README.md
4.9 KB · 110 lines · markdown Raw
1 ---
2 language: en
3 license: cc-by-4.0
4 datasets:
5 - squad_v2
6 model-index:
7 - name: deepset/bert-base-cased-squad2
8 results:
9 - task:
10 type: question-answering
11 name: Question Answering
12 dataset:
13 name: squad_v2
14 type: squad_v2
15 config: squad_v2
16 split: validation
17 metrics:
18 - type: exact_match
19 value: 71.1517
20 name: Exact Match
21 verified: true
22 verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZGZlNmQ1YzIzMWUzNTg4YmI4NWVhYThiMzE2ZGZmNWUzNDM3NWI0ZGJkNzliNGUxNTY2MDA5MWVkYjAwYWZiMCIsInZlcnNpb24iOjF9.iUvVdy5c4hoXkwlThJankQqG9QXzNilvfF1_4P0oL8X-jkY5Q6YSsZx6G6cpgXogqFpn7JlE_lP6_OT0VIamCg
23 - type: f1
24 value: 74.6714
25 name: F1
26 verified: true
27 verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiMWE5OGNjODhmY2Y0NWIyZDIzMmQ2NmRjZGYyYTYzOWMxZDUzYzg4YjBhNTRiNTY4NTc0M2IxNjI5NWI5ZDM0NCIsInZlcnNpb24iOjF9.IqU9rbzUcKmDEoLkwCUZTKSH0ZFhtqgnhOaEDKKnaRMGBJLj98D5V4VirYT6jLh8FlR0FiwvMTMjReBcfTisAQ
28 ---
29
30 This is a BERT base cased model trained on SQuAD v2
31
32 ## Overview
33 **Language model:** bert-base-cased
34 **Language:** English
35 **Downstream-task:** Extractive QA
36 **Training data:** SQuAD 2.0
37 **Eval data:** SQuAD 2.0
38 **Code:** See [an example extractive QA pipeline built with Haystack](https://haystack.deepset.ai/tutorials/34_extractive_qa_pipeline)
39
40 ## Usage
41
42 ### In Haystack
43 Haystack is an AI orchestration framework to build customizable, production-ready LLM applications. You can use this model in Haystack to do extractive question answering on documents.
44 To load and run the model with [Haystack](https://github.com/deepset-ai/haystack/):
45 ```python
46 # After running pip install haystack-ai "transformers[torch,sentencepiece]"
47
48 from haystack import Document
49 from haystack.components.readers import ExtractiveReader
50
51 docs = [
52 Document(content="Python is a popular programming language"),
53 Document(content="python ist eine beliebte Programmiersprache"),
54 ]
55
56 reader = ExtractiveReader(model="deepset/bert-base-cased-squad2")
57 reader.warm_up()
58
59 question = "What is a popular programming language?"
60 result = reader.run(query=question, documents=docs)
61 # {'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),...)]}
62 ```
63 For a complete example with an extractive question answering pipeline that scales over many documents, check out the [corresponding Haystack tutorial](https://haystack.deepset.ai/tutorials/34_extractive_qa_pipeline).
64
65 ### In Transformers
66 ```python
67 from transformers import AutoModelForQuestionAnswering, AutoTokenizer, pipeline
68
69 model_name = "deepset/bert-base-cased-squad2"
70
71 # a) Get predictions
72 nlp = pipeline('question-answering', model=model_name, tokenizer=model_name)
73 QA_input = {
74 'question': 'Why is model conversion important?',
75 'context': 'The option to convert models between FARM and transformers gives freedom to the user and let people easily switch between frameworks.'
76 }
77 res = nlp(QA_input)
78
79 # b) Load model & tokenizer
80 model = AutoModelForQuestionAnswering.from_pretrained(model_name)
81 tokenizer = AutoTokenizer.from_pretrained(model_name)
82 ```
83
84 ## About us
85
86 <div class="grid lg:grid-cols-2 gap-x-4 gap-y-3">
87 <div class="w-full h-40 object-cover mb-2 rounded-lg flex items-center justify-center">
88 <img alt="" src="https://raw.githubusercontent.com/deepset-ai/.github/main/deepset-logo-colored.png" class="w-40"/>
89 </div>
90 <div class="w-full h-40 object-cover mb-2 rounded-lg flex items-center justify-center">
91 <img alt="" src="https://raw.githubusercontent.com/deepset-ai/.github/main/haystack-logo-colored.png" class="w-40"/>
92 </div>
93 </div>
94
95 [deepset](http://deepset.ai/) is the company behind the production-ready open-source AI framework [Haystack](https://haystack.deepset.ai/).
96
97 Some of our other work:
98 - [Distilled roberta-base-squad2 (aka "tinyroberta-squad2")](https://huggingface.co/deepset/tinyroberta-squad2)
99 - [German BERT (aka "bert-base-german-cased")](https://deepset.ai/german-bert)
100 - [GermanQuAD and GermanDPR datasets and models (aka "gelectra-base-germanquad", "gbert-base-germandpr")](https://deepset.ai/germanquad)
101
102 ## Get in touch and join the Haystack community
103
104 <p>For more info on Haystack, visit our <strong><a href="https://github.com/deepset-ai/haystack">GitHub</a></strong> repo and <strong><a href="https://docs.haystack.deepset.ai">Documentation</a></strong>.
105
106 We also have a <strong><a class="h-7" href="https://haystack.deepset.ai/community">Discord community open to everyone!</a></strong></p>
107
108 [Twitter](https://twitter.com/Haystack_AI) | [LinkedIn](https://www.linkedin.com/company/deepset-ai/) | [Discord](https://haystack.deepset.ai/community) | [GitHub Discussions](https://github.com/deepset-ai/haystack/discussions) | [Website](https://haystack.deepset.ai/) | [YouTube](https://www.youtube.com/@deepset_ai)
109
110 By the way: [we're hiring!](http://www.deepset.ai/jobs)