标签:

• 自动自然语言处理
• 问答系统 语言: 未知 小部件示例:
• 文本: "谁喜欢AutoNLP?" 上下文: "大家都喜欢AutoNLP" 数据集:
• teacookies/autonlp-data-more_fine_tune_24465520 二氧化碳当量排放量: 81.7509252560808克

使用AutoNLP训练的模型

• 问题类型: 抽取式问答
• 模型ID: 26265897
• 二氧化碳排放量(克): 81.7509252560808

验证指标

• 损失值: 0.5754176378250122

使用方法

您可以使用cURL访问该模型:

$ curl -X POST -H "Authorization: Bearer 您的API密钥" -H "Content-Type: application/json" -d '{"question": "谁喜欢AutoNLP?", "context": "大家都喜欢AutoNLP"}' https://api-inference.huggingface.co/models/teacookies/autonlp-more_fine_tune_24465520-26265897

或使用Python API:

import torch

from transformers import AutoModelForQuestionAnswering, AutoTokenizer

model = AutoModelForQuestionAnswering.from_pretrained("teacookies/autonlp-more_fine_tune_24465520-26265897", use_auth_token=True)

tokenizer = AutoTokenizer.from_pretrained("teacookies/autonlp-more_fine_tune_24465520-26265897", use_auth_token=True)

from transformers import BertTokenizer, BertForQuestionAnswering

question, text = "谁喜欢AutoNLP?", "大家都喜欢AutoNLP"

inputs = tokenizer(question, text, return_tensors='pt')

start_positions = torch.tensor([1])

end_positions = torch.tensor([3])

outputs = model(**inputs, start_positions=start_positions, end_positions=end_positions)

loss = outputs.loss

start_scores = outputs.start_logits

end_scores = outputs.end_logits