|
--- |
|
library_name: transformers |
|
license: apache-2.0 |
|
base_model: moussaKam/AraBART |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: resultsara_bertscore |
|
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. --> |
|
|
|
# resultsara_bertscore |
|
|
|
This model is a fine-tuned version of [moussaKam/AraBART](https://huggingface.co/moussaKam/AraBART) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.5783 |
|
|
|
## 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: 128 |
|
- eval_batch_size: 128 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 500 |
|
- num_epochs: 8 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:------:|:----:|:---------------:| |
|
| 0.7346 | 0.4263 | 500 | 0.5939 | |
|
| 0.7367 | 0.8525 | 1000 | 0.5851 | |
|
| 0.7042 | 1.2788 | 1500 | 0.5825 | |
|
| 0.6854 | 1.7050 | 2000 | 0.5783 | |
|
| 0.6767 | 2.1313 | 2500 | 0.5775 | |
|
| 0.6561 | 2.5575 | 3000 | 0.5758 | |
|
| 0.6562 | 2.9838 | 3500 | 0.5748 | |
|
| 0.6457 | 3.4101 | 4000 | 0.5774 | |
|
| 0.6519 | 3.8363 | 4500 | 0.5755 | |
|
| 0.6396 | 4.2626 | 5000 | 0.5774 | |
|
| 0.626 | 4.6888 | 5500 | 0.5773 | |
|
| 0.6201 | 5.1151 | 6000 | 0.5789 | |
|
| 0.605 | 5.5413 | 6500 | 0.5776 | |
|
| 0.6044 | 5.9676 | 7000 | 0.5770 | |
|
| 0.5899 | 6.3939 | 7500 | 0.5786 | |
|
| 0.5917 | 6.8201 | 8000 | 0.5779 | |
|
| 0.5913 | 7.2464 | 8500 | 0.5783 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.45.0.dev0 |
|
- Pytorch 2.4.0+cu121 |
|
- Datasets 2.21.0 |
|
- Tokenizers 0.19.1 |
|
|
