encode_file (line 167) | def encode_file(self, path, ordered=False, verbose=False, add_eos=True,
method encode_sents (line 185) | def encode_sents(self, sents, ordered=False, verbose=False):
method add_special (line 198) | def add_special(self, sym):
method add_symbol (line 204) | def add_symbol(self, sym):
method _convert_id_to_token (line 209) | def _convert_id_to_token(self, idx):
method _convert_token_to_id (line 214) | def _convert_token_to_id(self, sym):
method convert_tokens_to_string (line 231) | def convert_tokens_to_string(self, tokens):
method convert_to_tensor (line 236) | def convert_to_tensor(self, symbols):
method vocab_size (line 240) | def vocab_size(self):
method _tokenize (line 243) | def _tokenize(self, line, add_eos=False, add_double_eos=False):
class LMOrderedIterator (line 263) | class LMOrderedIterator(object):
method __init__ (line 264) | def __init__(self, data, bsz, bptt, device='cpu', ext_len=None):
method get_batch (line 286) | def get_batch(self, i, bptt=None):
method get_fixlen_iter (line 301) | def get_fixlen_iter(self, start=0):
method get_varlen_iter (line 305) | def get_varlen_iter(self, start=0, std=5, min_len=5, max_deviation=3):
method __iter__ (line 317) | def __iter__(self):
class LMShuffledIterator (line 321) | class LMShuffledIterator(object):
method __init__ (line 322) | def __init__(self, data, bsz, bptt, device='cpu', ext_len=None, shuffl...
method get_sent_stream (line 335) | def get_sent_stream(self):
method stream_iterator (line 344) | def stream_iterator(self, sent_stream):
method __iter__ (line 393) | def __iter__(self):
class LMMultiFileIterator (line 401) | class LMMultiFileIterator(LMShuffledIterator):
method __init__ (line 402) | def __init__(self, paths, vocab, bsz, bptt, device='cpu', ext_len=None,
method get_sent_stream (line 415) | def get_sent_stream(self, path):
method __iter__ (line 423) | def __iter__(self):
class TransfoXLCorpus (line 434) | class TransfoXLCorpus(object):
method from_pretrained (line 436) | def from_pretrained(cls, pretrained_model_name_or_path, cache_dir=None...
method __init__ (line 478) | def __init__(self, *args, **kwargs):
method build_corpus (line 485) | def build_corpus(self, path, dataset):
method get_iterator (line 524) | def get_iterator(self, split, *args, **kwargs):
function get_lm_corpus (line 541) | def get_lm_corpus(datadir, dataset):
FILE: LAL-Parser/src_joint/pytorch_transformers/tokenization_utils.py
class PreTrainedTokenizer (line 32) | class PreTrainedTokenizer(object):
method bos_token (line 71) | def bos_token(self):
method eos_token (line 78) | def eos_token(self):
method unk_token (line 85) | def unk_token(self):
method sep_token (line 92) | def sep_token(self):
method pad_token (line 99) | def pad_token(self):
method cls_token (line 106) | def cls_token(self):
method mask_token (line 113) | def mask_token(self):
method additional_special_tokens (line 120) | def additional_special_tokens(self):
method bos_token (line 127) | def bos_token(self, value):
method eos_token (line 131) | def eos_token(self, value):
method unk_token (line 135) | def unk_token(self, value):
method sep_token (line 139) | def sep_token(self, value):
method pad_token (line 143) | def pad_token(self, value):
method cls_token (line 147) | def cls_token(self, value):
method mask_token (line 151) | def mask_token(self, value):
method additional_special_tokens (line 155) | def additional_special_tokens(self, value):
method __init__ (line 158) | def __init__(self, max_len=None, **kwargs):
method from_pretrained (line 182) | def from_pretrained(cls, *inputs, **kwargs):
method _from_pretrained (line 224) | def _from_pretrained(cls, pretrained_model_name_or_path, *inputs, **kw...
method save_pretrained (line 339) | def save_pretrained(self, save_directory):
method save_vocabulary (line 367) | def save_vocabulary(self, save_directory):
method vocab_size (line 376) | def vocab_size(self):
method __len__ (line 381) | def __len__(self):
method add_tokens (line 386) | def add_tokens(self, new_tokens):
method add_special_tokens (line 426) | def add_special_tokens(self, special_tokens_dict):
method tokenize (line 473) | def tokenize(self, text, **kwargs):
method _tokenize (line 494) | def _tokenize(self, text, **kwargs):
method convert_tokens_to_ids (line 503) | def convert_tokens_to_ids(self, tokens):
method _convert_token_to_id_with_added_voc (line 519) | def _convert_token_to_id_with_added_voc(self, token):
method _convert_token_to_id (line 524) | def _convert_token_to_id(self, token):
method encode (line 527) | def encode(self, text, text_pair=None, add_special_tokens=False):
method add_special_tokens_single_sentence (line 553) | def add_special_tokens_single_sentence(self, token_ids):
method add_special_tokens_sentences_pair (line 556) | def add_special_tokens_sentences_pair(self, token_ids_0, token_ids_1):
method convert_ids_to_tokens (line 559) | def convert_ids_to_tokens(self, ids, skip_special_tokens=False):
method _convert_id_to_token (line 581) | def _convert_id_to_token(self, index):
method convert_tokens_to_string (line 584) | def convert_tokens_to_string(self, tokens):
method decode (line 591) | def decode(self, token_ids, skip_special_tokens=False, clean_up_tokeni...
