-
Notifications
You must be signed in to change notification settings - Fork 394
/
Copy pathmodel.py
350 lines (298 loc) · 16.3 KB
/
model.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
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
# !/usr/bin/env python3
"""
==== No Bugs in code, just some Random Unexpected FEATURES ====
┌─────────────────────────────────────────────────────────────┐
│┌───┬───┬───┬───┬───┬───┬───┬───┬───┬───┬───┬───┬───┬───┬───┐│
││Esc│!1 │@2 │#3 │$4 │%5 │^6 │&7 │*8 │(9 │)0 │_- │+= │|\ │`~ ││
│├───┴─┬─┴─┬─┴─┬─┴─┬─┴─┬─┴─┬─┴─┬─┴─┬─┴─┬─┴─┬─┴─┬─┴─┬─┴─┬─┴───┤│
││ Tab │ Q │ W │ E │ R │ T │ Y │ U │ I │ O │ P │{[ │}] │ BS ││
│├─────┴┬──┴┬──┴┬──┴┬──┴┬──┴┬──┴┬──┴┬──┴┬──┴┬──┴┬──┴┬──┴─────┤│
││ Ctrl │ A │ S │ D │ F │ G │ H │ J │ K │ L │: ;│" '│ Enter ││
│├──────┴─┬─┴─┬─┴─┬─┴─┬─┴─┬─┴─┬─┴─┬─┴─┬─┴─┬─┴─┬─┴─┬─┴────┬───┤│
││ Shift │ Z │ X │ C │ V │ B │ N │ M │< ,│> .│? /│Shift │Fn ││
│└─────┬──┴┬──┴──┬┴───┴───┴───┴───┴───┴──┬┴───┴┬──┴┬─────┴───┘│
│ │Fn │ Alt │ Space │ Alt │Win│ HHKB │
│ └───┴─────┴───────────────────────┴─────┴───┘ │
└─────────────────────────────────────────────────────────────┘
各种匹配模型的实现。
Author: pankeyu
Date: 2022/10/26
"""
from typing import List
import torch
import torch.nn as nn
import numpy as np
class PointwiseMatching(nn.Module):
"""
PointWise 匹配实现, 高准度但匹配慢, 不适合用作大规模匹配且低时延的场景。
Code Reference: https://github.com/PaddlePaddle/PaddleNLP/blob/develop/examples/text_matching/ernie_matching/model.py
Args:
nn (_type_): _description_
"""
def __init__(self, encoder, dropout=None):
"""
init func.
Args:
encoder (transformers.AutoModel): backbone, 默认使用 ernie 3.0
dropout (float): dropout 比例
"""
super().__init__()
self.encoder = encoder
hidden_size = 768
self.dropout = nn.Dropout(dropout if dropout is not None else 0.1)
self.classifier = nn.Linear(768, 2)
def forward(self,
input_ids,
token_type_ids,
position_ids=None,
attention_mask=None) -> torch.tensor:
"""
Foward 函数，输入匹配好的pair对，返回二维向量（相似/不相似）。
Args:
input_ids (torch.LongTensor): (batch, seq_len)
token_type_ids (torch.LongTensor): (batch, seq_len)
position_ids (torch.LongTensor): (batch, seq_len)
attention_mask (torch.LongTensor): (batch, seq_len)
Returns:
torch.tensor: (batch, 2)
"""
pooled_embedding = self.encoder(
input_ids=input_ids,
token_type_ids=token_type_ids,
position_ids=position_ids,
attention_mask=attention_mask
)["pooler_output"] # (batch, hidden_size)
pooled_embedding = self.dropout(pooled_embedding) # (batch, hidden_size)
logits = self.classifier(pooled_embedding) # (batch, 2)
return logits
class DSSM(nn.Module):
"""
DSSM(Deep Structured Semantic Model) 模型实现, 采用cos值计算向量相似度, 精度稍低, 但计算速度快。
Paper Reference: https://posenhuang.github.io/papers/cikm2013_DSSM_fullversion.pdf
Args:
nn (_type_): _description_
"""
def __init__(self, encoder, dropout=None):
"""
init func.
Args:
encoder (transformers.AutoModel): backbone, 默认使用 ernie 3.0
dropout (float): dropout.
