arabert_cross_organization_task6_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: 0.4946
• Qwk: 0.7585
• Mse: 0.4958

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	0.125	2	1.3569	0.2073	1.3566
No log	0.25	4	0.8917	0.3804	0.8926
No log	0.375	6	1.3940	0.5002	1.3954
No log	0.5	8	1.1581	0.5869	1.1599
No log	0.625	10	0.8359	0.5005	0.8373
No log	0.75	12	0.8352	0.6240	0.8368
No log	0.875	14	0.7817	0.6808	0.7831
No log	1.0	16	0.6321	0.7284	0.6335
No log	1.125	18	0.5497	0.7013	0.5509
No log	1.25	20	0.6330	0.7733	0.6344
No log	1.375	22	0.9312	0.7286	0.9326
No log	1.5	24	0.9410	0.7373	0.9424
No log	1.625	26	0.6704	0.7704	0.6717
No log	1.75	28	0.5140	0.6654	0.5150
No log	1.875	30	0.5258	0.6207	0.5266
No log	2.0	32	0.4993	0.7208	0.5003
No log	2.125	34	0.5995	0.7661	0.6008
No log	2.25	36	0.6850	0.7821	0.6865
No log	2.375	38	0.6445	0.7839	0.6460
No log	2.5	40	0.5426	0.7571	0.5438
No log	2.625	42	0.5374	0.7584	0.5385
No log	2.75	44	0.5401	0.7508	0.5413
No log	2.875	46	0.5560	0.7716	0.5572
No log	3.0	48	0.5460	0.7794	0.5472
No log	3.125	50	0.5399	0.7800	0.5410
No log	3.25	52	0.4966	0.7520	0.4976
No log	3.375	54	0.4783	0.7484	0.4792
No log	3.5	56	0.5055	0.7654	0.5064
No log	3.625	58	0.4947	0.7569	0.4955
No log	3.75	60	0.5387	0.7681	0.5397
No log	3.875	62	0.6614	0.8077	0.6627
No log	4.0	64	0.6356	0.8243	0.6369
No log	4.125	66	0.4951	0.7545	0.4959
No log	4.25	68	0.4581	0.7123	0.4588
No log	4.375	70	0.4776	0.7450	0.4784
No log	4.5	72	0.5531	0.7823	0.5543
No log	4.625	74	0.5792	0.8103	0.5805
No log	4.75	76	0.5337	0.7801	0.5349
No log	4.875	78	0.4762	0.7597	0.4771
No log	5.0	80	0.4679	0.7390	0.4687
No log	5.125	82	0.4753	0.7488	0.4762
No log	5.25	84	0.5131	0.7689	0.5143
No log	5.375	86	0.5442	0.7925	0.5455
No log	5.5	88	0.5074	0.7624	0.5086
No log	5.625	90	0.4586	0.7435	0.4596
No log	5.75	92	0.4498	0.7269	0.4507
No log	5.875	94	0.4604	0.7354	0.4614
No log	6.0	96	0.5055	0.7753	0.5068
No log	6.125	98	0.5761	0.7991	0.5776
No log	6.25	100	0.5566	0.7942	0.5580
No log	6.375	102	0.5097	0.7509	0.5109
No log	6.5	104	0.4777	0.7454	0.4787
No log	6.625	106	0.4691	0.7225	0.4700
No log	6.75	108	0.4712	0.7283	0.4720
No log	6.875	110	0.4817	0.7509	0.4827
No log	7.0	112	0.4772	0.7454	0.4781
No log	7.125	114	0.4790	0.7490	0.4799
No log	7.25	116	0.5003	0.7688	0.5014
No log	7.375	118	0.5353	0.7753	0.5366
No log	7.5	120	0.5284	0.7670	0.5297
No log	7.625	122	0.5075	0.7556	0.5086
No log	7.75	124	0.4824	0.7527	0.4834
No log	7.875	126	0.4782	0.7527	0.4792
No log	8.0	128	0.4745	0.7554	0.4755
No log	8.125	130	0.4803	0.7523	0.4813
No log	8.25	132	0.4946	0.7614	0.4957
No log	8.375	134	0.4938	0.7558	0.4950
No log	8.5	136	0.4888	0.7558	0.4900
No log	8.625	138	0.4775	0.7507	0.4786
No log	8.75	140	0.4714	0.7474	0.4724
No log	8.875	142	0.4668	0.7410	0.4677
No log	9.0	144	0.4672	0.7382	0.4681
No log	9.125	146	0.4689	0.7433	0.4698
No log	9.25	148	0.4738	0.7571	0.4748
No log	9.375	150	0.4814	0.7511	0.4825
No log	9.5	152	0.4866	0.7567	0.4877
No log	9.625	154	0.4900	0.7585	0.4911
No log	9.75	156	0.4909	0.7585	0.4921
No log	9.875	158	0.4930	0.7585	0.4942
No log	10.0	160	0.4946	0.7585	0.4958

Framework versions

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