bert-base-arabertv2

This model is a fine-tuned version of aubmindlab/bert-base-arabertv2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8449
• Precision: 0.5
• Recall: 0.5938
• F1: 0.5429
• Accuracy: 0.6461

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	5	0.8166	0.0	0.0	0.0	0.6383
No log	2.0	10	0.8143	0.0	0.0	0.0	0.6915
No log	3.0	15	0.7780	0.0	0.0	0.0	0.6702
No log	4.0	20	0.7524	0.0	0.0	0.0	0.7234
No log	5.0	25	0.7472	0.0	0.0	0.0	0.7660
No log	6.0	30	0.7509	0.0	0.0	0.0	0.7340
No log	7.0	35	0.7163	0.0	0.0	0.0	0.7872
No log	8.0	40	0.7020	0.0	0.0	0.0	0.7872
No log	9.0	45	0.6960	0.0	0.0	0.0	0.7872
No log	10.0	50	0.6729	0.0	0.0	0.0	0.8085

Framework versions

• Transformers 4.48.1
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0