arabert_baseline_relevance_task8_fold0

This model is a fine-tuned version of aubmindlab/bert-base-arabertv02 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1603
• Qwk: 0.0
• Mse: 0.1603

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: 16
• 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	Qwk	Mse
No log	0.5	2	0.8396	0.0628	0.8396
No log	1.0	4	0.0941	0.0870	0.0941
No log	1.5	6	0.1025	0.0	0.1025
No log	2.0	8	0.1788	0.0	0.1788
No log	2.5	10	0.2166	0.1158	0.2166
No log	3.0	12	0.3700	0.2013	0.3700
No log	3.5	14	0.3262	0.2519	0.3262
No log	4.0	16	0.1386	0.2519	0.1386
No log	4.5	18	0.1148	0.0	0.1148
No log	5.0	20	0.1624	0.0	0.1624
No log	5.5	22	0.1332	0.0	0.1332
No log	6.0	24	0.1031	0.0411	0.1031
No log	6.5	26	0.1199	0.0	0.1199
No log	7.0	28	0.1398	0.0	0.1398
No log	7.5	30	0.1573	0.0	0.1573
No log	8.0	32	0.1592	0.1158	0.1592
No log	8.5	34	0.1587	0.1158	0.1587
No log	9.0	36	0.1588	0.0	0.1588
No log	9.5	38	0.1591	0.0	0.1591
No log	10.0	40	0.1603	0.0	0.1603

Framework versions

• Transformers 4.44.0
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1