|
--- |
|
base_model: google/pegasus-cnn_dailymail |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- xlsum |
|
model-index: |
|
- name: pegsasus_xlsum |
|
results: [] |
|
language: |
|
- en |
|
metrics: |
|
- rouge |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# pegsasus_xlsum |
|
|
|
This model is a fine-tuned version of [google/pegasus-cnn_dailymail](https://huggingface.co/google/pegasus-cnn_dailymail) on the [**csebuetnlp/xlsum/english**](https://huggingface.co/datasets/csebuetnlp/xlsum/viewer/english/train) dataset. |
|
|
|
## Model description |
|
|
|
The **pegasus_xlsum** is a state-of-the-art model fine-tuned on the **English** subset of the [**csebuetnlp/xlsum**](https://huggingface.co/datasets/csebuetnlp/xlsum) dataset. This dataset is one of the most comprehensive and diverse sets available, originally composed of **1.35 million** professional article-summary pairs sourced from the BBC across 45 languages. Despite its multilingual nature, we intentionally selected the **English** language subset, consisting of approximately **330k** records, as the focus for our fine-tuning process. |
|
|
|
The goal was to adapt the model for the text summarization task, and we're thrilled to report that the fine-tuned **pegasus_xlsum** model exceeded our expectations. It outperformed the established [**csebuetnlp/mT5_multilingual_XLSum**](https://huggingface.co/csebuetnlp/mT5_multilingual_XLSum) model in terms of [**ROUGE**](https://huggingface.co/spaces/evaluate-metric/rouge) scores, demonstrating superior summary generation capabilities. The **pegasus_xlsum** model leverages the powerful PEGASUS architecture, proving its efficiency and effectiveness in handling **English** text summarization tasks. |
|
|
|
|
|
## Intended uses |
|
|
|
The **pegasus_xlsum** is to provide a reliable, high-performance solution for **English** text summarization, making the most of the rich, professional, and diverse source dataset it was trained on. We hope you find this model as useful in your applications as we did in our experiments. |
|
|
|
|
|
|
|
## Benchmark comparisons |
|
|
|
**pegasus_xlsum** |
|
| Metric | Score | |
|
|--------|-------| |
|
| **ROUGE-1** | 39.121 | |
|
| **ROUGE-2** | 17.467 | |
|
| **ROUGE-L** | 30.894 | |
|
| **ROUGE-LSUM** | 30.892 | |
|
|
|
**csebuetnlp/mT5_multilingual_XLSum** |
|
| Metric | Score | |
|
|--------|-------| |
|
| **ROUGE-1** | 37.601 | |
|
| **ROUGE-2** | 15.153 | |
|
| **ROUGE-L** | 29.881 | |
|
| **ROUGE-LSUM** | -- | |
|
|
|
## Training and evaluation data |
|
|
|
## Training procedure |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 2e-05 |
|
- train_batch_size: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 4 |
|
|
|
### Framework versions |
|
|
|
- Transformers 4.31.0 |
|
- Pytorch 2.0.1+cu118 |
|
- Datasets 2.14.0 |
|
- Tokenizers 0.13.3 |
|
|
|
## Authors |
|
|
|
- Naimur Rahman ([@NaimurRahman](https://github.com/naimur900)) |
|
- Sadiul Arefin Rafi ([@SadiulArefinRafi](https://github.com/SadiulArefinRafi)) |
|
- Ha-mim Ahmad ([@HaMimAhmad](https://github.com/hamimz)) |
|
- Kazi Nazibul Islam ([@KaziNazibulIslam](https://github.com/nazibislam)) |
|
|
