Create bert_token_handler.py
Browse files- bert_token_handler.py +15 -0
bert_token_handler.py
ADDED
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from transformers import BertTokenizer
|
2 |
+
|
3 |
+
tokenizer = BertTokenizer.from_pretrained(
|
4 |
+
pretrained_model_name_or_path='bert-base-chinese',
|
5 |
+
cache_dir=None,
|
6 |
+
force_download=False
|
7 |
+
)
|
8 |
+
|
9 |
+
sents = ['选择珠江花园的原因就是方便。',
|
10 |
+
'笔记本的键盘确实爽。',
|
11 |
+
'房间太小。其他的都一般。',
|
12 |
+
'今天才知道这书还有第6卷,真有点郁闷。',
|
13 |
+
'机器背面似平被撕了张什么标签,残胶还在。']
|
14 |
+
|
15 |
+
tokenizer, sents
|