method special_tokens_map (line 616) | def special_tokens_map(self):
method all_special_tokens (line 628) | def all_special_tokens(self):
method all_special_ids (line 640) | def all_special_ids(self):
method clean_up_tokenization (line 649) | def clean_up_tokenization(out_string):
FILE: LAL-Parser/src_joint/pytorch_transformers/tokenization_xlm.py
function get_pairs (line 71) | def get_pairs(word):
function text_standardize (line 83) | def text_standardize(text):
class XLMTokenizer (line 98) | class XLMTokenizer(PreTrainedTokenizer):
method __init__ (line 115) | def __init__(self, vocab_file, merges_file, unk_token="<unk>", bos_tok...
method vocab_size (line 143) | def vocab_size(self):
method bpe (line 146) | def bpe(self, token):
method _tokenize (line 189) | def _tokenize(self, text):
method _convert_token_to_id (line 204) | def _convert_token_to_id(self, token):
method _convert_id_to_token (line 208) | def _convert_id_to_token(self, index):
method convert_tokens_to_string (line 212) | def convert_tokens_to_string(self, tokens):
method add_special_tokens_single_sentence (line 217) | def add_special_tokens_single_sentence(self, token_ids):
method add_special_tokens_sentences_pair (line 224) | def add_special_tokens_sentences_pair(self, token_ids_0, token_ids_1):
method save_vocabulary (line 233) | def save_vocabulary(self, save_directory):
FILE: LAL-Parser/src_joint/pytorch_transformers/tokenization_xlnet.py
class XLNetTokenizer (line 54) | class XLNetTokenizer(PreTrainedTokenizer):
method __init__ (line 64) | def __init__(self, vocab_file, max_len=None,
method vocab_size (line 89) | def vocab_size(self):
method __getstate__ (line 92) | def __getstate__(self):
method __setstate__ (line 97) | def __setstate__(self, d):
method preprocess_text (line 107) | def preprocess_text(self, inputs):
method _tokenize (line 125) | def _tokenize(self, text, return_unicode=True, sample=False):
method _convert_token_to_id (line 164) | def _convert_token_to_id(self, token):
method _convert_id_to_token (line 168) | def _convert_id_to_token(self, index, return_unicode=True):
method convert_tokens_to_string (line 175) | def convert_tokens_to_string(self, tokens):
method add_special_tokens_single_sentence (line 180) | def add_special_tokens_single_sentence(self, token_ids):
method add_special_tokens_sentences_pair (line 189) | def add_special_tokens_sentences_pair(self, token_ids_0, token_ids_1):
method save_vocabulary (line 198) | def save_vocabulary(self, save_directory):
FILE: LAL-Parser/src_joint/transliterate.py
function arabic (line 129) | def arabic(inp):
function hebrew (line 142) | def hebrew(inp):
FILE: LAL-Parser/src_joint/trees.py
class TreebankNode (line 8) | class TreebankNode(object):
class InternalTreebankNode (line 11) | class InternalTreebankNode(TreebankNode):
method __init__ (line 12) | def __init__(self, label, children):
method linearize (line 39) | def linearize(self):
method leaves (line 43) | def leaves(self):
method convert (line 47) | def convert(self, index=0, nocache=False):
class LeafTreebankNode (line 99) | class LeafTreebankNode(TreebankNode):
method __init__ (line 100) | def __init__(self, tag, word, head, father, type):
method linearize (line 111) | def linearize(self):
method leaves (line 114) | def leaves(self):
method convert (line 117) | def convert(self, index=0):
class ParseNode (line 120) | class ParseNode(object):
class InternalParseNode (line 123) | class InternalParseNode(ParseNode):
method __init__ (line 124) | def __init__(self, label, children, nocache=False):
method leaves (line 163) | def leaves(self):
method convert (line 167) | def convert(self):
method enclosing (line 174) | def enclosing(self, left, right):
method chil_enclosing (line 183) | def chil_enclosing(self, left, right):
method oracle_label (line 190) | def oracle_label(self, left, right):
method oracle_type (line 196) | def oracle_type(self, left, right):
method oracle_head (line 200) | def oracle_head(self, left, right):
method oracle_splits (line 204) | def oracle_splits(self, left, right):
class LeafParseNode (line 211) | class LeafParseNode(ParseNode):
method __init__ (line 212) | def __init__(self, index, tag, word, father, type):
method leaves (line 227) | def leaves(self):
method chil_enclosing (line 230) | def chil_enclosing(self, left, right):
method convert (line 234) | def convert(self):
function load_trees (line 237) | def load_trees(path, heads = None, types = None, wordss = None, strip_to...
