Total runs:
1.9M
Run Growth:
103.9K
Growth Rate:
5.43%
deepset / xlm-roberta-base-squad2
Model's Last Updated: September 26 2024
question-answering
Introduction of xlm-roberta-base-squad2
Model Details of xlm-roberta-base-squad2
Multilingual XLM-RoBERTa base for QA on various languages
Overview
Language model:
xlm-roberta-base
Language:
Multilingual
Downstream-task:
Extractive QA
Training data:
SQuAD 2.0
Eval data:
SQuAD 2.0 dev set - German MLQA - German XQuAD
Code:
See
example
in
FARM
Infrastructure
: 4x Tesla v100
Hyperparameters
batch_size = 22*4
n_epochs = 2
max_seq_len=256,
doc_stride=128,
learning_rate=2e-5,
Corresponding experiment logs in mlflow: link
Performance
Evaluated on the SQuAD 2.0 dev set with the official eval script .
"exact": 73.91560683904657
"f1": 77.14103746689592
Evaluated on German MLQA: test-context-de-question-de.json "exact": 33.67279167589108 "f1": 44.34437105434842 "total": 4517
Evaluated on German XQuAD: xquad.de.json "exact": 48.739495798319325 "f1": 62.552615701071495 "total": 1190
Usage
In Transformers
from transformers.pipelines import pipeline
from transformers.modeling_auto import AutoModelForQuestionAnswering
from transformers.tokenization_auto import AutoTokenizer
model_name = "deepset/xlm-roberta-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)
In FARM
from farm.modeling.adaptive_model import AdaptiveModel
from farm.modeling.tokenization import Tokenizer
from farm.infer import Inferencer
model_name = "deepset/xlm-roberta-base-squad2"
# a) Get predictions
nlp = Inferencer.load(model_name, task_type="question_answering")
QA_input = [{"questions": ["Why is model conversion important?"],
"text": "The option to convert models between FARM and transformers gives freedom to the user and let people easily switch between frameworks."}]
res = nlp.inference_from_dicts(dicts=QA_input, rest_api_schema=True)
# b) Load model & tokenizer
model = AdaptiveModel.convert_from_transformers(model_name, device="cpu", task_type="question_answering")
tokenizer = Tokenizer.load(model_name)
In haystack
For doing QA at scale (i.e. many docs instead of single paragraph), you can load the model also in haystack :
reader = FARMReader(model_name_or_path="deepset/xlm-roberta-base-squad2")
# or
reader = TransformersReader(model="deepset/roberta-base-squad2",tokenizer="deepset/xlm-roberta-base-squad2")
Authors
Branden Chan:
branden.chan [at] deepset.ai
Timo Möller:
timo.moeller [at] deepset.ai
Malte Pietsch:
malte.pietsch [at] deepset.ai
Tanay Soni:
tanay.soni [at] deepset.ai
About us
We bring NLP to the industry via open source!
Our focus: Industry specific language models & large scale QA systems.
Some of our work:
- German BERT (aka "bert-base-german-cased")
- GermanQuAD and GermanDPR datasets and models (aka "gelectra-base-germanquad", "gbert-base-germandpr")
- FARM
- Haystack
Get in touch: Twitter | LinkedIn | Discord | GitHub Discussions | Website
By the way: we're hiring!
Runs of deepset xlm-roberta-base-squad2 on huggingface.co
3.6K
Total runs
0
24-hour runs
67
3-day runs
1.1K
7-day runs
2.0K
30-day runs
More Information About xlm-roberta-base-squad2 huggingface.co Model
Url of xlm-roberta-base-squad2
xlm-roberta-base-squad2 huggingface.co Url
Provider of xlm-roberta-base-squad2 huggingface.co
Other API from deepset
Updated: September 24 2024
Total runs:
222.4K
Run Growth:
10.9K
Growth Rate:
4.90%
Total runs:
109.9K
Run Growth:
-1.2M
Growth Rate:
-795.98%
Total runs:
98.6K
Run Growth:
11.9K
Growth Rate:
12.39%
Total runs:
47.9K
Run Growth:
33.6K
Growth Rate:
71.94%
Total runs:
34.2K
Run Growth:
221
Growth Rate:
0.72%
Total runs:
32.6K
Run Growth:
18.2K
Growth Rate:
58.71%
Total runs:
28.1K
Run Growth:
-74.9K
Growth Rate:
-273.64%
Total runs:
27.4K
Run Growth:
4.8K
Growth Rate:
17.60%
Total runs:
9.6K
Run Growth:
3.8K
Growth Rate:
40.14%
Total runs:
8.5K
Run Growth:
-2.8K
Growth Rate:
-35.77%
Total runs:
5.3K
Run Growth:
1.9K
Growth Rate:
35.35%
Total runs:
5.2K
Run Growth:
-525
Growth Rate:
-10.22%
Total runs:
3.2K
Run Growth:
-2.5K
Growth Rate:
-78.03%
Total runs:
1.9K
Run Growth:
644
Growth Rate:
31.52%
Total runs:
881
Run Growth:
119
Growth Rate:
13.57%
Total runs:
869
Run Growth:
-109
Growth Rate:
-12.98%
Total runs:
602
Run Growth:
-1.5K
Growth Rate:
-248.64%
Total runs:
429
Run Growth:
396
Growth Rate:
91.88%
Total runs:
418
Run Growth:
-395
Growth Rate:
-91.86%
Total runs:
379
Run Growth:
-1.4K
Growth Rate:
-368.94%
Total runs:
316
Run Growth:
-113
Growth Rate:
-36.57%
Total runs:
229
Run Growth:
177
Growth Rate:
72.84%
Total runs:
196
Run Growth:
58
Growth Rate:
47.93%
Total runs:
186
Run Growth:
28
Growth Rate:
14.74%
Total runs:
179
Run Growth:
108
Growth Rate:
62.43%
Total runs:
146
Run Growth:
-204
Growth Rate:
-139.73%
Total runs:
133
Run Growth:
117
Growth Rate:
86.67%
Total runs:
131
Run Growth:
98
Growth Rate:
73.68%
Total runs:
120
Run Growth:
-28
Growth Rate:
-18.67%
Total runs:
112
Run Growth:
-2
Growth Rate:
-1.79%
Total runs:
63
Run Growth:
36
Growth Rate:
55.38%
Total runs:
56
Run Growth:
-102
Growth Rate:
-175.86%
Total runs:
54
Run Growth:
23
Growth Rate:
41.07%
Total runs:
48
Run Growth:
32
Growth Rate:
66.67%
Total runs:
30
Run Growth:
14
Growth Rate:
43.75%
Total runs:
21
Run Growth:
-100
Growth Rate:
-476.19%
Total runs:
21
Run Growth:
-91
Growth Rate:
-433.33%
Total runs:
20
Run Growth:
-40
Growth Rate:
-190.48%
Total runs:
19
Run Growth:
6
Growth Rate:
31.58%
Total runs:
19
Run Growth:
-55
Growth Rate:
-289.47%
Total runs:
17
Run Growth:
-12
Growth Rate:
-66.67%
Total runs:
5
Run Growth:
-188
Growth Rate:
-3760.00%