MiniLM-L12-H384-uncased 用于抽取式问答

概述

语言模型： microsoft/MiniLM-L12-H384-uncased
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： 参见使用 Haystack 构建的抽取式问答管道示例
基础设施： 1x Tesla v100

超参数

seed=42
batch_size = 12
n_epochs = 4
base_LM_model = "microsoft/MiniLM-L12-H384-uncased"
max_seq_len = 384
learning_rate = 4e-5
lr_schedule = LinearWarmup
warmup_proportion = 0.2
doc_stride=128
max_query_length=64
grad_acc_steps=4

使用方法

在 Haystack 中

Haystack 是一个 AI 编排框架，用于构建可定制、生产级的 LLM 应用程序。您可以在 Haystack 中使用此模型对文档进行抽取式问答。要使用 Haystack 加载并运行模型：

# After running pip install haystack-ai "transformers[torch,sentencepiece]"

from haystack import Document
from haystack.components.readers import ExtractiveReader

docs = [
    Document(content="Python is a popular programming language"),
    Document(content="python ist eine beliebte Programmiersprache"),
]

reader = ExtractiveReader(model="deepset/minilm-uncased-squad2")
reader.warm_up()

question = "What is a popular programming language?"
result = reader.run(query=question, documents=docs)
# {'answers': [ExtractedAnswer(query='What is a popular programming language?', score=0.5740374326705933, data='python', document=Document(id=..., content: '...'), context=None, document_offset=ExtractedAnswer.Span(start=0, end=6),...)]}

如需查看包含可扩展至多个文档的抽取式问答流水线完整示例，请查看相应的 Haystack 教程。

在 Transformers 中

from transformers import AutoModelForQuestionAnswering, AutoTokenizer, pipeline

model_name = "deepset/minilm-uncased-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)

性能

使用官方评估脚本在 SQuAD 2.0 开发集上进行评估。

"exact": 76.13071675229513,
"f1": 79.49786500219953,
"total": 11873,
"HasAns_exact": 78.35695006747639,
"HasAns_f1": 85.10090269418276,
"HasAns_total": 5928,
"NoAns_exact": 73.91084945332211,
"NoAns_f1": 73.91084945332211,
"NoAns_total": 5945

作者

Vaishali Pal: vaishali.pal@deepset.ai
Branden Chan: branden.chan@deepset.ai
Timo Möller: timo.moeller@deepset.ai
Malte Pietsch: malte.pietsch@deepset.ai
Tanay Soni: tanay.soni@deepset.ai

关于我们

deepset 是生产级开源 AI 框架 Haystack 的开发公司。

我们的其他部分成果：

联系我们并加入 Haystack 社区

想了解更多关于 Haystack 的信息，请访问我们的 GitHub 代码库和文档。

我们还有一个向所有人开放的 Discord 社区！

顺便说一下：我们正在招聘！

seed=42 batch_size = 12 n_epochs = 4 base_LM_model = "microsoft/MiniLM-L12-H384-uncased" max_seq_len = 384 learning_rate = 4e-5 lr_schedule = LinearWarmup warmup_proportion = 0.2 doc_stride=128 max_query_length=64 grad_acc_steps=4

使用方法

在 Haystack 中

# After running pip install haystack-ai "transformers[torch,sentencepiece]"

from haystack import Document
from haystack.components.readers import ExtractiveReader

docs = [
    Document(content="Python is a popular programming language"),
    Document(content="python ist eine beliebte Programmiersprache"),
]

reader = ExtractiveReader(model="deepset/minilm-uncased-squad2")
reader.warm_up()

question = "What is a popular programming language?"
result = reader.run(query=question, documents=docs)
# {'answers': [ExtractedAnswer(query='What is a popular programming language?', score=0.5740374326705933, data='python', document=Document(id=..., content: '...'), context=None, document_offset=ExtractedAnswer.Span(start=0, end=6),...)]}

如需查看包含可扩展至多个文档的抽取式问答流水线完整示例，请查看相应的 Haystack 教程。

在 Transformers 中

from transformers import AutoModelForQuestionAnswering, AutoTokenizer, pipeline

model_name = "deepset/minilm-uncased-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)

性能

使用官方评估脚本在 SQuAD 2.0 开发集上进行评估。

"exact": 76.13071675229513,
"f1": 79.49786500219953,
"total": 11873,
"HasAns_exact": 78.35695006747639,
"HasAns_f1": 85.10090269418276,
"HasAns_total": 5928,
"NoAns_exact": 73.91084945332211,
"NoAns_f1": 73.91084945332211,
"NoAns_total": 5945