File size: 8,822 Bytes
fb238e8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
import csv
import json
import torch
from transformers import BertTokenizer


class CNerTokenizer(BertTokenizer):
    def __init__(self, vocab_file, do_lower_case=True):
        super().__init__(vocab_file=str(vocab_file), do_lower_case=do_lower_case)
        self.vocab_file = str(vocab_file)
        self.do_lower_case = do_lower_case

    def tokenize(self, text):
        _tokens = []
        for c in text:
            if self.do_lower_case:
                c = c.lower()
            if c in self.vocab:
                _tokens.append(c)
            else:
                _tokens.append('[UNK]')
        return _tokens


class DataProcessor(object):
    """Base class for data converters for sequence classification data sets."""

    def get_train_examples(self, data_dir):
        """Gets a collection of `InputExample`s for the train set."""
        raise NotImplementedError()

    def get_dev_examples(self, data_dir):
        """Gets a collection of `InputExample`s for the dev set."""
        raise NotImplementedError()

    def get_labels(self):
        """Gets the list of labels for this data set."""
        raise NotImplementedError()

    @classmethod
    def _read_tsv(cls, input_file, quotechar=None):
        """Reads a tab separated value file."""
        with open(input_file, "r", encoding="utf-8-sig") as f:
            reader = csv.reader(f, delimiter="\t", quotechar=quotechar)
            lines = []
            for line in reader:
                lines.append(line)
            return lines

    @classmethod
    def _read_text(self, input_file):
        lines = []
        with open(input_file, 'r') as f:
            words = []
            labels = []
            for line in f:
                if line.startswith("-DOCSTART-") or line == "" or line == "\n":
                    if words:
                        lines.append({"words": words, "labels": labels})
                        words = []
                        labels = []
                else:
                    splits = line.split(" ")
                    words.append(splits[0])
                    if len(splits) > 1:
                        labels.append(splits[-1].replace("\n", ""))
                    else:
                        # Examples could have no label for mode = "test"
                        labels.append("O")
            if words:
                lines.append({"words": words, "labels": labels})
        return lines

    @classmethod
    def _read_json(self, input_file):
        lines = []
        with open(input_file, 'r', encoding='utf8') as f:
            for line in f:
                line = json.loads(line.strip())
                text = line['text']
                label_entities = line.get('label', None)
                words = list(text)
                labels = ['O'] * len(words)
                if label_entities is not None:
                    for key, value in label_entities.items():
                        for sub_name, sub_index in value.items():
                            for start_index, end_index in sub_index:
                                assert ''.join(words[start_index:end_index+1]) == sub_name
                                if start_index == end_index:
                                    labels[start_index] = 'S-'+key
                                else:
                                    if end_index - start_index == 1:
                                        labels[start_index] = 'B-' + key
                                        labels[end_index] = 'E-' + key
                                    else:
                                        labels[start_index] = 'B-' + key
                                        labels[start_index + 1:end_index] = ['I-' + key] * (len(sub_name) - 2)
                                        labels[end_index] = 'E-' + key
                lines.append({"words": words, "labels": labels})
        return lines


def get_entity_bios(seq, id2label, middle_prefix='I-'):
    """Gets entities from sequence.
    note: BIOS
    Args:
        seq (list): sequence of labels.
    Returns:
        list: list of (chunk_type, chunk_start, chunk_end).
    Example:
        # >>> seq = ['B-PER', 'I-PER', 'O', 'S-LOC']
        # >>> get_entity_bios(seq)
        [['PER', 0,1], ['LOC', 3, 3]]
    """
    chunks = []
    chunk = [-1, -1, -1]
    for indx, tag in enumerate(seq):
        if not isinstance(tag, str):
            tag = id2label[tag]
        if tag.startswith("S-"):
            if chunk[2] != -1:
                chunks.append(chunk)
            chunk = [-1, -1, -1]
            chunk[1] = indx
            chunk[2] = indx
            chunk[0] = tag.split('-')[1]
            chunks.append(chunk)
            chunk = (-1, -1, -1)
        if tag.startswith("B-"):
            if chunk[2] != -1:
                chunks.append(chunk)
            chunk = [-1, -1, -1]
            chunk[1] = indx
            chunk[0] = tag.split('-')[1]
        elif tag.startswith(middle_prefix) and chunk[1] != -1:
            _type = tag.split('-')[1]
            if _type == chunk[0]:
                chunk[2] = indx
            if indx == len(seq) - 1:
                chunks.append(chunk)
        else:
            if chunk[2] != -1:
                chunks.append(chunk)
            chunk = [-1, -1, -1]
    return chunks