FILE: LAL-Parser/src_joint/utils.py
function load_embedding_dict (line 10) | def load_embedding_dict(embedding, embedding_path, normalize_digits= Fal...
FILE: LAL-Parser/src_joint/vocabulary.py
class Vocabulary (line 3) | class Vocabulary(object):
method __init__ (line 4) | def __init__(self):
method size (line 11) | def size(self):
method value (line 14) | def value(self, index):
method index (line 18) | def index(self, value):
method index_or_unk (line 33) | def index_or_unk(self, value, unk_value):
method count (line 40) | def count(self, value):
method freeze (line 43) | def freeze(self):
Copy disabled (too large)
Download .json
Condensed preview — 133 files, each showing path, character count, and a content snippet. Download the .json file for the full structured content (34,425K chars).
[
{
"path": "DualGCN/build_vocab.sh",
"chars": 1159,
"preview": "#!/bin/bash\n# build vocab for different datasets\n\npython ./DualGCN/prepare_vocab.py --data_dir DualGCN/dataset/Restauran"
},
{
"path": "DualGCN/data_utils.py",
"chars": 17233,
"preview": "'''\nDescription: \nversion: \nAuthor: chenhao\nDate: 2021-06-09 14:17:37\n'''\nimport os\nimport sys\nsys.path.append(r'./LAL-P"
},
{
"path": "DualGCN/dataset/Laptops_allennlp/test.json",
"chars": 695265,
"preview": "[\n {\n \"token\": [\n \"Boot\",\n \"time\",\n \"is\",\n \"super\",\n \"f"
},
{
"path": "DualGCN/dataset/Laptops_allennlp/train.json",
"chars": 2870109,
"preview": "[\n {\n \"token\": [\n \"I\",\n \"charge\",\n \"it\",\n \"at\",\n \"night"
},
{
"path": "DualGCN/dataset/Laptops_corenlp/test.json",
"chars": 718961,
"preview": "[\r\n {\r\n \"token\": [\r\n \"Boot\",\r\n \"time\",\r\n \"is\",\r\n \"super\",\r\n "
},
{
"path": "DualGCN/dataset/Laptops_corenlp/train.json",
"chars": 2973937,
"preview": "[\r\n {\r\n \"token\": [\r\n \"I\",\r\n \"charge\",\r\n \"it\",\r\n \"at\",\r\n "
},
{
"path": "DualGCN/dataset/Laptops_stanza/test.json",
"chars": 687345,
"preview": "[\n {\n \"token\": [\n \"Boot\",\n \"time\",\n \"is\",\n \"super\",\n \"f"
},
{
"path": "DualGCN/dataset/Laptops_stanza/train.json",
"chars": 2839121,
"preview": "[\n {\n \"token\": [\n \"I\",\n \"charge\",\n \"it\",\n \"at\",\n \"night"
},
{
"path": "DualGCN/dataset/Restaurants_allennlp/test.json",
"chars": 1076979,
"preview": "[\n {\n \"token\": [\n \"The\",\n \"bread\",\n \"is\",\n \"top\",\n \"not"
},
{
"path": "DualGCN/dataset/Restaurants_allennlp/train.json",
"chars": 3653908,
"preview": "[\n {\n \"token\": [\n \"But\",\n \"the\",\n \"staff\",\n \"was\",\n \"so"
},
{
"path": "DualGCN/dataset/Restaurants_corenlp/test.json",
"chars": 1110951,
"preview": "[\r\n {\r\n \"token\": [\r\n \"The\",\r\n \"bread\",\r\n \"is\",\r\n \"top\",\r\n "
},
{
"path": "DualGCN/dataset/Restaurants_corenlp/train.json",
"chars": 3778504,
"preview": "[\r\n {\r\n \"token\": [\r\n \"But\",\r\n \"the\",\r\n \"staff\",\r\n \"was\",\r\n "
},
{
"path": "DualGCN/dataset/Restaurants_stanza/test.json",
"chars": 1062625,
"preview": "[\n {\n \"token\": [\n \"The\",\n \"bread\",\n \"is\",\n \"top\",\n \"not"
},
{
"path": "DualGCN/dataset/Restaurants_stanza/train.json",
"chars": 3604337,
"preview": "[\n {\n \"token\": [\n \"But\",\n \"the\",\n \"staff\",\n \"was\",\n \"so"
},
{
"path": "DualGCN/dataset/Tweets_allennlp/test.json",
"chars": 1351826,
"preview": "[\n {\n \"token\": [\n \"shaquille\",\n \"o'neal\",\n \"to\",\n \"miss\",\n "
},
{
"path": "DualGCN/dataset/Tweets_corenlp/test.json",
"chars": 1330823,
"preview": "[\n {\n \"token\": [\n \"shaquille\",\n \"o'neal\",\n \"to\",\n \"miss\",\n "
},
{
"path": "DualGCN/dataset/Tweets_stanza/test.json",
"chars": 1340201,
"preview": "[\n {\n \"token\": [\n \"shaquille\",\n \"o'neal\",\n \"to\",\n \"miss\",\n "
},
{
"path": "DualGCN/glove/readme.md",
"chars": 341,
"preview": "### Please put the pre-trained word vectors in this folder.\n\n[https://nlp.stanford.edu/projects/glove/](https://nlp.stan"
},
{
"path": "DualGCN/layers.py",
"chars": 7837,
"preview": "import math\nimport torch\nimport torch.nn as nn\nimport torch.nn.functional as F\n\nclass DynamicLSTM(nn.Module):\n '''\n "
},
{
"path": "DualGCN/models/atae_lstm.py",
"chars": 1482,
"preview": "'''\nDescription: \nversion: \nAuthor: chenhao\nDate: 2021-06-09 14:17:37\n'''\nimport torch\nimport torch.nn as nn\nimport torc"
},
{
"path": "DualGCN/models/dualgcn.py",
"chars": 9891,
"preview": "'''\nDescription: \nversion: \nAuthor: chenhao\nDate: 2021-06-09 14:17:37\n'''\nimport copy\nimport math\nimport torch\nimport nu"
