arabert_cross_relevance_task3_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: 0.2596
• Qwk: 0.4029
• Mse: 0.2596

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	0.5262	0.1787	0.5262
No log	0.25	4	0.4596	0.2239	0.4596
No log	0.375	6	0.3362	0.2878	0.3362
No log	0.5	8	0.3282	0.2999	0.3282
No log	0.625	10	0.3253	0.3165	0.3253
No log	0.75	12	0.3138	0.2765	0.3138
No log	0.875	14	0.2956	0.2948	0.2956
No log	1.0	16	0.2879	0.3245	0.2879
No log	1.125	18	0.2688	0.3611	0.2688
No log	1.25	20	0.2662	0.3671	0.2662
No log	1.375	22	0.2655	0.3995	0.2655
No log	1.5	24	0.2543	0.3783	0.2543
No log	1.625	26	0.2646	0.3143	0.2646
No log	1.75	28	0.2991	0.2752	0.2991
No log	1.875	30	0.3254	0.2955	0.3254
No log	2.0	32	0.3327	0.3366	0.3327
No log	2.125	34	0.2988	0.3162	0.2988
No log	2.25	36	0.2557	0.3439	0.2557
No log	2.375	38	0.2529	0.3569	0.2529
No log	2.5	40	0.2597	0.3841	0.2597
No log	2.625	42	0.3219	0.4642	0.3219
No log	2.75	44	0.3567	0.4279	0.3567
No log	2.875	46	0.3379	0.3457	0.3379
No log	3.0	48	0.2854	0.3137	0.2854
No log	3.125	50	0.2618	0.3617	0.2618
No log	3.25	52	0.2662	0.4309	0.2662
No log	3.375	54	0.2959	0.4928	0.2959
No log	3.5	56	0.2933	0.5115	0.2933
No log	3.625	58	0.2656	0.5054	0.2656
No log	3.75	60	0.2489	0.4257	0.2489
No log	3.875	62	0.2494	0.3892	0.2494
No log	4.0	64	0.2669	0.4115	0.2669
No log	4.125	66	0.3092	0.4465	0.3092
No log	4.25	68	0.3149	0.4704	0.3149
No log	4.375	70	0.2852	0.4302	0.2852
No log	4.5	72	0.2544	0.4010	0.2544
No log	4.625	74	0.2597	0.3835	0.2597
No log	4.75	76	0.2622	0.3893	0.2622
No log	4.875	78	0.2879	0.3962	0.2879
No log	5.0	80	0.2944	0.4007	0.2944
No log	5.125	82	0.2653	0.3953	0.2653
No log	5.25	84	0.2441	0.3922	0.2441
No log	5.375	86	0.2428	0.4033	0.2428
No log	5.5	88	0.2581	0.4593	0.2581
No log	5.625	90	0.2908	0.4953	0.2908
No log	5.75	92	0.2948	0.4521	0.2948
No log	5.875	94	0.2783	0.3762	0.2783
No log	6.0	96	0.2471	0.3316	0.2471
No log	6.125	98	0.2369	0.3492	0.2369
No log	6.25	100	0.2362	0.3548	0.2362
No log	6.375	102	0.2417	0.3613	0.2417
No log	6.5	104	0.2591	0.4177	0.2591
No log	6.625	106	0.2639	0.4581	0.2639
No log	6.75	108	0.2534	0.4358	0.2534
No log	6.875	110	0.2433	0.4075	0.2433
No log	7.0	112	0.2434	0.3823	0.2434
No log	7.125	114	0.2521	0.3648	0.2521
No log	7.25	116	0.2750	0.3827	0.2750
No log	7.375	118	0.2989	0.3637	0.2989
No log	7.5	120	0.2984	0.3630	0.2984
No log	7.625	122	0.2963	0.3628	0.2963
No log	7.75	124	0.2861	0.3757	0.2861
No log	7.875	126	0.2762	0.3896	0.2762
No log	8.0	128	0.2661	0.4135	0.2661
No log	8.125	130	0.2529	0.4042	0.2529
No log	8.25	132	0.2492	0.4148	0.2492
No log	8.375	134	0.2525	0.4679	0.2525
No log	8.5	136	0.2600	0.4853	0.2600
No log	8.625	138	0.2656	0.4866	0.2656
No log	8.75	140	0.2658	0.4866	0.2658
No log	8.875	142	0.2693	0.4694	0.2693
No log	9.0	144	0.2682	0.4416	0.2682
No log	9.125	146	0.2635	0.4304	0.2635
No log	9.25	148	0.2593	0.4140	0.2593
No log	9.375	150	0.2591	0.4029	0.2591
No log	9.5	152	0.2617	0.4082	0.2617
No log	9.625	154	0.2621	0.4026	0.2621
No log	9.75	156	0.2609	0.4026	0.2609
No log	9.875	158	0.2598	0.4029	0.2598
No log	10.0	160	0.2596	0.4029	0.2596

Framework versions

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