ArSUM-ai
This model is a fine-tuned version of abdalrahmanshahrour/arabartsummarization on the xlsum dataset. It achieves the following results on the evaluation set:
- Loss: 2.5951
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
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.3947 | 1.0 | 9380 | 2.4121 |
| 2.182 | 2.0 | 18760 | 2.4303 |
| 1.9303 | 3.0 | 28140 | 2.4727 |
| 1.7455 | 4.0 | 37520 | 2.5470 |
| 1.5804 | 5.0 | 46900 | 2.5951 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.0+cu116
- Datasets 2.8.0
- Tokenizers 0.13.2
- Downloads last month
- 62
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.