},
{
"path": "DualGCN/models/dualgcn_bert.py",
"chars": 7858,
"preview": "'''\nDescription: \nversion: \nAuthor: chenhao\nDate: 2021-06-09 14:17:37\n'''\nimport copy\nimport math\nimport torch\nimport nu"
},
{
"path": "DualGCN/models/ian.py",
"chars": 2079,
"preview": "'''\nDescription: \nversion: \nAuthor: chenhao\nDate: 2021-06-09 14:17:37\n'''\nimport torch\nimport torch.nn as nn\nimport torc"
},
{
"path": "DualGCN/models/semgcn.py",
"chars": 7108,
"preview": "'''\nDescription: \nversion: \nAuthor: chenhao\nDate: 2021-06-09 14:17:37\n'''\nimport copy\nimport math\nimport torch\nimport nu"
},
{
"path": "DualGCN/models/syngcn.py",
"chars": 5273,
"preview": "'''\nDescription: \nversion: \nAuthor: chenhao\nDate: 2021-06-09 14:17:37\n'''\nimport torch\nimport numpy as np\nimport torch.n"
},
{
"path": "DualGCN/prepare_vocab.py",
"chars": 4522,
"preview": "\"\"\"\nPrepare vocabulary and initial word vectors.\n\"\"\"\nimport json\nimport tqdm\nimport pickle\nimport argparse\nimport numpy "
},
{
"path": "DualGCN/run.sh",
"chars": 1554,
"preview": "#!/bin/bash\n\n# * laptop\n\n# * DualGCN\nCUDA_VISIBLE_DEVICES=0 python ./DualGCN/train.py --model_name dualgcn --dataset lap"
},
{
"path": "DualGCN/train.py",
"chars": 18411,
"preview": "'''\nDescription: \nversion: \nAuthor: chenhao\nDate: 2021-06-09 14:17:37\n'''\nimport os\nimport sys\nimport copy\nimport random"
},
{
"path": "DualGCN/tree.py",
"chars": 2386,
"preview": "\"\"\"\nBasic operations on trees.\n\"\"\"\n\nimport numpy as np\nfrom collections import defaultdict\n\nimport copy\n\nclass Tree(obje"
},
{
"path": "LAL-Parser/EVALB/COLLINS.prm",
"chars": 2374,
"preview": "##------------------------------------------##\n## Debug mode ##\n## 0: No debugging "
},
{
"path": "LAL-Parser/EVALB/LICENSE",
"chars": 1211,
"preview": "This is free and unencumbered software released into the public domain.\n\nAnyone is free to copy, modify, publish, use, c"
},
{
"path": "LAL-Parser/EVALB/Makefile",
"chars": 58,
"preview": "all: evalb\n\nevalb: evalb.c\n\tgcc -Wall -g -o evalb evalb.c\n"
},
{
"path": "LAL-Parser/EVALB/README",
"chars": 12140,
"preview": "#################################################################\n# "
},
{
"path": "LAL-Parser/EVALB/bug/bug.gld",
"chars": 2795,
"preview": "(TOP (S (NP-SBJ (DT The) (NN Thy-1) (NN gene) (NN promoter) ) (VP (VBZ resembles) (NP (DT a) (`` \") (JJ housekeepi"
},
{
"path": "LAL-Parser/EVALB/bug/bug.rsl-new",
"chars": 1629,
"preview": "Sent. Matched Bracket Cross Correct Tag\nID Len. Stat. Recal Prec. Bracket gold test"
},
{
"path": "LAL-Parser/EVALB/bug/bug.rsl-old",
"chars": 1729,
"preview": "Sent. Matched Bracket Cross Correct Tag\nID Len. Stat. Recal Prec. Bracket gold test"
},
{
"path": "LAL-Parser/EVALB/bug/bug.tst",
"chars": 2553,
"preview": "(S1 (S (NP (DT The) (JJ Thy-1) (NN gene) (NN promoter)) (VP (VP (VBZ resembles) (NP (NP (DT a) (ADJP (CD \") (NN housekee"
},
{
"path": "LAL-Parser/EVALB/evalb.c",
"chars": 39869,
"preview": "/*****************************************************************/\n/* evalb [-p param_file] [-dh] [-e n] gold-file test"
},
{
"path": "LAL-Parser/EVALB/evalb.dSYM/Contents/Info.plist",
"chars": 634,
"preview": "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n<!DOCTYPE plist PUBLIC \"-//Apple Computer//DTD PLIST 1.0//EN\" \"http://www.apple.c"
},
{
"path": "LAL-Parser/EVALB/new.prm",
"chars": 2957,
"preview": "##------------------------------------------##\n## Debug mode ##\n## 0: No debugging "
},
{
"path": "LAL-Parser/EVALB/nk.prm",
"chars": 3227,
"preview": "# Based on new.prm (and by extension COLLINS.prm)\n# The only change from new.prm is increasing MAX_ERROR. The evaluation"
},
{
"path": "LAL-Parser/EVALB/sample/sample.gld",
"chars": 1590,
"preview": "(S (A (P this)) (B (Q is) (A (R a) (T test))))\n(S (A (P this)) (B (Q is) (A (R a) (T test))))\n(S (A (P this)) (B (Q is) "
},
{
"path": "LAL-Parser/EVALB/sample/sample.prm",
"chars": 2301,
"preview": "##------------------------------------------##\n## Debug mode ##\n## print out data for in"
},
{
"path": "LAL-Parser/EVALB/sample/sample.rsl",
"chars": 3044,
"preview": " Sent. Matched Bracket Cross Correct Tag\n ID Len. Stat. Recal Prec. Bracket gold t"
},
{
"path": "LAL-Parser/EVALB/sample/sample.tst",
"chars": 1532,
"preview": "(S (A (P this)) (B (Q is) (A (R a) (T test))))\n(S (A (P this)) (B (Q is) (C (R a) (T test))))\n(S (A (P this)) (B (Q is) "
},
{
