elnasharomar2/Bert-base-Keyword-Extraction

bert-base-arabic-camelbert-ca-ner_oknashar

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

• Loss: 0.2492
• Precision: 0.7007
• Recall: 0.7475
• F1: 0.7234
• Accuracy: 0.9683

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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.1376	1.0	1357	0.1260	0.5569	0.5220	0.5389	0.9565
0.0874	2.0	2714	0.1150	0.6264	0.6198	0.6231	0.9616
0.0578	3.0	4071	0.1162	0.6598	0.6567	0.6583	0.9663
0.0372	4.0	5428	0.1336	0.6695	0.6837	0.6765	0.9662
0.0257	5.0	6785	0.1667	0.6276	0.6989	0.6613	0.9617
0.0183	6.0	8142	0.1537	0.6876	0.7258	0.7062	0.9684
0.0139	7.0	9499	0.1813	0.6860	0.7258	0.7054	0.9672
0.0096	8.0	10856	0.1935	0.6802	0.7165	0.6979	0.9673
0.0076	9.0	12213	0.1880	0.7335	0.7270	0.7302	0.9699
0.005	10.0	13570	0.2266	0.7070	0.7352	0.7209	0.9683
0.0041	11.0	14927	0.2340	0.7011	0.7405	0.7202	0.9681
0.0032	12.0	16284	0.2335	0.7146	0.7364	0.7253	0.9687
0.0031	13.0	17641	0.2461	0.7060	0.7499	0.7273	0.9683
0.0027	14.0	18998	0.2421	0.7088	0.7428	0.7254	0.9686
0.0015	15.0	20355	0.2492	0.7007	0.7475	0.7234	0.9683

Framework versions

• Transformers 4.34.1
• Pytorch 2.1.0+cu118
• Datasets 2.14.5
• Tokenizers 0.14.1