arabert_cross_relevance_task6_fold2

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: 0.2957
• Qwk: 0.0
• Mse: 0.2962

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	0.4557	0.0551	0.4551
No log	0.2353	4	0.7639	0.1621	0.7645
No log	0.3529	6	0.6785	0.0828	0.6793
No log	0.4706	8	0.2858	0.0	0.2858
No log	0.5882	10	0.3157	0.0	0.3153
No log	0.7059	12	0.2872	0.0	0.2872
No log	0.8235	14	0.3422	0.0491	0.3427
No log	0.9412	16	0.4804	0.1093	0.4812
No log	1.0588	18	0.5008	0.0622	0.5017
No log	1.1765	20	0.4329	0.1899	0.4337
No log	1.2941	22	0.3437	-0.0329	0.3443
No log	1.4118	24	0.3016	0.0	0.3020
No log	1.5294	26	0.3038	0.0	0.3043
No log	1.6471	28	0.3038	0.0	0.3043
No log	1.7647	30	0.3490	0.0491	0.3496
No log	1.8824	32	0.3388	0.0368	0.3394
No log	2.0	34	0.3023	0.0	0.3028
No log	2.1176	36	0.2791	0.0	0.2795
No log	2.2353	38	0.2726	0.0	0.2729
No log	2.3529	40	0.2754	0.0	0.2758
No log	2.4706	42	0.2899	0.0	0.2904
No log	2.5882	44	0.2963	0.0	0.2968
No log	2.7059	46	0.3008	0.0	0.3013
No log	2.8235	48	0.3022	0.0	0.3027
No log	2.9412	50	0.3239	-0.0164	0.3245
No log	3.0588	52	0.3244	0.0	0.3250
No log	3.1765	54	0.3271	0.0122	0.3277
No log	3.2941	56	0.3066	0.0	0.3070
No log	3.4118	58	0.2914	0.0	0.2918
No log	3.5294	60	0.2879	0.0	0.2883
No log	3.6471	62	0.3009	0.0122	0.3013
No log	3.7647	64	0.3097	-0.0208	0.3102
No log	3.8824	66	0.2902	0.0	0.2906
No log	4.0	68	0.2854	0.0	0.2857
No log	4.1176	70	0.3075	-0.0042	0.3080
No log	4.2353	72	0.3194	0.0080	0.3199
No log	4.3529	74	0.3140	0.0	0.3145
No log	4.4706	76	0.3268	0.0245	0.3273
No log	4.5882	78	0.3408	0.0491	0.3413
No log	4.7059	80	0.3249	0.0122	0.3254
No log	4.8235	82	0.3217	0.0122	0.3222
No log	4.9412	84	0.3068	0.0	0.3073
No log	5.0588	86	0.2957	0.0	0.2961
No log	5.1765	88	0.2935	0.0	0.2939
No log	5.2941	90	0.2907	0.0	0.2911
No log	5.4118	92	0.2928	0.0	0.2932
No log	5.5294	94	0.3137	0.0245	0.3143
No log	5.6471	96	0.3485	0.0706	0.3491
No log	5.7647	98	0.3569	0.0961	0.3576
No log	5.8824	100	0.3355	0.0450	0.3361
No log	6.0	102	0.3087	0.0122	0.3093
No log	6.1176	104	0.2963	0.0	0.2967
No log	6.2353	106	0.2984	0.0	0.2989
No log	6.3529	108	0.3049	0.0122	0.3055
No log	6.4706	110	0.2955	0.0	0.2959
No log	6.5882	112	0.2913	0.0	0.2918
No log	6.7059	114	0.2962	0.0	0.2966
No log	6.8235	116	0.2953	0.0	0.2958
No log	6.9412	118	0.2980	0.0	0.2985
No log	7.0588	120	0.2992	0.0	0.2997
No log	7.1765	122	0.3007	0.0	0.3012
No log	7.2941	124	0.3091	0.0368	0.3097
No log	7.4118	126	0.3122	0.0368	0.3128
No log	7.5294	128	0.3042	0.0368	0.3048
No log	7.6471	130	0.2990	0.0	0.2995
No log	7.7647	132	0.2979	0.0	0.2984
No log	7.8824	134	0.2985	0.0	0.2990
No log	8.0	136	0.3036	0.0245	0.3042
No log	8.1176	138	0.3069	0.0245	0.3074
No log	8.2353	140	0.3063	0.0245	0.3068
No log	8.3529	142	0.3012	0.0122	0.3017
No log	8.4706	144	0.2968	0.0	0.2973
No log	8.5882	146	0.2951	0.0	0.2956
No log	8.7059	148	0.2938	0.0	0.2942
No log	8.8235	150	0.2931	0.0	0.2936
No log	8.9412	152	0.2931	0.0	0.2935
No log	9.0588	154	0.2943	0.0	0.2948
No log	9.1765	156	0.2964	0.0	0.2969
No log	9.2941	158	0.2975	0.0122	0.2980
No log	9.4118	160	0.2970	0.0	0.2975
No log	9.5294	162	0.2962	0.0	0.2967
No log	9.6471	164	0.2957	0.0	0.2962
No log	9.7647	166	0.2956	0.0	0.2960
No log	9.8824	168	0.2955	0.0	0.2960
No log	10.0	170	0.2957	0.0	0.2962

Framework versions

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