"path": "LAL-Parser/EVALB/tgrep_proc.prl",
"chars": 101,
"preview": "#!/usr/local/bin/perl\n\nwhile(<>)\n{\n if(m/TOP/) #skip lines which are blank\n {\n\tprint;\n } \n}\n"
},
{
"path": "LAL-Parser/best_model/readme.md",
"chars": 244,
"preview": "### Please put the best_parser.pt in this folder.\n\nBaidu Netdisk: [[Link]](https://pan.baidu.com/s/1N1C3HFL4Yjn--WO9ImHt"
},
{
"path": "LAL-Parser/src_joint/KM_parser.py",
"chars": 86330,
"preview": "import functools\n\nimport numpy as np\nimport sys\nimport torch\nfrom torch.autograd import Variable\nimport torch.nn as nn\ni"
},
{
"path": "LAL-Parser/src_joint/absa_parser.py",
"chars": 1638,
"preview": "import uuid\nimport torch\nimport numpy as np\nimport KM_parser\ntokens = KM_parser\nimport nltk\n# from nltk import word_toke"
},
{
"path": "LAL-Parser/src_joint/const_decoder.pyx",
"chars": 6504,
"preview": "import numpy as np\ncimport numpy as np\nfrom numpy cimport ndarray\ncimport cython\n\nctypedef np.float32_t DTYPE_t\n\nORACLE_"
},
{
"path": "LAL-Parser/src_joint/dep_eval.py",
"chars": 2371,
"preview": "__author__ = 'max'\n\nimport re\nimport numpy as np\n\ndef is_uni_punctuation(word):\n match = re.match(\"^[^\\w\\s]+$]\", word"
},
{
"path": "LAL-Parser/src_joint/dep_reader.py",
"chars": 2087,
"preview": "__author__ = 'max'\n\nimport KM_parser\ntokens = KM_parser\n\nclass Sentence(object):\n def __init__(self, words, postags):"
},
{
"path": "LAL-Parser/src_joint/evaluate.py",
"chars": 3682,
"preview": "import math\nimport os.path\nimport re\nimport subprocess\nimport tempfile\n\nimport trees\n\nclass FScore(object):\n def __in"
},
{
"path": "LAL-Parser/src_joint/hpsg_decoder.pyx",
"chars": 11979,
"preview": "import numpy as np\ncimport numpy as np\nfrom numpy cimport ndarray\ncimport cython\n\nctypedef np.float32_t DTYPE_t\n\nORACLE_"
},
{
"path": "LAL-Parser/src_joint/main.py",
"chars": 36631,
"preview": "import argparse\nimport itertools\nimport os.path\n# import os\nimport time\nimport uuid\n\nimport torch\nimport torch.optim.lr_"
},
{
"path": "LAL-Parser/src_joint/makehp.py",
"chars": 1700,
"preview": "class HParams():\n _skip_keys = ['populate_arguments', 'set_from_args', 'print', 'to_dict']\n def __init__(self, **k"
},
{
"path": "LAL-Parser/src_joint/pretrained_bert/__init__.py",
"chars": 478,
"preview": "__version__ = \"0.4.0\"\nfrom .tokenization import BertTokenizer, BasicTokenizer, WordpieceTokenizer\nfrom .modeling import "
},
{
"path": "LAL-Parser/src_joint/pretrained_bert/__main__.py",
"chars": 932,
"preview": "# coding: utf8\ndef main():\n import sys\n try:\n from .convert_tf_checkpoint_to_pytorch import convert_tf_chec"
},
{
"path": "LAL-Parser/src_joint/pretrained_bert/convert_tf_checkpoint_to_pytorch.py",
"chars": 4463,
"preview": "# coding=utf-8\n# Copyright 2018 The HugginFace Inc. team.\n#\n# Licensed under the Apache License, Version 2.0 (the \"Licen"
},
{
"path": "LAL-Parser/src_joint/pretrained_bert/file_utils.py",
"chars": 8021,
"preview": "\"\"\"\nUtilities for working with the local dataset cache.\nThis file is adapted from the AllenNLP library at https://github"
},
{
"path": "LAL-Parser/src_joint/pretrained_bert/modeling.py",
"chars": 55243,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors and The HugginFace Inc. team.\n# Copyright (c) 2018, "
},
{
"path": "LAL-Parser/src_joint/pretrained_bert/optimization.py",
"chars": 6803,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors and The HugginFace Inc. team.\n#\n# Licensed under the"
},
{
"path": "LAL-Parser/src_joint/pretrained_bert/tokenization.py",
"chars": 14026,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors and The HugginFace Inc. team.\n#\n# Licensed under the"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/__init__.py",
"chars": 3326,
"preview": "__version__ = \"1.1.0\"\nfrom .tokenization_auto import AutoTokenizer\nfrom .tokenization_bert import BertTokenizer, BasicTo"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/__main__.py",
"chars": 7021,
"preview": "# coding: utf8\ndef main():\n import sys\n if (len(sys.argv) < 4 or len(sys.argv) > 6) or sys.argv[1] not in [\"bert\","
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/convert_gpt2_checkpoint_to_pytorch.py",
"chars": 3081,
"preview": "# coding=utf-8\n# Copyright 2018 The HuggingFace Inc. team.\n#\n# Licensed under the Apache License, Version 2.0 (the \"Lice"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/convert_openai_checkpoint_to_pytorch.py",
