Spaces:
Sleeping
Sleeping
Gla-AI4BioMed-Lab
commited on
Commit
•
f61c828
1
Parent(s):
0312a01
Delete utils/.ipynb_checkpoints
Browse files
utils/.ipynb_checkpoints/drug_tokenizer-checkpoint.py
DELETED
@@ -1,66 +0,0 @@
|
|
1 |
-
import json
|
2 |
-
import re
|
3 |
-
import torch
|
4 |
-
import torch.nn as nn
|
5 |
-
from torch.nn import functional as F
|
6 |
-
|
7 |
-
class DrugTokenizer:
|
8 |
-
def __init__(self, vocab_path="tokenizer/vocab.json", special_tokens_path="tokenizer/special_tokens_map.json"):
|
9 |
-
self.vocab, self.special_tokens = self.load_vocab_and_special_tokens(vocab_path, special_tokens_path)
|
10 |
-
self.cls_token_id = self.vocab[self.special_tokens['cls_token']]
|
11 |
-
self.sep_token_id = self.vocab[self.special_tokens['sep_token']]
|
12 |
-
self.unk_token_id = self.vocab[self.special_tokens['unk_token']]
|
13 |
-
self.pad_token_id = self.vocab[self.special_tokens['pad_token']]
|
14 |
-
self.id_to_token = {v: k for k, v in self.vocab.items()}
|
15 |
-
|
16 |
-
def load_vocab_and_special_tokens(self, vocab_path, special_tokens_path):
|
17 |
-
with open(vocab_path, 'r', encoding='utf-8') as vocab_file:
|
18 |
-
vocab = json.load(vocab_file)
|
19 |
-
with open(special_tokens_path, 'r', encoding='utf-8') as special_tokens_file:
|
20 |
-
special_tokens_raw = json.load(special_tokens_file)
|
21 |
-
|
22 |
-
special_tokens = {key: value['content'] for key, value in special_tokens_raw.items()}
|
23 |
-
return vocab, special_tokens
|
24 |
-
|
25 |
-
def encode(self, sequence):
|
26 |
-
tokens = re.findall(r'\[([^\[\]]+)\]', sequence)
|
27 |
-
input_ids = [self.cls_token_id] + [self.vocab.get(token, self.unk_token_id) for token in tokens] + [self.sep_token_id]
|
28 |
-
attention_mask = [1] * len(input_ids)
|
29 |
-
return {
|
30 |
-
'input_ids': input_ids,
|
31 |
-
'attention_mask': attention_mask
|
32 |
-
}
|
33 |
-
|
34 |
-
def batch_encode_plus(self, sequences, max_length, padding, truncation, add_special_tokens, return_tensors):
|
35 |
-
input_ids_list = []
|
36 |
-
attention_mask_list = []
|
37 |
-
|
38 |
-
for sequence in sequences:
|
39 |
-
encoded = self.encode(sequence)
|
40 |
-
input_ids = encoded['input_ids']
|
41 |
-
attention_mask = encoded['attention_mask']
|
42 |
-
|
43 |
-
if len(input_ids) > max_length:
|
44 |
-
input_ids = input_ids[:max_length]
|
45 |
-
attention_mask = attention_mask[:max_length]
|
46 |
-
elif len(input_ids) < max_length:
|
47 |
-
pad_length = max_length - len(input_ids)
|
48 |
-
input_ids = input_ids + [self.vocab[self.special_tokens['pad_token']]] * pad_length
|
49 |
-
attention_mask = attention_mask + [0] * pad_length
|
50 |
-
|
51 |
-
input_ids_list.append(input_ids)
|
52 |
-
attention_mask_list.append(attention_mask)
|
53 |
-
|
54 |
-
return {
|
55 |
-
'input_ids': torch.tensor(input_ids_list, dtype=torch.long),
|
56 |
-
'attention_mask': torch.tensor(attention_mask_list, dtype=torch.long)
|
57 |
-
}
|
58 |
-
|
59 |
-
def decode(self, input_ids, skip_special_tokens=False):
|
60 |
-
tokens = []
|
61 |
-
for id in input_ids:
|
62 |
-
if skip_special_tokens and id in [self.cls_token_id, self.sep_token_id, self.pad_token_id]:
|
63 |
-
continue
|
64 |
-
tokens.append(self.id_to_token.get(id, self.special_tokens['unk_token']))
|
65 |
-
sequence = ''.join([f'[{token}]' for token in tokens])
|
66 |
-
return sequence
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|