"""
super().__init__()
self.encoder = encoder
hidden_size = 768
self.dropout = nn.Dropout(dropout if dropout is not None else 0.1)
def forward(
self,
input_ids: torch.tensor,
token_type_ids: torch.tensor,
attention_mask: torch.tensor
) -> torch.tensor:
"""
forward 函数，输入单句子，获得单句子的embedding。
Args:
input_ids (torch.LongTensor): (batch, seq_len)
token_type_ids (torch.LongTensor): (batch, seq_len)
attention_mask (torch.LongTensor): (batch, seq_len)
Returns:
torch.tensor: embedding -> (batch, hidden_size)
"""
embedding = self.encoder(
input_ids=input_ids,
token_type_ids=token_type_ids,
attention_mask=attention_mask
)["pooler_output"] # (batch, hidden_size)
return embedding
def get_similarity(
self,
query_input_ids: torch.tensor,
query_token_type_ids: torch.tensor,
query_attention_mask: torch.tensor,
doc_input_ids: torch.tensor,
doc_token_type_ids: torch.tensor,
doc_attention_mask: torch.tensor
) -> torch.tensor:
"""
输入query和doc的向量，返回query和doc两个向量的余弦相似度。
Args:
query_input_ids (torch.LongTensor): (batch, seq_len)
query_token_type_ids (torch.LongTensor): (batch, seq_len)
query_attention_mask (torch.LongTensor): (batch, seq_len)
doc_input_ids (torch.LongTensor): (batch, seq_len)
doc_token_type_ids (torch.LongTensor): (batch, seq_len)
doc_attention_mask (torch.LongTensor): (batch, seq_len)
Returns:
torch.tensor: (batch, 1)
"""
query_embedding = self.encoder(
input_ids=query_input_ids,
token_type_ids=query_token_type_ids,
attention_mask=query_attention_mask
)["pooler_output"] # (batch, hidden_size)
query_embedding = self.dropout(query_embedding)
doc_embedding = self.encoder(
input_ids=doc_input_ids,
token_type_ids=doc_token_type_ids,
attention_mask=doc_attention_mask
)["pooler_output"] # (batch, hidden_size)
doc_embedding = self.dropout(doc_embedding)
similarity = nn.functional.cosine_similarity(query_embedding, doc_embedding)
return similarity
class SentenceTransformer(nn.Module):
"""
Sentence Transomer实现, 双塔网络, 精度适中, 计算速度快。
Paper Reference: https://arxiv.org/pdf/1908.10084.pdf
Code Reference: https://github.com/PaddlePaddle/PaddleNLP/blob/develop/examples/text_matching/sentence_transformers/model.py
Args:
nn (_type_): _description_
"""
def __init__(self, encoder, dropout=0.1):
"""
init func.
Args:
encoder (transformers.PretrainedModel): backbone, 默认使用 ernie 3.0
dropout (float): dropout.
"""
super().__init__()
self.encoder = encoder
self.dropout = nn.Dropout(dropout if dropout is not None else 0.1)
self.classifier = nn.Linear(768 * 3, 2) # concat(u, v, u - v) -> 2, 相似/不相似
def forward(
self,
query_input_ids: torch.tensor,
query_token_type_ids: torch.tensor,
query_attention_mask: torch.tensor,
doc_embeddings: torch.tensor,
) -> torch.tensor:
"""
forward 函数，输入query句子和doc_embedding向量，将query句子过一遍模型得到
query embedding再和doc_embedding做二分类。
Args:
input_ids (torch.LongTensor): (batch, seq_len)
token_type_ids (torch.LongTensor): (batch, seq_len)
attention_mask (torch.LongTensor): (batch, seq_len)
doc_embedding (torch.LongTensor): 所有需要匹配的doc_embedding -> (batch, doc_embedding_numbers, hidden_size)
Returns:
torch.tensor: embedding_match_logits -> (batch, doc_embedding_numbers, 2)
"""
query_embedding = self.encoder(
input_ids=query_input_ids,
token_type_ids=query_token_type_ids,
attention_mask=query_attention_mask
)["last_hidden_state"] # (batch, seq_len, hidden_size)
query_attention_mask = torch.unsqueeze(query_attention_mask, dim=-1) # (batch, seq_len, 1)
query_embedding = query_embedding * query_attention_mask # (batch, seq_len, hidden_size)
query_sum_embedding = torch.sum(query_embedding, dim=1) # (batch, hidden_size)
query_sum_mask = torch.sum(query_attention_mask, dim=1) # (batch, 1)
query_mean = query_sum_embedding / query_sum_mask # (batch, hidden_size)
query_mean = query_mean.unsqueeze(dim=1).repeat(1, doc_embeddings.size()[1], 1) # (batch, doc_embedding_numbers, hidden_size)
sub = torch.abs(torch.subtract(query_mean, doc_embeddings)) # (batch, doc_embedding_numbers, hidden_size)
concat = torch.cat([query_mean, doc_embeddings, sub], dim=-1) # (batch, doc_embedding_numbers, hidden_size * 3)
logits = self.classifier(concat) # (batch, doc_embedding_numbers, 2)
return logits
def get_embedding(
self,
input_ids: torch.tensor,
token_type_ids: torch.tensor,
attention_mask: torch.tensor,
) -> torch.tensor:
"""
输入句子，返回这个句子的embedding，用于事先计算doc embedding并存储。
Args:
input_ids (torch.LongTensor): (batch, seq_len)
token_type_ids (torch.LongTensor): (batch, seq_len)