"chars": 3170,
"preview": "# coding=utf-8\n# Copyright 2018 The HuggingFace Inc. team.\n#\n# Licensed under the Apache License, Version 2.0 (the \"Lice"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/convert_pytorch_checkpoint_to_tf.py",
"chars": 4495,
"preview": "# coding=utf-8\n# Copyright 2018 The HuggingFace Inc. team.\n#\n# Licensed under the Apache License, Version 2.0 (the \"Lice"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/convert_roberta_checkpoint_to_pytorch.py",
"chars": 8720,
"preview": "# coding=utf-8\n# Copyright 2018 The HuggingFace Inc. team.\n#\n# Licensed under the Apache License, Version 2.0 (the \"Lice"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/convert_tf_checkpoint_to_pytorch.py",
"chars": 2599,
"preview": "# coding=utf-8\n# Copyright 2018 The HuggingFace Inc. team.\n#\n# Licensed under the Apache License, Version 2.0 (the \"Lice"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/convert_transfo_xl_checkpoint_to_pytorch.py",
"chars": 5555,
"preview": "# coding=utf-8\n# Copyright 2018 The HuggingFace Inc. team.\n#\n# Licensed under the Apache License, Version 2.0 (the \"Lice"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/convert_xlm_checkpoint_to_pytorch.py",
"chars": 2955,
"preview": "# coding=utf-8\n# Copyright 2018 The HuggingFace Inc. team.\n#\n# Licensed under the Apache License, Version 2.0 (the \"Lice"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/convert_xlnet_checkpoint_to_pytorch.py",
"chars": 4357,
"preview": "# coding=utf-8\n# Copyright 2018 The HuggingFace Inc. team.\n#\n# Licensed under the Apache License, Version 2.0 (the \"Lice"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/file_utils.py",
"chars": 9100,
"preview": "\"\"\"\nUtilities for working with the local dataset cache.\nThis file is adapted from the AllenNLP library at https://github"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/modeling_auto.py",
"chars": 15108,
"preview": "# coding=utf-8\n# Copyright 2018 The HuggingFace Inc. team.\n#\n# Licensed under the Apache License, Version 2.0 (the \"Lice"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/modeling_bert.py",
"chars": 66519,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.\n# Copyright (c) 2018,"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/modeling_gpt2.py",
"chars": 36429,
"preview": "# coding=utf-8\n# Copyright 2018 The OpenAI Team Authors and HuggingFace Inc. team.\n# Copyright (c) 2018, NVIDIA CORPORAT"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/modeling_openai.py",
"chars": 34785,
"preview": "# coding=utf-8\n# Copyright 2018 The OpenAI Team Authors and HuggingFace Inc. team.\n# Copyright (c) 2018, NVIDIA CORPORAT"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/modeling_roberta.py",
"chars": 18024,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.\n# Copyright (c) 2018,"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/modeling_transfo_xl.py",
"chars": 58525,
"preview": "# coding=utf-8\n# Copyright 2018 Google AI, Google Brain and Carnegie Mellon University Authors and the HuggingFace Inc. "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/modeling_transfo_xl_utilities.py",
"chars": 13568,
"preview": "# coding=utf-8\n# Copyright 2018 Google AI, Google Brain and Carnegie Mellon University Authors and the HuggingFace Inc. "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/modeling_utils.py",
"chars": 47379,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.\n# Copyright (c) 2018,"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/modeling_xlm.py",
"chars": 44843,
"preview": "# coding=utf-8\n# Copyright 2019-present, Facebook, Inc and the HuggingFace Inc. team.\n#\n# Licensed under the Apache Lice"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/modeling_xlnet.py",
"chars": 63593,
"preview": "# coding=utf-8\n# Copyright 2018 Google AI, Google Brain and Carnegie Mellon University Authors and the HuggingFace Inc. "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/optimization.py",
"chars": 8635,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.\n#\n# Licensed under th"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/__init__.py",
"chars": 0,
"preview": ""
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/conftest.py",
"chars": 511,
"preview": "# content of conftest.py\n\nimport pytest\n\n\ndef pytest_addoption(parser):\n parser.addoption(\n \"--runslow\", actio"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/fixtures/input.txt",
