-
Notifications
You must be signed in to change notification settings - Fork 3
/
dataloader.py
326 lines (259 loc) · 13.5 KB
/
dataloader.py
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
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
from util import *
import torch.utils.data as util_data
from torch.utils.data import Dataset
def set_seed(seed):
random.seed(seed)
np.random.seed(10)
torch.manual_seed(seed)
torch.cuda.manual_seed(seed)
torch.cuda.manual_seed_all(seed)
class Data:
def __init__(self, args):
# 随机初始化
set_seed(args.seed)
# 载入数据集的关键信息
max_seq_lengths = {'clinc':30, 'banking':55, 'snips': 35, "HWU64":25}
args.max_seq_length = max_seq_lengths[args.dataset]
# 随机选取已知类和未知类(得到IND和OOD的类别list)
self.data_dir = os.path.join(args.data_dir, args.dataset)
self.all_label_list = self.get_labels(self.data_dir)
print("the numbers of all labels:", len(self.all_label_list))
self.n_known_cls = round(len(self.all_label_list) * args.known_cls_ratio)
self.known_label_list = list(np.random.choice(np.array(self.all_label_list), self.n_known_cls, replace=False))
print("the numbers of IND labels:", len(self.known_label_list), self.n_known_cls)
self.n_unknown_cls = len(self.all_label_list) - len(self.known_label_list)
self.unknown_label_list = list(set(self.all_label_list).difference(set(self.known_label_list)))
print("the numbers of OOD labels:", len(self.unknown_label_list), self.n_unknown_cls)
print(self.unknown_label_list)
for k in range(len(self.unknown_label_list)):
print(self.unknown_label_list[k])
if args.IND_ratio!=1.0:
self.n_known_cls = round(len(self.known_label_list) * args.IND_ratio)
self.known_label_list = list(np.random.choice(np.array(self.known_label_list), self.n_known_cls, replace=False))
print("revised: the numbers of IND labels:", len(self.known_label_list), self.n_known_cls)
self.num_labels = int(len(self.unknown_label_list))*2
# 载入数据集(tsv文件的表格,二维列表形式)
train_sets = self.get_datasets(self.data_dir, 'train')
eval_sets = self.get_datasets(self.data_dir, 'eval')
test_sets = self.get_datasets(self.data_dir, 'test')
# 划分OOD和IND (至此,数据结构都是list,元素为数据集中的一行(也是一个小的list),并且还是字符形式的)
self.train_labeled_examples, self.train_unlabeled_examples = self.divide_datasets_2(train_sets, args)
print('train_num_labeled_samples', len(self.train_labeled_examples))
print('train_num_unlabeled_samples', len(self.train_unlabeled_examples))
self.eval_labeled_examples, self.eval_unlabeled_examples = self.divide_datasets(eval_sets)
print('eval_num_labeled_samples', len(self.eval_labeled_examples))
print('eval_num_unlabeled_samples', len(self.eval_unlabeled_examples))
self.test_labeled_examples, self.test_unlabeled_examples = self.divide_datasets(test_sets)
print('test_num_samples', len(self.test_labeled_examples))
print('test_num_unlabeled_samples', len(self.test_unlabeled_examples))
# (此时仍然还是字符形式的)
self.train_labeled_examples = self.get_samples(self.train_labeled_examples, args, "train")
self.train_unlabeled_examples = self.get_samples(self.train_unlabeled_examples, args, "train")
self.eval_labeled_examples = self.get_samples(self.eval_labeled_examples, args, "eval")
self.eval_unlabeled_examples = self.get_samples(self.eval_unlabeled_examples, args, "eval")
self.test_labeled_examples = self.get_samples(self.test_labeled_examples, args, "test")
self.test_unlabeled_examples = self.get_samples(self.test_unlabeled_examples, args, "test")
# 封装成dataloader格式(此时需要vectorization)
self.train_labeled_dataloader = self.get_loader(self.train_labeled_examples,self.known_label_list,args,"train")
self.train_unlabeled_dataloader = self.augment_loader(self.train_unlabeled_examples, self.unknown_label_list, args,"train")
self.eval_labeled_dataloader = self.get_loader(self.eval_labeled_examples,self.known_label_list, args, "eval")
self.eval_unlabeled_dataloader = self.get_loader(self.eval_unlabeled_examples,self.unknown_label_list, args, "eval")
self.test_labeled_dataloader = self.get_loader(self.test_labeled_examples,self.known_label_list, args, "test")
self.test_unlabeled_dataloader = self.get_loader(self.test_unlabeled_examples,self.unknown_label_list, args, "test")
def get_labels(self, data_dir):
"""See base class."""
