arabert_cross_organization_task5_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: 0.8536
• Qwk: 0.5305
• Mse: 0.8521

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	0.1333	2	2.8325	0.0355	2.8291
No log	0.2667	4	1.6790	0.2071	1.6769
No log	0.4	6	1.2491	0.3232	1.2479
No log	0.5333	8	1.0946	0.4006	1.0940
No log	0.6667	10	0.9240	0.5244	0.9238
No log	0.8	12	1.0994	0.4812	1.0991
No log	0.9333	14	1.0985	0.4785	1.0981
No log	1.0667	16	0.9091	0.5213	0.9088
No log	1.2	18	0.8691	0.5235	0.8687
No log	1.3333	20	1.1009	0.4480	1.1002
No log	1.4667	22	0.9916	0.4796	0.9909
No log	1.6	24	0.8206	0.5231	0.8200
No log	1.7333	26	0.8819	0.5255	0.8812
No log	1.8667	28	1.1165	0.4625	1.1156
No log	2.0	30	0.9537	0.5260	0.9531
No log	2.1333	32	0.7775	0.6092	0.7773
No log	2.2667	34	0.8262	0.5792	0.8258
No log	2.4	36	0.9527	0.5326	0.9519
No log	2.5333	38	0.8990	0.5382	0.8983
No log	2.6667	40	0.7673	0.6124	0.7669
No log	2.8	42	0.7340	0.6181	0.7335
No log	2.9333	44	0.8826	0.5543	0.8815
No log	3.0667	46	1.2982	0.4128	1.2965
No log	3.2	48	1.1748	0.4370	1.1731
No log	3.3333	50	0.7851	0.5568	0.7840
No log	3.4667	52	0.6933	0.6150	0.6924
No log	3.6	54	0.7593	0.5656	0.7582
No log	3.7333	56	0.9243	0.4943	0.9229
No log	3.8667	58	0.9098	0.5117	0.9084
No log	4.0	60	0.7896	0.5674	0.7885
No log	4.1333	62	0.7231	0.6204	0.7224
No log	4.2667	64	0.7654	0.6006	0.7646
No log	4.4	66	0.8645	0.5394	0.8632
No log	4.5333	68	0.8997	0.5331	0.8983
No log	4.6667	70	0.7948	0.5651	0.7936
No log	4.8	72	0.7556	0.5765	0.7545
No log	4.9333	74	0.7815	0.5752	0.7803
No log	5.0667	76	0.7767	0.5824	0.7755
No log	5.2	78	0.8240	0.5627	0.8228
No log	5.3333	80	0.8544	0.5524	0.8531
No log	5.4667	82	0.8799	0.5383	0.8785
No log	5.6	84	0.8068	0.5637	0.8055
No log	5.7333	86	0.7866	0.5790	0.7853
No log	5.8667	88	0.8602	0.5505	0.8588
No log	6.0	90	0.8403	0.5508	0.8390
No log	6.1333	92	0.7915	0.5599	0.7903
No log	6.2667	94	0.7576	0.5889	0.7565
No log	6.4	96	0.7814	0.5709	0.7802
No log	6.5333	98	0.7840	0.5682	0.7828
No log	6.6667	100	0.8376	0.5500	0.8362
No log	6.8	102	0.8823	0.5259	0.8808
No log	6.9333	104	0.8409	0.5388	0.8395
No log	7.0667	106	0.7908	0.5700	0.7895
No log	7.2	108	0.8370	0.5398	0.8356
No log	7.3333	110	0.9465	0.5061	0.9448
No log	7.4667	112	0.9805	0.4909	0.9787
No log	7.6	114	0.8809	0.5217	0.8793
No log	7.7333	116	0.7546	0.5828	0.7533
No log	7.8667	118	0.7282	0.5855	0.7270
No log	8.0	120	0.7740	0.5703	0.7727
No log	8.1333	122	0.8664	0.5129	0.8649
No log	8.2667	124	0.8853	0.5069	0.8838
No log	8.4	126	0.8189	0.5483	0.8175
No log	8.5333	128	0.7717	0.5683	0.7704
No log	8.6667	130	0.7625	0.5761	0.7613
No log	8.8	132	0.7903	0.5682	0.7890
No log	8.9333	134	0.8518	0.5384	0.8503
No log	9.0667	136	0.9168	0.5056	0.9152
No log	9.2	138	0.9669	0.4922	0.9652
No log	9.3333	140	0.9617	0.4944	0.9600
No log	9.4667	142	0.9232	0.5081	0.9216
No log	9.6	144	0.8847	0.5249	0.8831
No log	9.7333	146	0.8666	0.5296	0.8651
No log	9.8667	148	0.8562	0.5305	0.8547
No log	10.0	150	0.8536	0.5305	0.8521

Framework versions

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