"chars": 52,
"preview": "Who was Jim Henson ? ||| Jim Henson was a puppeteer\n"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/fixtures/sample_text.txt",
"chars": 4364,
"preview": "This text is included to make sure Unicode is handled properly: 力加勝北区ᴵᴺᵀᵃছজটডণত\nText should be one-sentence-per-line, wi"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/modeling_auto_test.py",
"chars": 1782,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors.\n#\n# Licensed under the Apache License, Version 2.0 "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/modeling_bert_test.py",
"chars": 14581,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors.\n#\n# Licensed under the Apache License, Version 2.0 "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/modeling_common_test.py",
"chars": 26105,
"preview": "# coding=utf-8\n# Copyright 2019 HuggingFace Inc.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# y"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/modeling_gpt2_test.py",
"chars": 1934,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors.\n#\n# Licensed under the Apache License, Version 2.0 "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/modeling_openai_test.py",
"chars": 2001,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors.\n#\n# Licensed under the Apache License, Version 2.0 "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/modeling_roberta_test.py",
"chars": 10235,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors.\n#\n# Licensed under the Apache License, Version 2.0 "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/modeling_transfo_xl_test.py",
"chars": 8536,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors.\n#\n# Licensed under the Apache License, Version 2.0 "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/modeling_xlm_test.py",
"chars": 12674,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors.\n#\n# Licensed under the Apache License, Version 2.0 "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/modeling_xlnet_test.py",
"chars": 14185,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors.\n#\n# Licensed under the Apache License, Version 2.0 "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/optimization_test.py",
"chars": 6088,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors.\n#\n# Licensed under the Apache License, Version 2.0 "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/tokenization_auto_test.py",
"chars": 1825,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors.\n#\n# Licensed under the Apache License, Version 2.0 "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/tokenization_bert_test.py",
"chars": 5280,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors.\n#\n# Licensed under the Apache License, Version 2.0 "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/tokenization_gpt2_test.py",
"chars": 2656,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors.\n#\n# Licensed under the Apache License, Version 2.0 "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/tokenization_openai_test.py",
"chars": 2636,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors.\n#\n# Licensed under the Apache License, Version 2.0 "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/tokenization_roberta_test.py",
"chars": 3782,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors.\n#\n# Licensed under the Apache License, Version 2.0 "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/tokenization_tests_commons.py",
"chars": 5934,
"preview": "# coding=utf-8\n# Copyright 2019 HuggingFace Inc.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# y"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/tokenization_transfo_xl_test.py",
"chars": 2677,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors.\n#\n# Licensed under the Apache License, Version 2.0 "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/tokenization_utils_test.py",
"chars": 1838,
"preview": "# coding=utf-8\n# Copyright 2018 HuggingFace Inc..\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/tokenization_xlm_test.py",
"chars": 3181,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors.\n#\n# Licensed under the Apache License, Version 2.0 "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tests/tokenization_xlnet_test.py",
