ahmeddbahaa's picture
update model card README.md
3104a1c
|
raw
history blame
2.68 kB
---
tags:
- summarization
- ar
- encoder-decoder
- roberta
- xlmroberta2xlmroberta
- Abstractive Summarization
- generated_from_trainer
datasets:
- wiki_lingua
model-index:
- name: xlmroberta2xlmroberta-finetuned-ar-wikilingua
results: []
---
<!-- 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. -->
# xlmroberta2xlmroberta-finetuned-ar-wikilingua
This model is a fine-tuned version of [](https://huggingface.co/) on the wiki_lingua dataset.
It achieves the following results on the evaluation set:
- Loss: 4.7757
- Rouge-1: 11.2
- Rouge-2: 1.96
- Rouge-l: 10.28
- Gen Len: 19.8
- Bertscore: 66.27
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 250
- num_epochs: 10
- label_smoothing_factor: 0.1
### Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge-1 | Rouge-2 | Rouge-l | Gen Len | Bertscore |
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:-------:|:---------:|
| 8.03 | 1.0 | 312 | 7.3208 | 0.19 | 0.0 | 0.19 | 20.0 | 54.84 |
| 7.2309 | 2.0 | 624 | 7.1107 | 1.17 | 0.03 | 1.16 | 20.0 | 60.0 |
| 7.0752 | 3.0 | 936 | 7.0061 | 2.58 | 0.15 | 2.55 | 20.0 | 63.52 |
| 6.7538 | 4.0 | 1248 | 6.4189 | 5.75 | 0.46 | 5.55 | 19.95 | 62.83 |
| 6.1513 | 5.0 | 1560 | 5.8402 | 8.46 | 1.04 | 8.08 | 19.2 | 64.25 |
| 5.6639 | 6.0 | 1872 | 5.3938 | 8.62 | 1.17 | 8.16 | 19.28 | 64.81 |
| 5.2857 | 7.0 | 2184 | 5.0719 | 9.34 | 1.41 | 8.61 | 19.71 | 65.29 |
| 5.027 | 8.0 | 2496 | 4.9047 | 10.42 | 1.52 | 9.57 | 19.57 | 65.75 |
| 4.8747 | 9.0 | 2808 | 4.8032 | 10.79 | 1.71 | 9.91 | 19.42 | 66.2 |
| 4.7855 | 10.0 | 3120 | 4.7757 | 11.01 | 1.73 | 10.04 | 19.55 | 66.24 |
### Framework versions
- Transformers 4.19.4
- Pytorch 1.11.0+cu113
- Datasets 2.2.2
- Tokenizers 0.12.1