import pandas as pd
test = pd.read_csv(os.path.join(data_dir, "train.tsv"), sep="\t")
labels = np.unique(np.array(test['label']))
return labels
def get_datasets(self, data_dir, mode = 'train', quotechar=None):
with open(os.path.join(data_dir, mode+".tsv"), "r") as f:
reader = csv.reader(f, delimiter="\t", quotechar=quotechar)
lines = []
i=0
for line in reader:
if (i==0):
i+=1
continue
line[0] = line[0].strip()
lines.append(line)
return lines
def divide_datasets(self, origin_data):
labeled_examples, unlabeled_examples = [], []
for example in origin_data:
if example[-1] in self.known_label_list:
labeled_examples.append(example)
elif example[-1] in self.unknown_label_list:
unlabeled_examples.append(example)
return labeled_examples, unlabeled_examples
def divide_datasets_2(self, origin_data, args):
train_labels = np.array([example[-1] for example in origin_data])
train_labeled_ids = []
for label in self.known_label_list:
num = round(len(train_labels[train_labels == label]) * args.labeled_ratio)
pos = list(np.where(train_labels == label)[0])
train_labeled_ids.extend(random.sample(pos, num))
labeled_examples, unlabeled_examples = [], []
for idx, example in enumerate(origin_data):
if idx in train_labeled_ids:
labeled_examples.append(example)
elif example[-1] in self.unknown_label_list:
unlabeled_examples.append(example)
return labeled_examples, unlabeled_examples
def get_samples(self, labelled_examples, args, mode):
content_list, labels_list = [], []
for example in labelled_examples:
text = example[0]
label = example[-1]
content_list.append(text)
labels_list.append(label)
data = OriginSamples(content_list,labels_list)
return data
def get_embedding(self, labelled_examples, label_list, args, mode="train"):
tokenizer = BertTokenizer.from_pretrained(args.bert_model, do_lower_case=True)
features = convert_examples_to_features(labelled_examples, label_list, args.max_seq_length,
tokenizer)
data=[]
for f in features:
results={
"input_ids":f.input_ids,
"input_mask":f.input_mask,
"segment_ids":f.segment_ids,
"label_id":f.label_id
}
#print("input_ids",f.input_ids)
#print("input_mask",f.input_mask)
#print("segment_ids",f.segment_ids)
#print("label_id",f.label_id)
data.append(results)
#input_ids = torch.tensor([f.input_ids for f in features], dtype=torch.long)
#input_mask = torch.tensor([f.input_mask for f in features], dtype=torch.long)
#segment_ids = torch.tensor([f.segment_ids for f in features], dtype=torch.long)
#label_ids = torch.tensor([f.label_id for f in features], dtype=torch.long)
#print(input_ids)
#exit()
#data = TensorDataset(input_ids, input_mask, segment_ids, label_ids)
return data
def get_loader(self, labelled_examples, label_list, args, mode="train"):
tokenizer = BertTokenizer.from_pretrained(args.bert_model, do_lower_case=True)
features = convert_examples_to_features(labelled_examples, label_list, args.max_seq_length, tokenizer)
input_ids = torch.tensor([f.input_ids for f in features], dtype=torch.long)
input_mask = torch.tensor([f.input_mask for f in features], dtype=torch.long)
segment_ids = torch.tensor([f.segment_ids for f in features], dtype=torch.long)
label_ids = torch.tensor([f.label_id for f in features], dtype=torch.long)
data = TensorDataset(input_ids, input_mask, segment_ids, label_ids)
if mode == 'train':
sampler = RandomSampler(data)
dataloader = DataLoader(data, sampler=sampler, batch_size=args.pre_train_batch_size)
else:
sampler = SequentialSampler(data)
dataloader = DataLoader(data, sampler=sampler, batch_size=args.eval_batch_size)
return dataloader
def augment_loader(self, unlabelled_examples, label_list, args, mode="train"):
tokenizer = BertTokenizer.from_pretrained(args.bert_model, do_lower_case=True)
features = convert_examples_to_features(unlabelled_examples, label_list, args.max_seq_length, tokenizer)
input_ids = torch.tensor([f.input_ids for f in features], dtype=torch.long)
input_mask = torch.tensor([f.input_mask for f in features], dtype=torch.long)
segment_ids = torch.tensor([f.segment_ids for f in features], dtype=torch.long)
label_ids = torch.tensor([f.label_id for f in features], dtype=torch.long)
data = TensorDataset(input_ids, input_mask, segment_ids, label_ids)
if mode == 'train':
sampler = RandomSampler(data)
dataloader = DataLoader(data, sampler=sampler, batch_size=args.train_batch_size)
else:
sampler = SequentialSampler(data)
dataloader = DataLoader(data, sampler=sampler, batch_size=args.eval_batch_size)
return dataloader
class OriginSamples(Dataset):
def __init__(self, train_x, train_y):
assert len(train_y) == len(train_x)
self.train_x = train_x
self.train_y = train_y
class InputFeatures(object):
"""A single set of features of data."""
