Total runs:
554.1K
Run Growth:
37.7K
Growth Rate:
6.81%
Sponsored by
YouArt
deepset / gelectra-base-germanquad-distilled
Model's Last Updated: September 26 2024
question-answering
Introduction of gelectra-base-germanquad-distilled
Model Details of gelectra-base-germanquad-distilled
Overview
Language model:
gelectra-base-germanquad-distilled
Language:
German
Training data:
GermanQuAD train set (~ 12MB)
Eval data:
GermanQuAD test set (~ 5MB)
Infrastructure
: 1x V100 GPU
Published
: Apr 21st, 2021
Details
- We trained a German question answering model with a gelectra-base model as its basis.
- The dataset is GermanQuAD, a new, German language dataset, which we hand-annotated and published online .
- The training dataset is one-way annotated and contains 11518 questions and 11518 answers, while the test dataset is three-way annotated so that there are 2204 questions and with 2204·3−76 = 6536answers, because we removed 76 wrong answers.
- In addition to the annotations in GermanQuAD, haystack's distillation feature was used for training. deepset/gelectra-large-germanquad was used as the teacher model.
See https://deepset.ai/germanquad for more details and dataset download in SQuAD format.
Hyperparameters
batch_size = 24
n_epochs = 6
max_seq_len = 384
learning_rate = 3e-5
lr_schedule = LinearWarmup
embeds_dropout_prob = 0.1
temperature = 2
distillation_loss_weight = 0.75
Performance
We evaluated the extractive question answering performance on our GermanQuAD test set. Model types and training data are included in the model name. For finetuning XLM-Roberta, we use the English SQuAD v2.0 dataset. The GELECTRA models are warm started on the German translation of SQuAD v1.1 and finetuned on \\germanquad. The human baseline was computed for the 3-way test set by taking one answer as prediction and the other two as ground truth.
"exact": 62.4773139745916
"f1": 80.9488017070188
Authors
-
Timo Möller:
timo.moeller [at] deepset.ai -
Julian Risch:
julian.risch [at] deepset.ai -
Malte Pietsch:
malte.pietsch [at] deepset.ai -
Michel Bartels:
michel.bartels [at] deepset.ai
About us
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 | Slack | GitHub Discussions | Website
By the way: we're hiring!
Runs of deepset gelectra-base-germanquad-distilled on huggingface.co
87
Total runs
0
24-hour runs
-1
3-day runs
1
7-day runs
26
30-day runs
More Information About gelectra-base-germanquad-distilled huggingface.co Model
Url of gelectra-base-germanquad-distilled
gelectra-base-germanquad-distilled huggingface.co Url
Provider of gelectra-base-germanquad-distilled huggingface.co
Other API from deepset
Updated:September 24 2024
Total runs:
251.8K
Run Growth:
244.6K
Growth Rate:
97.16%
Total runs:
204.6K
Run Growth:
8.2K
Growth Rate:
4.02%
Total runs:
191.1K
Run Growth:
-3.1K
Growth Rate:
-1.62%
Total runs:
69.8K
Run Growth:
20.3K
Growth Rate:
29.19%
Total runs:
47.4K
Run Growth:
-33.3K
Growth Rate:
-70.10%
Total runs:
32.8K
Run Growth:
6.1K
Growth Rate:
16.90%
Total runs:
22.8K
Run Growth:
-56.9K
Growth Rate:
-249.52%
Total runs:
20.0K
Run Growth:
5.0K
Growth Rate:
24.90%
Total runs:
16.4K
Run Growth:
9.6K
Growth Rate:
59.46%
Total runs:
11.9K
Run Growth:
-1.5K
Growth Rate:
-12.45%
Total runs:
9.5K
Run Growth:
-1.7K
Growth Rate:
-18.19%
Total runs:
9.4K
Run Growth:
137
Growth Rate:
1.46%
Total runs:
7.5K
Run Growth:
-2.0K
Growth Rate:
-27.37%
Total runs:
3.5K
Run Growth:
1.9K
Growth Rate:
54.28%
Total runs:
3.3K
Run Growth:
536
Growth Rate:
16.39%
Total runs:
1.9K
Run Growth:
19
Growth Rate:
1.02%
Total runs:
1.8K
Run Growth:
-3.9K
Growth Rate:
-212.72%
Total runs:
1.7K
Run Growth:
-2.3K
Growth Rate:
-135.16%
Total runs:
1.7K
Run Growth:
-4.0K
Growth Rate:
-375.88%
Total runs:
1.0K
Run Growth:
-3.0K
Growth Rate:
-453.52%
Total runs:
1.0K
Run Growth:
-3.0K
Growth Rate:
-459.48%
Total runs:
832
Run Growth:
-335
Growth Rate:
-45.21%
Total runs:
825
Run Growth:
45
Growth Rate:
5.31%
Total runs:
781
Run Growth:
-6.5K
Growth Rate:
-874.29%
Total runs:
713
Run Growth:
-1.6K
Growth Rate:
-226.42%
Total runs:
568
Run Growth:
-3.8K
Growth Rate:
-685.95%
Total runs:
251
Run Growth:
-17
Growth Rate:
-6.14%
Total runs:
210
Run Growth:
-12
Growth Rate:
-5.71%
Total runs:
155
Run Growth:
-49
Growth Rate:
-31.82%
Total runs:
118
Run Growth:
78
Growth Rate:
66.10%
Total runs:
114
Run Growth:
-93
Growth Rate:
-81.58%
Total runs:
97
Run Growth:
10
Growth Rate:
10.53%
Total runs:
74
Run Growth:
-22
Growth Rate:
-31.43%
Total runs:
62
Run Growth:
28
Growth Rate:
51.85%
Total runs:
52
Run Growth:
28
Growth Rate:
53.85%
Total runs:
41
Run Growth:
22
Growth Rate:
61.11%
Total runs:
40
Run Growth:
18
Growth Rate:
51.43%
Total runs:
37
Run Growth:
31
Growth Rate:
83.78%
Total runs:
32
Run Growth:
14
Growth Rate:
56.00%
Total runs:
14
Run Growth:
8
Growth Rate:
57.14%
Total runs:
11
Run Growth:
4
Growth Rate:
36.36%