asahi417's picture
model update
a7101d2
---
license: cc-by-4.0
metrics:
- bleu4
- meteor
- rouge-l
- bertscore
- moverscore
language: de
datasets:
- lmqg/qag_dequad
pipeline_tag: text2text-generation
tags:
- questions and answers generation
widget:
- text: "Empfangs- und Sendeantenne sollen in ihrer Polarisation übereinstimmen, andernfalls wird die Signalübertragung stark gedämpft. "
example_title: "Questions & Answers Generation Example 1"
model-index:
- name: lmqg/mbart-large-cc25-dequad-qag
results:
- task:
name: Text2text Generation
type: text2text-generation
dataset:
name: lmqg/qag_dequad
type: default
args: default
metrics:
- name: QAAlignedF1Score-BERTScore (Question & Answer Generation)
type: qa_aligned_f1_score_bertscore_question_answer_generation
value: 69.25
- name: QAAlignedRecall-BERTScore (Question & Answer Generation)
type: qa_aligned_recall_bertscore_question_answer_generation
value: 68.05
- name: QAAlignedPrecision-BERTScore (Question & Answer Generation)
type: qa_aligned_precision_bertscore_question_answer_generation
value: 70.69
- name: QAAlignedF1Score-MoverScore (Question & Answer Generation)
type: qa_aligned_f1_score_moverscore_question_answer_generation
value: 50.71
- name: QAAlignedRecall-MoverScore (Question & Answer Generation)
type: qa_aligned_recall_moverscore_question_answer_generation
value: 49.78
- name: QAAlignedPrecision-MoverScore (Question & Answer Generation)
type: qa_aligned_precision_moverscore_question_answer_generation
value: 51.81
---
# Model Card of `lmqg/mbart-large-cc25-dequad-qag`
This model is fine-tuned version of [facebook/mbart-large-cc25](https://huggingface.co/facebook/mbart-large-cc25) for question & answer pair generation task on the [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) (dataset_name: default) via [`lmqg`](https://github.com/asahi417/lm-question-generation).
### Overview
- **Language model:** [facebook/mbart-large-cc25](https://huggingface.co/facebook/mbart-large-cc25)
- **Language:** de
- **Training data:** [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) (default)
- **Online Demo:** [https://autoqg.net/](https://autoqg.net/)
- **Repository:** [https://github.com/asahi417/lm-question-generation](https://github.com/asahi417/lm-question-generation)
- **Paper:** [https://arxiv.org/abs/2210.03992](https://arxiv.org/abs/2210.03992)
### Usage
- With [`lmqg`](https://github.com/asahi417/lm-question-generation#lmqg-language-model-for-question-generation-)
```python
from lmqg import TransformersQG
# initialize model
model = TransformersQG(language="de", model="lmqg/mbart-large-cc25-dequad-qag")
# model prediction
question_answer_pairs = model.generate_qa("das erste weltweit errichtete Hermann Brehmer 1855 im niederschlesischen ''Görbersdorf'' (heute Sokołowsko, Polen).")
```
- With `transformers`
```python
from transformers import pipeline
pipe = pipeline("text2text-generation", "lmqg/mbart-large-cc25-dequad-qag")
output = pipe("Empfangs- und Sendeantenne sollen in ihrer Polarisation übereinstimmen, andernfalls wird die Signalübertragung stark gedämpft. ")
```
## Evaluation
- ***Metric (Question & Answer Generation)***: [raw metric file](https://huggingface.co/lmqg/mbart-large-cc25-dequad-qag/raw/main/eval/metric.first.answer.paragraph.questions_answers.lmqg_qag_dequad.default.json)
| | Score | Type | Dataset |
|:--------------------------------|--------:|:--------|:-------------------------------------------------------------------|
| QAAlignedF1Score (BERTScore) | 69.25 | default | [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) |
| QAAlignedF1Score (MoverScore) | 50.71 | default | [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) |
| QAAlignedPrecision (BERTScore) | 70.69 | default | [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) |
| QAAlignedPrecision (MoverScore) | 51.81 | default | [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) |
| QAAlignedRecall (BERTScore) | 68.05 | default | [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) |
| QAAlignedRecall (MoverScore) | 49.78 | default | [lmqg/qag_dequad](https://huggingface.co/datasets/lmqg/qag_dequad) |
## Training hyperparameters
The following hyperparameters were used during fine-tuning:
- dataset_path: lmqg/qag_dequad
- dataset_name: default
- input_types: ['paragraph']
- output_types: ['questions_answers']
- prefix_types: None
- model: facebook/mbart-large-cc25
- max_length: 512
- max_length_output: 256
- epoch: 6
- batch: 2
- lr: 0.0001
- fp16: False
- random_seed: 1
- gradient_accumulation_steps: 32
- label_smoothing: 0.15
The full configuration can be found at [fine-tuning config file](https://huggingface.co/lmqg/mbart-large-cc25-dequad-qag/raw/main/trainer_config.json).
## Citation
```
@inproceedings{ushio-etal-2022-generative,
title = "{G}enerative {L}anguage {M}odels for {P}aragraph-{L}evel {Q}uestion {G}eneration",
author = "Ushio, Asahi and
Alva-Manchego, Fernando and
Camacho-Collados, Jose",
booktitle = "Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing",
month = dec,
year = "2022",
address = "Abu Dhabi, U.A.E.",
publisher = "Association for Computational Linguistics",
}
```