deberta-v3-base for QA

这是 deberta-v3-base 模型，使用 SQuAD2.0 数据集进行微调。它是使用问答对进行训练的，包括不可回答的问题，用于问答任务。

概述

语言模型：deberta-v3-base 语言：英语下游任务：抽取式问答训练数据：SQuAD 2.0 评估数据：SQuAD 2.0 代码：请参阅 an example QA pipeline on Haystack 基础设施：1x NVIDIA A10G

超参数

batch_size = 12
n_epochs = 4
base_LM_model = "deberta-v3-base"
max_seq_len = 512
learning_rate = 2e-5
lr_schedule = LinearWarmup
warmup_proportion = 0.2
doc_stride = 128
max_query_length = 64

用法

在Haystack中

Haystack是由deepset开发的NLP框架。您可以在Haystack流水线中使用此模型进行大规模的问答（多个文档）。要在 Haystack 中加载模型：

reader = FARMReader(model_name_or_path="deepset/deberta-v3-base-squad2")
# or 
reader = TransformersReader(model_name_or_path="deepset/deberta-v3-base-squad2",tokenizer="deepset/deberta-v3-base-squad2")

在Transformers中

from transformers import AutoModelForQuestionAnswering, AutoTokenizer, pipeline
model_name = "deepset/deberta-v3-base-squad2"
# a) Get predictions
nlp = pipeline('question-answering', model=model_name, tokenizer=model_name)
QA_input = {
    'question': 'Why is model conversion important?',
    'context': 'The option to convert models between FARM and transformers gives freedom to the user and let people easily switch between frameworks.'
}
res = nlp(QA_input)
# b) Load model & tokenizer
model = AutoModelForQuestionAnswering.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)