arabert_cross_relevance_task5_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: 0.3040
• Qwk: 0.0224
• Mse: 0.3040

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	0.3814	0.0319	0.3814
No log	0.2353	4	0.4935	0.0934	0.4935
No log	0.3529	6	0.3556	0.1346	0.3556
No log	0.4706	8	0.2761	0.0	0.2761
No log	0.5882	10	0.3024	0.0	0.3024
No log	0.7059	12	0.2997	-0.0517	0.2997
No log	0.8235	14	0.3140	-0.1029	0.3140
No log	0.9412	16	0.2979	-0.0473	0.2979
No log	1.0588	18	0.2994	0.0	0.2994
No log	1.1765	20	0.2824	0.0	0.2824
No log	1.2941	22	0.2570	0.0	0.2570
No log	1.4118	24	0.2638	-0.0235	0.2638
No log	1.5294	26	0.2681	-0.0235	0.2681
No log	1.6471	28	0.2587	-0.0235	0.2587
No log	1.7647	30	0.2781	0.0	0.2781
No log	1.8824	32	0.3225	0.0	0.3225
No log	2.0	34	0.2895	0.0	0.2895
No log	2.1176	36	0.2779	0.0	0.2779
No log	2.2353	38	0.2807	0.0	0.2807
No log	2.3529	40	0.2808	0.0	0.2808
No log	2.4706	42	0.2648	0.0	0.2648
No log	2.5882	44	0.2627	0.0	0.2627
No log	2.7059	46	0.2723	0.0	0.2723
No log	2.8235	48	0.2669	0.0	0.2669
No log	2.9412	50	0.2817	0.0	0.2817
No log	3.0588	52	0.2862	0.0	0.2862
No log	3.1765	54	0.2710	0.0	0.2710
No log	3.2941	56	0.2691	0.0	0.2691
No log	3.4118	58	0.2758	0.0	0.2758
No log	3.5294	60	0.2757	0.0	0.2757
No log	3.6471	62	0.2734	0.0	0.2734
No log	3.7647	64	0.2619	0.0	0.2619
No log	3.8824	66	0.2661	0.0	0.2661
No log	4.0	68	0.2833	0.0	0.2833
No log	4.1176	70	0.2841	0.0	0.2841
No log	4.2353	72	0.2826	0.0	0.2826
No log	4.3529	74	0.2721	0.0	0.2721
No log	4.4706	76	0.2658	0.0	0.2658
No log	4.5882	78	0.2740	0.0	0.2740
No log	4.7059	80	0.2745	0.0	0.2745
No log	4.8235	82	0.2783	0.0	0.2783
No log	4.9412	84	0.2774	0.0	0.2774
No log	5.0588	86	0.2885	0.0	0.2885
No log	5.1765	88	0.3222	0.0	0.3222
No log	5.2941	90	0.3391	0.0	0.3391
No log	5.4118	92	0.3189	0.0	0.3189
No log	5.5294	94	0.3185	0.0224	0.3185
No log	5.6471	96	0.2891	0.0	0.2891
No log	5.7647	98	0.2697	0.0	0.2697
No log	5.8824	100	0.2703	0.0	0.2703
No log	6.0	102	0.2749	0.0	0.2749
No log	6.1176	104	0.2900	0.0	0.2900
No log	6.2353	106	0.3272	0.0	0.3272
No log	6.3529	108	0.3347	0.0224	0.3347
No log	6.4706	110	0.3020	0.0	0.3020
No log	6.5882	112	0.2731	0.0	0.2731
No log	6.7059	114	0.2683	0.0	0.2683
No log	6.8235	116	0.2736	0.0	0.2736
No log	6.9412	118	0.2940	0.0	0.2940
No log	7.0588	120	0.3391	0.0224	0.3391
No log	7.1765	122	0.3471	0.0224	0.3471
No log	7.2941	124	0.3232	0.0224	0.3232
No log	7.4118	126	0.2886	0.0	0.2886
No log	7.5294	128	0.2755	0.0	0.2755
No log	7.6471	130	0.2736	0.0	0.2736
No log	7.7647	132	0.2796	0.0	0.2796
No log	7.8824	134	0.2901	0.0	0.2901
No log	8.0	136	0.3054	0.0224	0.3054
No log	8.1176	138	0.3155	0.0224	0.3155
No log	8.2353	140	0.3163	0.0224	0.3163
No log	8.3529	142	0.3326	0.0224	0.3326
No log	8.4706	144	0.3464	0.0224	0.3464
No log	8.5882	146	0.3603	0.0123	0.3603
No log	8.7059	148	0.3747	-0.0057	0.3747
No log	8.8235	150	0.3703	0.0123	0.3703
No log	8.9412	152	0.3512	0.0224	0.3512
No log	9.0588	154	0.3262	0.0224	0.3262
No log	9.1765	156	0.3095	0.0224	0.3095
No log	9.2941	158	0.3054	0.0224	0.3054
No log	9.4118	160	0.3038	0.0224	0.3038
No log	9.5294	162	0.3034	0.0224	0.3034
No log	9.6471	164	0.3016	0.0224	0.3016
No log	9.7647	166	0.3019	0.0224	0.3019
No log	9.8824	168	0.3032	0.0224	0.3032
No log	10.0	170	0.3040	0.0224	0.3040

Framework versions

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