def get_entity_bio(seq, id2label, middle_prefix='I-'):
    """Gets entities from sequence.
    note: BIO
    Args:
        seq (list): sequence of labels.
    Returns:
        list: list of (chunk_type, chunk_start, chunk_end).
    Example:
        seq = ['B-PER', 'I-PER', 'O', 'B-LOC']
        get_entity_bio(seq)
        #output
        [['PER', 0,1], ['LOC', 3, 3]]
    """
    chunks = []
    chunk = [-1, -1, -1]
    for indx, tag in enumerate(seq):
        if not isinstance(tag, str):
            tag = id2label[tag]
        if tag.startswith("B-"):
            if chunk[2] != -1:
                chunks.append(chunk)
            chunk = [-1, -1, -1]
            chunk[1] = indx
            chunk[0] = tag.split('-')[1]
            chunk[2] = indx
            if indx == len(seq) - 1:
                chunks.append(chunk)
        elif tag.startswith(middle_prefix) and chunk[1] != -1:
            _type = tag.split('-')[1]
            if _type == chunk[0]:
                chunk[2] = indx

            if indx == len(seq) - 1:
                chunks.append(chunk)
        else:
            if chunk[2] != -1:
                chunks.append(chunk)
            chunk = [-1, -1, -1]
    return chunks


def get_entity_bioes(seq, id2label, middle_prefix='I-'):
    """Gets entities from sequence.
    note: BIOS
    Args:
        seq (list): sequence of labels.
    Returns:
        list: list of (chunk_type, chunk_start, chunk_end).
    Example:
        # >>> seq = ['B-PER', 'I-PER', 'O', 'S-LOC']
        # >>> get_entity_bios(seq)
        [['PER', 0,1], ['LOC', 3, 3]]
    """
    chunks = []
    chunk = [-1, -1, -1]
    for indx, tag in enumerate(seq):
        if not isinstance(tag, str):
            tag = id2label[tag]
        if tag.startswith("S-"):
            if chunk[2] != -1:
                chunks.append(chunk)
            chunk = [-1, -1, -1]
            chunk[1] = indx
            chunk[2] = indx
            chunk[0] = tag.split('-')[1]
            chunks.append(chunk)
            chunk = (-1, -1, -1)
        if tag.startswith("B-"):
            if chunk[2] != -1:
                chunks.append(chunk)
            chunk = [-1, -1, -1]
            chunk[1] = indx
            chunk[0] = tag.split('-')[1]
        elif (tag.startswith(middle_prefix) or tag.startswith("E-")) and chunk[1] != -1:
            _type = tag.split('-')[1]
            if _type == chunk[0]:
                chunk[2] = indx
            if indx == len(seq) - 1:
                chunks.append(chunk)
        else:
            if chunk[2] != -1:
                chunks.append(chunk)
            chunk = [-1, -1, -1]
    return chunks


def get_entities(seq, id2label, markup='bio', middle_prefix='I-'):
    '''
    :param seq:
    :param id2label:
    :param markup:
    :return:
    '''
    assert markup in ['bio', 'bios', 'bioes']
    if markup == 'bio':
        return get_entity_bio(seq, id2label, middle_prefix)
    elif markup == 'bios':
        return get_entity_bios(seq, id2label, middle_prefix)
    else:
        return get_entity_bioes(seq, id2label, middle_prefix)


def bert_extract_item(start_logits, end_logits):
    S = []
    start_pred = torch.argmax(start_logits, -1).cpu().numpy()[0][1:-1]
    end_pred = torch.argmax(end_logits, -1).cpu().numpy()[0][1:-1]
    for i, s_l in enumerate(start_pred):
        if s_l == 0:
            continue
        for j, e_l in enumerate(end_pred[i:]):
            if s_l == e_l:
                S.append((s_l, i, i + j))
                break
    return S