results

This model is a fine-tuned version of CAMeL-Lab/bert-base-arabic-camelbert-msa-ner on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0671
• Precision: 0.9393
• Recall: 0.9453
• F1: 0.9423
• Accuracy: 0.9849

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: Use OptimizerNames.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	21	0.7573	0.688	0.5432	0.6071	0.8476
No log	2.0	42	0.4541	0.7551	0.6295	0.6866	0.8865
No log	3.0	63	0.3165	0.8498	0.7979	0.8230	0.9365
No log	4.0	84	0.2295	0.8636	0.8	0.8306	0.9456
No log	5.0	105	0.1669	0.9089	0.8611	0.8843	0.9599
No log	6.0	126	0.1222	0.8966	0.8947	0.8957	0.9679
No log	7.0	147	0.0967	0.9242	0.9242	0.9242	0.9754
No log	8.0	168	0.0775	0.9349	0.9368	0.9359	0.9825
No log	9.0	189	0.0685	0.9351	0.9411	0.9381	0.9841
No log	10.0	210	0.0671	0.9393	0.9453	0.9423	0.9849

Framework versions

• Transformers 4.52.4
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.2