AraBERT_token_classification_AraEval24_back_translation_augmented

This model is a fine-tuned version of aubmindlab/bert-base-arabert on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8791
• Precision: 0.0803
• Recall: 0.0309
• F1: 0.0447
• Accuracy: 0.8670

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: 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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.6238	1.0	3770	0.7220	0.2083	0.0009	0.0018	0.8767
0.5368	2.0	7540	0.7270	0.1944	0.0012	0.0024	0.8766
0.4918	3.0	11310	0.7255	0.0519	0.0042	0.0078	0.8718
0.4385	4.0	15080	0.7570	0.0817	0.0060	0.0111	0.8758
0.3833	5.0	18850	0.7710	0.0753	0.0220	0.0340	0.8714
0.3617	6.0	22620	0.7781	0.0689	0.0185	0.0291	0.8693
0.336	7.0	26390	0.8184	0.0888	0.0248	0.0388	0.8711
0.3176	8.0	30160	0.8289	0.0771	0.0232	0.0357	0.8706
0.2943	9.0	33930	0.8600	0.0768	0.0251	0.0379	0.8709
0.2856	10.0	37700	0.8791	0.0803	0.0309	0.0447	0.8670

Framework versions

• Transformers 4.30.2
• Pytorch 1.12.1
• Datasets 2.13.2
• Tokenizers 0.13.3