arabert_baseline_relevance_task7_fold1

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: 0.1811
• Qwk: 0.4857
• Mse: 0.1829

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	0.3333	2	0.2365	0.3424	0.2296
No log	0.6667	4	0.2621	0.4112	0.2567
No log	1.0	6	0.3169	0.5670	0.3168
No log	1.3333	8	0.3023	0.5145	0.2938
No log	1.6667	10	0.2629	0.2310	0.2513
No log	2.0	12	0.2208	0.4676	0.2109
No log	2.3333	14	0.2071	0.5410	0.2005
No log	2.6667	16	0.1961	0.5410	0.1905
No log	3.0	18	0.1888	0.5093	0.1831
No log	3.3333	20	0.1855	0.5093	0.1806
No log	3.6667	22	0.1838	0.4967	0.1793
No log	4.0	24	0.1870	0.4857	0.1820
No log	4.3333	26	0.1864	0.4857	0.1822
No log	4.6667	28	0.1812	0.4857	0.1788
No log	5.0	30	0.1754	0.4857	0.1742
No log	5.3333	32	0.1723	0.4967	0.1722
No log	5.6667	34	0.1695	0.5695	0.1704
No log	6.0	36	0.1654	0.5695	0.1667
No log	6.3333	38	0.1652	0.5093	0.1667
No log	6.6667	40	0.1707	0.4857	0.1722
No log	7.0	42	0.1771	0.4857	0.1786
No log	7.3333	44	0.1814	0.4857	0.1829
No log	7.6667	46	0.1782	0.4857	0.1801
No log	8.0	48	0.1775	0.5389	0.1799
No log	8.3333	50	0.1799	0.5389	0.1824
No log	8.6667	52	0.1801	0.5389	0.1825
No log	9.0	54	0.1806	0.4857	0.1828
No log	9.3333	56	0.1816	0.4857	0.1837
No log	9.6667	58	0.1812	0.4857	0.1831
No log	10.0	60	0.1811	0.4857	0.1829

Framework versions

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