|
--- |
|
base_model: elnasharomar2/AraBert_arabic_keyword_extraction |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: AraBert_arabic_keyword_extraction |
|
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. --> |
|
|
|
# AraBert_arabic_keyword_extraction |
|
|
|
This model is a fine-tuned version of [aubmindlab/bert-base-arabertv2](https://huggingface.co/aubmindlab/bert-base-arabertv2)n) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.4622 |
|
- Precision: 0.5583 |
|
- Recall: 0.6294 |
|
- F1: 0.5917 |
|
- Accuracy: 0.9297 |
|
|
|
## 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: 2e-05 |
|
- train_batch_size: 16 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 15 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
|
|:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| 0.2524 | 1.0 | 750 | 0.1969 | 0.3638 | 0.4022 | 0.3820 | 0.9161 | |
|
| 0.1705 | 2.0 | 1500 | 0.1793 | 0.4424 | 0.4749 | 0.4581 | 0.9240 | |
|
| 0.1386 | 3.0 | 2250 | 0.1834 | 0.4547 | 0.5438 | 0.4953 | 0.9240 | |
|
| 0.1091 | 4.0 | 3000 | 0.1987 | 0.4805 | 0.5650 | 0.5193 | 0.9243 | |
|
| 0.0892 | 5.0 | 3750 | 0.2164 | 0.4951 | 0.5661 | 0.5282 | 0.9259 | |
|
| 0.0737 | 6.0 | 4500 | 0.2130 | 0.5101 | 0.5635 | 0.5355 | 0.9282 | |
|
| 0.0579 | 7.0 | 5250 | 0.2301 | 0.4890 | 0.5810 | 0.5311 | 0.9266 | |
|
| 0.0481 | 8.0 | 6000 | 0.2479 | 0.5025 | 0.6041 | 0.5486 | 0.9269 | |
|
| 0.0411 | 9.0 | 6750 | 0.2496 | 0.5353 | 0.5739 | 0.5539 | 0.9298 | |
|
| 0.0348 | 10.0 | 7500 | 0.2719 | 0.5150 | 0.6063 | 0.5570 | 0.9286 | |
|
| 0.0304 | 11.0 | 8250 | 0.2881 | 0.5252 | 0.6015 | 0.5608 | 0.9283 | |
|
| 0.0258 | 12.0 | 9000 | 0.3088 | 0.5129 | 0.6093 | 0.5569 | 0.9266 | |
|
| 0.0231 | 13.0 | 9750 | 0.3110 | 0.5230 | 0.5922 | 0.5555 | 0.9284 | |
|
| 0.0199 | 14.0 | 10500 | 0.3196 | 0.5243 | 0.6030 | 0.5609 | 0.9282 | |
|
| 0.0188 | 15.0 | 11250 | 0.3194 | 0.5169 | 0.6041 | 0.5571 | 0.9279 | |
|
| 0.0146 | 16.0 | 750 | 0.3524 | 0.5119 | 0.5993 | 0.5522 | 0.9237 | |
|
| 0.011 | 17.0 | 1500 | 0.3849 | 0.4895 | 0.6410 | 0.5551 | 0.9214 | |
|
| 0.0087 | 18.0 | 2250 | 0.3469 | 0.5353 | 0.6153 | 0.5725 | 0.9311 | |
|
| 0.0113 | 19.0 | 3000 | 0.3471 | 0.5150 | 0.6212 | 0.5631 | 0.9268 | |
|
| 0.0088 | 20.0 | 3750 | 0.3677 | 0.5493 | 0.5929 | 0.5703 | 0.9302 | |
|
| 0.0068 | 21.0 | 4500 | 0.3867 | 0.5313 | 0.6071 | 0.5667 | 0.9270 | |
|
| 0.0056 | 22.0 | 5250 | 0.3843 | 0.5435 | 0.6186 | 0.5786 | 0.9293 | |
|
| 0.0049 | 23.0 | 6000 | 0.4145 | 0.5491 | 0.6272 | 0.5855 | 0.9295 | |
|
| 0.0043 | 24.0 | 6750 | 0.4290 | 0.5396 | 0.6339 | 0.5830 | 0.9280 | |
|
| 0.0035 | 25.0 | 7500 | 0.4532 | 0.5322 | 0.6369 | 0.5799 | 0.9274 | |
|
| 0.0033 | 26.0 | 8250 | 0.4273 | 0.5570 | 0.6227 | 0.5880 | 0.9309 | |
|
| 0.0032 | 27.0 | 9000 | 0.4415 | 0.5541 | 0.6317 | 0.5903 | 0.9297 | |
|
| 0.0025 | 28.0 | 9750 | 0.4509 | 0.5518 | 0.6272 | 0.5871 | 0.9291 | |
|
| 0.0021 | 29.0 | 10500 | 0.4652 | 0.5668 | 0.6179 | 0.5912 | 0.9308 | |
|
| 0.0026 | 30.0 | 11250 | 0.4622 | 0.5583 | 0.6294 | 0.5917 | 0.9297 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.35.2 |
|
- Pytorch 2.1.0+cu118 |
|
- Datasets 2.15.0 |
|
- Tokenizers 0.15.0 |
|
|
