|
|
|
@@ -1,15 +0,0 @@ |
|
|
|
from modelscope.models import SbertForNLI |
|
|
|
from modelscope.pipelines import pipeline |
|
|
|
from modelscope.preprocessors import NLIPreprocessor |
|
|
|
|
|
|
|
model = SbertForNLI('../nlp_structbert_nli_chinese-base') |
|
|
|
print(model) |
|
|
|
tokenizer = NLIPreprocessor(model.model_dir) |
|
|
|
|
|
|
|
semantic_cls = pipeline('nli', model=model, preprocessor=tokenizer) |
|
|
|
print(type(semantic_cls)) |
|
|
|
|
|
|
|
print( |
|
|
|
semantic_cls( |
|
|
|
input=('我想还有一件事也伤害到了老师的招聘,那就是他们在课堂上失去了很多的权威', |
|
|
|
'教师在课堂上失去权威,导致想要进入这一职业的人减少了。'))) |