"chars": 5242,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors.\n#\n# Licensed under the Apache License, Version 2.0 "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tokenization_auto.py",
"chars": 5711,
"preview": "# coding=utf-8\n# Copyright 2018 The HuggingFace Inc. team.\n#\n# Licensed under the Apache License, Version 2.0 (the \"Lice"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tokenization_bert.py",
"chars": 20131,
"preview": "# coding=utf-8\n# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.\n#\n# Licensed under th"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tokenization_gpt2.py",
"chars": 8187,
"preview": "# coding=utf-8\n# Copyright 2018 The Open AI Team Authors and The HuggingFace Inc. team.\n#\n# Licensed under the Apache Li"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tokenization_openai.py",
"chars": 7494,
"preview": "# coding=utf-8\n# Copyright 2018 The Open AI Team Authors and The HuggingFace Inc. team.\n#\n# Licensed under the Apache Li"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tokenization_roberta.py",
"chars": 8329,
"preview": "# coding=utf-8\n# Copyright 2018 The Open AI Team Authors and The HuggingFace Inc. team.\n#\n# Licensed under the Apache Li"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tokenization_transfo_xl.py",
"chars": 21459,
"preview": "# coding=utf-8\n# Copyright 2018 Google AI, Google Brain and Carnegie Mellon University Authors and the HuggingFace Inc. "
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tokenization_utils.py",
"chars": 31331,
"preview": "# coding=utf-8\n# Copyright 2018 The Open AI Team Authors and The HuggingFace Inc. team.\n#\n# Licensed under the Apache Li"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tokenization_xlm.py",
"chars": 11071,
"preview": "# coding=utf-8\n# Copyright 2019 The Open AI Team Authors and The HuggingFace Inc. team.\n#\n# Licensed under the Apache Li"
},
{
"path": "LAL-Parser/src_joint/pytorch_transformers/tokenization_xlnet.py",
"chars": 8190,
"preview": "# coding=utf-8\n# Copyright 2018 Google AI, Google Brain and Carnegie Mellon University Authors and the HuggingFace Inc. "
},
{
"path": "LAL-Parser/src_joint/test.py",
"chars": 304,
"preview": "import numpy as np\nimport pyximport\npyximport.install(setup_args={\"include_dirs\": np.get_include()})\n#import src_dep_con"
},
{
"path": "LAL-Parser/src_joint/transliterate.py",
"chars": 3105,
"preview": "BUCKWALTER_MAP = {\n '\\'': '\\u0621',\n '|': '\\u0622',\n '>': '\\u0623',\n 'O': '\\u0623',\n '&': '\\u0624',\n '"
},
{
"path": "LAL-Parser/src_joint/trees.py",
"chars": 10384,
"preview": "import collections.abc\nimport gzip\nimport numpy as np\nSub_Head = \"<H>\"\nNo_Head = \"<N>\"\nHtype = 1\nNtype = 0\nclass Treeban"
},
{
"path": "LAL-Parser/src_joint/utils.py",
"chars": 2399,
"preview": "__author__ = 'max'\n\nimport pickle\nimport numpy as np\nimport gzip\n\nimport utils_io\n\n\ndef load_embedding_dict(embedding, e"
},
{
"path": "LAL-Parser/src_joint/utils_io.py",
"chars": 147,
"preview": "__author__ = 'max'\n\nimport re\nMAX_CHAR_LENGTH = 45\nNUM_CHAR_PAD = 2\n\n# Regular expressions used to normalize digits.\nDIG"
},
{
"path": "LAL-Parser/src_joint/vocabulary.py",
"chars": 1101,
"preview": "import collections\n\nclass Vocabulary(object):\n def __init__(self):\n self.frozen = False\n self.values = "
},
{
"path": "LICENSE",
"chars": 1064,
"preview": "MIT License\n\nCopyright (c) 2021 Chenhao\n\nPermission is hereby granted, free of charge, to any person obtaining a copy\nof"
},
{
"path": "README.md",
"chars": 1761,
"preview": "# DualGCN\n\nCode and datasets of our paper \"[Dual Graph Convolutional Networks for Aspect-based Sentiment Analysis](https"
},
{
"path": "requirements.txt",
"chars": 79,
"preview": "torch==1.4.0\nscikit-learn==0.23.2\ntransformers==3.2.0\ncython==0.29.13\nnltk==3.5"
}
]
// ... and 6 more files (download for full content)
About this extraction
This page contains the full source code of the CCChenhao997/DualGCN-ABSA GitHub repository, extracted and formatted as plain text for AI agents and large language models (LLMs). The extraction includes 133 files (62.8 MB), approximately 7.6M tokens, and a symbol index with 1257 extracted functions, classes, methods, constants, and types. Use this with OpenClaw, Claude, ChatGPT, Cursor, Windsurf, or any other AI tool that accepts text input. You can copy the full output to your clipboard or download it as a .txt file.
Extracted by GitExtract — free GitHub repo to text converter for AI. Built by Nikandr Surkov.