def __init__(self, input_ids, input_mask, segment_ids, label_id):
self.input_ids = input_ids
self.input_mask = input_mask
self.segment_ids = segment_ids
self.label_id = label_id
def convert_examples_to_features(examples, label_list, max_seq_length, tokenizer):
"""Loads a data file into a list of `InputBatch`s."""
label_map = {}
for i, label in enumerate(label_list):
label_map[label] = i
'''
if len(label_list)==15:
label_map = {'smart_home': 0, 'spending_history': 1, 'tire_pressure': 2, 'lost_luggage': 3, 'cancel': 4, 'reset_settings': 5, 'where_are_you_from': 6, 'book_flight': 7, 'bill_due': 8,
'accept_reservations': 9, 'expiration_date': 10, 'timezone': 11, 'new_card': 12, 'cancel_reservation': 13, 'income': 14}
print(label_map)
'''
'''
if len(label_list) == 23:
label_map = { "top_up_by_bank_transfer_charge": 0,
"card_acceptance": 1,
"Refund_not_showing_up":2,
"card_not_working":3,
"transfer_fee_charged":4,
"verify_top_up":5,
"unable_to_verify_identity":6,
"beneficiary_not_allowed":7,
"card_linking":8,
"supported_cards_and_currencies":9,
"getting_spare_card":10,
"transfer_into_account":11,
"receiving_money":12,
"card_payment_fee_charged":13,
"automatic_top_up":14,
"declined_transfer":15,
"direct_debit_payment_not_recognised":16,
"pending_transfer":17,
"failed_transfer":18,
"card_delivery_estimate":19,
"cancel_transfer":20,
"topping_up_by_card":21,
"exchange_rate":22,
}
print(label_map)
elif len(label_list) == 15:
label_map = {"getting_spare_card": 0,
"card_not_working": 1,
"exchange_rate": 2,
"card_acceptance": 3,
"topping_up_by_card": 4,
"declined_transfer": 5,
"supported_cards_and_currencies": 6,
"direct_debit_payment_not_recognised": 7,
"failed_transfer": 8,
"cancel_transfer": 9,
"card_payment_fee_charged": 10,
"Refund_not_showing_up": 11,
"verify_top_up": 12,
"beneficiary_not_allowed": 13,
"transfer_fee_charged": 14}
print(label_map)
'''
features = []
content_list = examples.train_x
label_list = examples.train_y
for i in range(len(content_list)):
tokens_a = tokenizer.tokenize(content_list[i])
if len(tokens_a) > max_seq_length - 2:
tokens_a = tokens_a[:(max_seq_length - 2)]
tokens = ["[CLS]"] + tokens_a + ["[SEP]"]
segment_ids = [0] * len(tokens)
input_ids = tokenizer.convert_tokens_to_ids(tokens)
input_mask = [1] * len(input_ids)
padding = [0] * (max_seq_length - len(input_ids))
input_ids += padding
input_mask += padding
segment_ids += padding
assert len(input_ids) == max_seq_length
assert len(input_mask) == max_seq_length
assert len(segment_ids) == max_seq_length
label_id = label_map[label_list[i]]
features.append(
InputFeatures(input_ids=input_ids,
input_mask=input_mask,
segment_ids=segment_ids,
label_id=label_id))
return features