attention_mask (torch.LongTensor): (batch, seq_len)
Returns:
torch.tensor: embedding向量 -> (batch, hidden_size)
"""
embedding = self.encoder(
input_ids=input_ids,
token_type_ids=token_type_ids,
attention_mask=attention_mask
)["last_hidden_state"] # (batch, seq_len, hidden_size)
attention_mask = torch.unsqueeze(attention_mask, dim=-1) # (batch, seq_len, 1)
embedding = embedding * attention_mask # (batch, seq_len, hidden_size)
sum_embedding = torch.sum(embedding, dim=1) # (batch, hidden_size)
sum_mask = torch.sum(attention_mask, dim=1) # (batch, 1)
mean = sum_embedding / sum_mask # (batch, hidden_size)
return mean
def get_similarity_label(
self,
query_input_ids: torch.tensor,
query_token_type_ids: torch.tensor,
query_attention_mask: torch.tensor,
doc_input_ids: torch.tensor,
doc_token_type_ids: torch.tensor,
doc_attention_mask: torch.tensor
) -> torch.tensor:
"""
forward 函数，输入query和doc的向量，返回两个向量相似/不相似的二维向量。
Args:
query_input_ids (torch.LongTensor): (batch, seq_len)
query_token_type_ids (torch.LongTensor): (batch, seq_len)
query_attention_mask (torch.LongTensor): (batch, seq_len)
doc_input_ids (torch.LongTensor): (batch, seq_len)
doc_token_type_ids (torch.LongTensor): (batch, seq_len)
doc_attention_mask (torch.LongTensor): (batch, seq_len)
Returns:
torch.tensor: (batch, 2)
"""
query_embedding = self.encoder(
input_ids=query_input_ids,
token_type_ids=query_token_type_ids,
attention_mask=query_attention_mask
)["last_hidden_state"] # (batch, seq_len, hidden_size)
query_embedding = self.dropout(query_embedding) # (batch, seq_len, hidden_size)
query_attention_mask = torch.unsqueeze(query_attention_mask, dim=-1) # (batch, seq_len, 1)
query_embedding = query_embedding * query_attention_mask # (batch, seq_len, hidden_size)
query_sum_embedding = torch.sum(query_embedding, dim=1) # (batch, hidden_size)
query_sum_mask = torch.sum(query_attention_mask, dim=1) # (batch, 1)
query_mean = query_sum_embedding / query_sum_mask # (batch, hidden_size)
doc_embedding = self.encoder(
input_ids=doc_input_ids,
token_type_ids=doc_token_type_ids,
attention_mask=doc_attention_mask
)["last_hidden_state"] # (batch, seq_len, hidden_size)
doc_embedding = self.dropout(doc_embedding) # (batch, seq_len, hidden_size)
doc_attention_mask = torch.unsqueeze(doc_attention_mask, dim=-1) # (batch, seq_len, 1)
doc_embedding = doc_embedding * doc_attention_mask # (batch, seq_len, hidden_size)
doc_sum_embdding = torch.sum(doc_embedding, dim=1) # (batch, hidden_size)
doc_sum_mask = torch.sum(doc_attention_mask, dim=1) # (batch, 1)
doc_mean = doc_sum_embdding / doc_sum_mask # (batch, hidden_size)
sub = torch.abs(torch.subtract(query_mean, doc_mean)) # (batch, hidden_size)
concat = torch.cat([query_mean, doc_mean, sub], dim=-1) # (batch, hidden_size * 3)
logits = self.classifier(concat) # (batch, 2)
return logits
if __name__ == '__main__':
from rich import print
from transformers import AutoTokenizer, AutoModel
from utils import convert_dssm_example
device = 'cpu'
tokenizer = AutoTokenizer.from_pretrained('nghuyong/ernie-3.0-base-zh')
encoder = AutoModel.from_pretrained('nghuyong/ernie-3.0-base-zh')
model = SentenceTransformer(encoder).to(device)
example = {
"text": [
'今天天气好吗 今天天气怎样 1'
]
}
batch = convert_dssm_example(example, tokenizer, 10)
print(batch)
# * 测试sentence bert训练输出logits
# output = model.get_similarity_label(query_input_ids=torch.LongTensor(batch['query_input_ids']),
# query_token_type_ids=torch.LongTensor(batch['query_token_type_ids']),
# query_attention_mask=torch.LongTensor(batch['query_attention_mask']),
# doc_input_ids=torch.LongTensor(batch['doc_input_ids']),
# doc_token_type_ids=torch.LongTensor(batch['doc_token_type_ids']),
# doc_attention_mask=torch.LongTensor(batch['doc_attention_mask']))
# * 测试sentence bert的inference功能
# output = model(query_input_ids=torch.LongTensor(batch['query_input_ids']).to(device),
# query_token_type_ids=torch.LongTensor(batch['query_token_type_ids']).to(device),
# query_attention_mask=torch.LongTensor(batch['query_attention_mask']).to(device),
# doc_embeddings=torch.randn(1, 10, 768).to(device))
# * 测试sentence bert获取sentence embedding功能
output = model.get_embedding(input_ids=torch.LongTensor(batch['query_input_ids']).to(device),
token_type_ids=torch.LongTensor(batch['query_token_type_ids']).to(device),
attention_mask=torch.LongTensor(batch['query_attention_mask']).to(device))
print(output, output.size())