Usage
from transformers import AutoTokenizer,AutoModelForSequenceClassification,AutoModelForCausalLM
nli_tokenizer = AutoTokenizer.from_pretrained(nli_v2_model_name,trust_remote_code=True)
nli_model = AutoModelForCausalLM.from_pretrained(nli_v2_model_name,device_map="auto", trust_remote_code=True).eval()
query = f"以下提供两个句子，你的工作是选择这两个句子是否明确一致（蕴含）、不一致（矛盾）或者是否无法确定（中立）。你的答案必须是entailment（蕴含）、neutral（中性）或contradiction（矛盾）。\n句子1：{premise}\n句子2：{hypothesis}"
response, history = self.nli_v2_model.chat(self.nli_v2_tokenizer,query,history=None)
- Downloads last month
- 2
Inference API (serverless) does not yet support model repos that contain custom code.