arabert_cross_organization_task6_fold6

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: 0.6971
• Qwk: 0.5467
• Mse: 0.6953

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: 64
• eval_batch_size: 64
• 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	0.1176	2	2.1812	0.0813	2.1796
No log	0.2353	4	1.1581	0.1541	1.1560
No log	0.3529	6	0.9175	0.4300	0.9172
No log	0.4706	8	0.7853	0.5387	0.7849
No log	0.5882	10	0.7421	0.3500	0.7415
No log	0.7059	12	0.7475	0.3575	0.7469
No log	0.8235	14	0.5277	0.6055	0.5272
No log	0.9412	16	0.6059	0.7252	0.6066
No log	1.0588	18	0.5616	0.7136	0.5620
No log	1.1765	20	0.5211	0.6304	0.5204
No log	1.2941	22	0.6860	0.5190	0.6841
No log	1.4118	24	0.6738	0.5216	0.6721
No log	1.5294	26	0.5480	0.6313	0.5472
No log	1.6471	28	0.5424	0.7060	0.5428
No log	1.7647	30	0.4918	0.6873	0.4920
No log	1.8824	32	0.5098	0.5684	0.5094
No log	2.0	34	0.5437	0.5294	0.5430
No log	2.1176	36	0.5312	0.5901	0.5302
No log	2.2353	38	0.5616	0.5966	0.5604
No log	2.3529	40	0.5882	0.5838	0.5868
No log	2.4706	42	0.5423	0.6000	0.5413
No log	2.5882	44	0.5067	0.6211	0.5059
No log	2.7059	46	0.4934	0.6349	0.4926
No log	2.8235	48	0.4940	0.6329	0.4932
No log	2.9412	50	0.5291	0.5677	0.5279
No log	3.0588	52	0.6166	0.5158	0.6151
No log	3.1765	54	0.6014	0.5588	0.5998
No log	3.2941	56	0.5316	0.5878	0.5303
No log	3.4118	58	0.5135	0.5990	0.5124
No log	3.5294	60	0.5285	0.5827	0.5273
No log	3.6471	62	0.5943	0.5492	0.5929
No log	3.7647	64	0.5882	0.5620	0.5868
No log	3.8824	66	0.5237	0.5937	0.5227
No log	4.0	68	0.5270	0.6150	0.5261
No log	4.1176	70	0.5820	0.5589	0.5806
No log	4.2353	72	0.6445	0.5284	0.6429
No log	4.3529	74	0.6153	0.5627	0.6139
No log	4.4706	76	0.6066	0.5783	0.6054
No log	4.5882	78	0.6378	0.5639	0.6363
No log	4.7059	80	0.7155	0.5342	0.7135
No log	4.8235	82	0.7123	0.5305	0.7104
No log	4.9412	84	0.6786	0.5363	0.6769
No log	5.0588	86	0.6340	0.5611	0.6326
No log	5.1765	88	0.6050	0.5630	0.6038
No log	5.2941	90	0.6307	0.5564	0.6293
No log	5.4118	92	0.6603	0.5449	0.6588
No log	5.5294	94	0.6765	0.5483	0.6748
No log	5.6471	96	0.6364	0.5686	0.6351
No log	5.7647	98	0.6144	0.5967	0.6132
No log	5.8824	100	0.6315	0.5826	0.6300
No log	6.0	102	0.6964	0.5217	0.6946
No log	6.1176	104	0.6906	0.5310	0.6887
No log	6.2353	106	0.6656	0.5513	0.6639
No log	6.3529	108	0.6273	0.5829	0.6259
No log	6.4706	110	0.6354	0.5748	0.6340
No log	6.5882	112	0.6855	0.5397	0.6839
No log	6.7059	114	0.7228	0.5179	0.7211
No log	6.8235	116	0.6976	0.5206	0.6960
No log	6.9412	118	0.6558	0.5456	0.6544
No log	7.0588	120	0.6618	0.5569	0.6605
No log	7.1765	122	0.7088	0.5397	0.7072
No log	7.2941	124	0.8015	0.4900	0.7996
No log	7.4118	126	0.8354	0.4798	0.8334
No log	7.5294	128	0.7861	0.4973	0.7842
No log	7.6471	130	0.7081	0.5399	0.7065
No log	7.7647	132	0.6756	0.5725	0.6741
No log	7.8824	134	0.6874	0.5524	0.6859
No log	8.0	136	0.7225	0.5459	0.7207
No log	8.1176	138	0.7336	0.5368	0.7317
No log	8.2353	140	0.7330	0.5258	0.7312
No log	8.3529	142	0.7088	0.5474	0.7070
No log	8.4706	144	0.7009	0.5474	0.6991
No log	8.5882	146	0.6880	0.5489	0.6863
No log	8.7059	148	0.6695	0.5530	0.6679
No log	8.8235	150	0.6654	0.5587	0.6638
No log	8.9412	152	0.6764	0.5464	0.6747
No log	9.0588	154	0.6945	0.5419	0.6927
No log	9.1765	156	0.7081	0.5278	0.7063
No log	9.2941	158	0.7099	0.5292	0.7080
No log	9.4118	160	0.7043	0.5419	0.7025
No log	9.5294	162	0.7049	0.5430	0.7031
No log	9.6471	164	0.7019	0.5430	0.7000
No log	9.7647	166	0.6996	0.5467	0.6978
No log	9.8824	168	0.6977	0.5467	0.6959
No log	10.0	170	0.6971	0.5467	0.6953

Framework versions

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