ANER_arabic_keyword_extraction_dataset1

This model is a fine-tuned version of elnasharomar2/ANER_arabic_keyword_extraction_dataset1 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1708
• Precision: 0.7713
• Recall: 0.7690
• F1: 0.7701
• Accuracy: 0.9767

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
0.0062	1.0	675	0.1415	0.74	0.7404	0.7402	0.9737
0.0074	2.0	1350	0.1452	0.7086	0.7726	0.7392	0.9727
0.0061	3.0	2025	0.1356	0.7296	0.7447	0.7371	0.9742
0.0051	4.0	2700	0.1448	0.7456	0.7374	0.7415	0.9743
0.0038	5.0	3375	0.1437	0.7696	0.7453	0.7572	0.9763
0.0029	6.0	4050	0.1555	0.7702	0.7562	0.7632	0.9763
0.0028	7.0	4725	0.1500	0.7636	0.7483	0.7559	0.9757
0.0028	8.0	5400	0.1522	0.7648	0.7574	0.7611	0.9761
0.0019	9.0	6075	0.1585	0.7584	0.7671	0.7627	0.9757
0.002	10.0	6750	0.1637	0.7567	0.7659	0.7613	0.9754
0.0019	11.0	7425	0.1686	0.7783	0.7514	0.7646	0.9760
0.0013	12.0	8100	0.1659	0.7877	0.7538	0.7704	0.9770
0.0011	13.0	8775	0.1636	0.7759	0.7683	0.7721	0.9767
0.001	14.0	9450	0.1720	0.7733	0.7635	0.7684	0.9765
0.0009	15.0	10125	0.1708	0.7713	0.7690	0.7701	0.9767

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0