arabert_cross_organization_task4_fold5

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: 0.4843
• Qwk: 0.7666
• Mse: 0.4857

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	0.9490	0.2150	0.9491
No log	0.2353	4	0.8716	0.5434	0.8727
No log	0.3529	6	0.6659	0.6268	0.6673
No log	0.4706	8	0.5369	0.6777	0.5386
No log	0.5882	10	0.6440	0.7813	0.6460
No log	0.7059	12	0.5256	0.6777	0.5270
No log	0.8235	14	0.4966	0.7039	0.4980
No log	0.9412	16	0.7392	0.7069	0.7410
No log	1.0588	18	0.8304	0.7187	0.8325
No log	1.1765	20	0.6276	0.7906	0.6292
No log	1.2941	22	0.4520	0.7197	0.4531
No log	1.4118	24	0.4614	0.7447	0.4625
No log	1.5294	26	0.5417	0.7758	0.5430
No log	1.6471	28	0.4689	0.7534	0.4700
No log	1.7647	30	0.4572	0.7731	0.4583
No log	1.8824	32	0.6532	0.7981	0.6546
No log	2.0	34	0.7591	0.7855	0.7607
No log	2.1176	36	0.5854	0.7951	0.5867
No log	2.2353	38	0.4474	0.7384	0.4483
No log	2.3529	40	0.4483	0.7531	0.4492
No log	2.4706	42	0.5873	0.8112	0.5887
No log	2.5882	44	0.6933	0.8094	0.6948
No log	2.7059	46	0.5657	0.7996	0.5670
No log	2.8235	48	0.4349	0.7582	0.4358
No log	2.9412	50	0.4131	0.7274	0.4139
No log	3.0588	52	0.4231	0.7410	0.4241
No log	3.1765	54	0.4993	0.7809	0.5006
No log	3.2941	56	0.5623	0.8152	0.5637
No log	3.4118	58	0.4818	0.7688	0.4831
No log	3.5294	60	0.4394	0.7381	0.4405
No log	3.6471	62	0.4350	0.7249	0.4361
No log	3.7647	64	0.4743	0.7662	0.4756
No log	3.8824	66	0.5838	0.8110	0.5853
No log	4.0	68	0.6906	0.7966	0.6921
No log	4.1176	70	0.6463	0.7987	0.6478
No log	4.2353	72	0.4721	0.7595	0.4733
No log	4.3529	74	0.4408	0.7408	0.4419
No log	4.4706	76	0.4907	0.7768	0.4920
No log	4.5882	78	0.5961	0.7936	0.5977
No log	4.7059	80	0.5925	0.7830	0.5941
No log	4.8235	82	0.5242	0.7638	0.5257
No log	4.9412	84	0.4430	0.7330	0.4442
No log	5.0588	86	0.4332	0.7022	0.4342
No log	5.1765	88	0.4549	0.7664	0.4561
No log	5.2941	90	0.5563	0.7685	0.5577
No log	5.4118	92	0.6249	0.7924	0.6265
No log	5.5294	94	0.5569	0.7721	0.5585
No log	5.6471	96	0.4954	0.7701	0.4969
No log	5.7647	98	0.4434	0.7420	0.4446
No log	5.8824	100	0.4310	0.7481	0.4320
No log	6.0	102	0.4622	0.7629	0.4634
No log	6.1176	104	0.5374	0.7903	0.5387
No log	6.2353	106	0.5235	0.7829	0.5248
No log	6.3529	108	0.4601	0.7688	0.4612
No log	6.4706	110	0.4260	0.7327	0.4269
No log	6.5882	112	0.4318	0.7362	0.4327
No log	6.7059	114	0.4747	0.7719	0.4758
No log	6.8235	116	0.5753	0.7740	0.5767
No log	6.9412	118	0.5967	0.7736	0.5981
No log	7.0588	120	0.5343	0.7711	0.5356
No log	7.1765	122	0.4499	0.7556	0.4510
No log	7.2941	124	0.4329	0.7295	0.4337
No log	7.4118	126	0.4364	0.7376	0.4373
No log	7.5294	128	0.4672	0.7700	0.4683
No log	7.6471	130	0.5413	0.7754	0.5427
No log	7.7647	132	0.5644	0.7761	0.5659
No log	7.8824	134	0.5227	0.7749	0.5241
No log	8.0	136	0.4791	0.7688	0.4804
No log	8.1176	138	0.4693	0.7580	0.4705
No log	8.2353	140	0.4639	0.7580	0.4652
No log	8.3529	142	0.4809	0.7685	0.4822
No log	8.4706	144	0.4940	0.7700	0.4954
No log	8.5882	146	0.4956	0.7700	0.4970
No log	8.7059	148	0.4916	0.7753	0.4930
No log	8.8235	150	0.4799	0.7651	0.4812
No log	8.9412	152	0.4697	0.7580	0.4709
No log	9.0588	154	0.4612	0.7573	0.4624
No log	9.1765	156	0.4541	0.7549	0.4552
No log	9.2941	158	0.4549	0.7566	0.4561
No log	9.4118	160	0.4603	0.7527	0.4616
No log	9.5294	162	0.4654	0.7544	0.4667
No log	9.6471	164	0.4728	0.7597	0.4742
No log	9.7647	166	0.4786	0.7597	0.4800
No log	9.8824	168	0.4825	0.7685	0.4838
No log	10.0	170	0.4843	0.7666	0.4857

Framework versions

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