arabert_cross_relevance_task2_fold3

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: 0.2868
• Qwk: 0.4931
• Mse: 0.2868

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	0.5163	0.2249	0.5163
No log	0.2667	4	0.4386	0.1958	0.4386
No log	0.4	6	0.4172	0.1958	0.4172
No log	0.5333	8	0.3488	0.2555	0.3488
No log	0.6667	10	0.3483	0.2974	0.3483
No log	0.8	12	0.3327	0.2718	0.3327
No log	0.9333	14	0.3570	0.2289	0.3570
No log	1.0667	16	0.3857	0.3184	0.3857
No log	1.2	18	0.3553	0.2711	0.3553
No log	1.3333	20	0.3072	0.2825	0.3072
No log	1.4667	22	0.2870	0.3248	0.2870
No log	1.6	24	0.2719	0.3375	0.2719
No log	1.7333	26	0.2776	0.3974	0.2776
No log	1.8667	28	0.2783	0.4688	0.2783
No log	2.0	30	0.2844	0.5250	0.2844
No log	2.1333	32	0.2948	0.5193	0.2948
No log	2.2667	34	0.2802	0.375	0.2802
No log	2.4	36	0.2657	0.3297	0.2657
No log	2.5333	38	0.2635	0.3258	0.2635
No log	2.6667	40	0.2700	0.3347	0.2700
No log	2.8	42	0.2862	0.3299	0.2862
No log	2.9333	44	0.3057	0.3666	0.3057
No log	3.0667	46	0.3164	0.4048	0.3164
No log	3.2	48	0.3074	0.3977	0.3074
No log	3.3333	50	0.2918	0.3729	0.2918
No log	3.4667	52	0.2907	0.4232	0.2907
No log	3.6	54	0.2835	0.4066	0.2835
No log	3.7333	56	0.2768	0.3894	0.2768
No log	3.8667	58	0.2866	0.4187	0.2866
No log	4.0	60	0.3010	0.4186	0.3010
No log	4.1333	62	0.3104	0.4066	0.3104
No log	4.2667	64	0.2965	0.3773	0.2965
No log	4.4	66	0.2718	0.3398	0.2718
No log	4.5333	68	0.2585	0.3586	0.2585
No log	4.6667	70	0.2610	0.3481	0.2610
No log	4.8	72	0.2741	0.3533	0.2741
No log	4.9333	74	0.2979	0.4156	0.2979
No log	5.0667	76	0.3007	0.4214	0.3007
No log	5.2	78	0.2959	0.3980	0.2959
No log	5.3333	80	0.2864	0.3369	0.2864
No log	5.4667	82	0.2716	0.3458	0.2716
No log	5.6	84	0.2611	0.3307	0.2611
No log	5.7333	86	0.2596	0.3373	0.2596
No log	5.8667	88	0.2680	0.3583	0.2680
No log	6.0	90	0.2914	0.3969	0.2914
No log	6.1333	92	0.3171	0.4789	0.3171
No log	6.2667	94	0.3159	0.5187	0.3159
No log	6.4	96	0.3015	0.4948	0.3015
No log	6.5333	98	0.2857	0.4598	0.2857
No log	6.6667	100	0.2723	0.4572	0.2723
No log	6.8	102	0.2683	0.4454	0.2683
No log	6.9333	104	0.2753	0.4473	0.2753
No log	7.0667	106	0.2829	0.4644	0.2829
No log	7.2	108	0.2793	0.4531	0.2793
No log	7.3333	110	0.2787	0.4531	0.2787
No log	7.4667	112	0.2754	0.4531	0.2754
No log	7.6	114	0.2767	0.4600	0.2767
No log	7.7333	116	0.2854	0.4880	0.2854
No log	7.8667	118	0.2920	0.5047	0.2920
No log	8.0	120	0.2949	0.5212	0.2949
No log	8.1333	122	0.2935	0.5204	0.2935
No log	8.2667	124	0.2888	0.5040	0.2888
No log	8.4	126	0.2843	0.4986	0.2843
No log	8.5333	128	0.2811	0.5017	0.2811
No log	8.6667	130	0.2800	0.5017	0.2800
No log	8.8	132	0.2856	0.5125	0.2856
No log	8.9333	134	0.2900	0.4986	0.2900
No log	9.0667	136	0.2898	0.4986	0.2898
No log	9.2	138	0.2908	0.5040	0.2908
No log	9.3333	140	0.2914	0.5095	0.2914
No log	9.4667	142	0.2916	0.5047	0.2916
No log	9.6	144	0.2901	0.4986	0.2901
No log	9.7333	146	0.2879	0.4931	0.2879
No log	9.8667	148	0.2871	0.4931	0.2871
No log	10.0	150	0.2868	0.4931	0.2868

Framework versions

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