arabert_cross_relevance_task6_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: 0.4618
• Qwk: 0.0432
• Mse: 0.4618

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	0.1111	2	0.6215	0.0598	0.6215
No log	0.2222	4	0.2878	0.1667	0.2878
No log	0.3333	6	0.3256	0.125	0.3256
No log	0.4444	8	0.2842	0.0	0.2842
No log	0.5556	10	0.2870	0.0	0.2870
No log	0.6667	12	0.3049	0.0	0.3049
No log	0.7778	14	0.2824	0.0	0.2824
No log	0.8889	16	0.2859	0.0	0.2859
No log	1.0	18	0.2773	0.0	0.2773
No log	1.1111	20	0.2760	0.0	0.2760
No log	1.2222	22	0.2840	0.0	0.2840
No log	1.3333	24	0.2933	0.0	0.2933
No log	1.4444	26	0.2910	0.0	0.2910
No log	1.5556	28	0.2848	0.0	0.2848
No log	1.6667	30	0.2936	0.0	0.2936
No log	1.7778	32	0.2964	0.0	0.2964
No log	1.8889	34	0.3217	0.0	0.3217
No log	2.0	36	0.3074	0.0	0.3074
No log	2.1111	38	0.2822	0.0	0.2822
No log	2.2222	40	0.2781	0.0	0.2781
No log	2.3333	42	0.2711	0.0	0.2711
No log	2.4444	44	0.2765	0.0	0.2765
No log	2.5556	46	0.2987	0.0	0.2987
No log	2.6667	48	0.3302	0.0	0.3302
No log	2.7778	50	0.3306	0.0	0.3306
No log	2.8889	52	0.3144	0.0	0.3144
No log	3.0	54	0.2775	0.0	0.2775
No log	3.1111	56	0.2698	0.0	0.2698
No log	3.2222	58	0.2930	0.0	0.2930
No log	3.3333	60	0.3368	0.0	0.3368
No log	3.4444	62	0.3202	0.0	0.3202
No log	3.5556	64	0.2917	0.0	0.2917
No log	3.6667	66	0.2915	0.0	0.2915
No log	3.7778	68	0.3126	0.0	0.3126
No log	3.8889	70	0.3240	0.0	0.3240
No log	4.0	72	0.3470	0.0	0.3470
No log	4.1111	74	0.3430	0.0	0.3430
No log	4.2222	76	0.3647	0.0	0.3647
No log	4.3333	78	0.3957	0.0	0.3957
No log	4.4444	80	0.3840	0.0	0.3840
No log	4.5556	82	0.3521	0.0	0.3521
No log	4.6667	84	0.3619	0.0	0.3619
No log	4.7778	86	0.3569	0.0	0.3569
No log	4.8889	88	0.3310	0.0	0.3310
No log	5.0	90	0.3382	0.0	0.3382
No log	5.1111	92	0.3534	0.0	0.3534
No log	5.2222	94	0.3516	0.0	0.3516
No log	5.3333	96	0.3252	0.0	0.3252
No log	5.4444	98	0.3135	0.0	0.3135
No log	5.5556	100	0.3412	0.0	0.3412
No log	5.6667	102	0.4012	0.0	0.4012
No log	5.7778	104	0.4263	0.0224	0.4263
No log	5.8889	106	0.3916	0.0	0.3916
No log	6.0	108	0.3330	0.0	0.3330
No log	6.1111	110	0.2996	0.0	0.2996
No log	6.2222	112	0.3013	0.0	0.3013
No log	6.3333	114	0.3241	0.0	0.3241
No log	6.4444	116	0.3864	0.0	0.3864
No log	6.5556	118	0.4424	0.0	0.4424
No log	6.6667	120	0.4730	0.0	0.4730
No log	6.7778	122	0.4565	0.0	0.4565
No log	6.8889	124	0.4114	0.0	0.4114
No log	7.0	126	0.3989	0.0	0.3989
No log	7.1111	128	0.3922	0.0	0.3922
No log	7.2222	130	0.3994	0.0	0.3994
No log	7.3333	132	0.4033	0.0	0.4033
No log	7.4444	134	0.3909	0.0	0.3909
No log	7.5556	136	0.3949	0.0	0.3949
No log	7.6667	138	0.3949	0.0	0.3949
No log	7.7778	140	0.4170	0.0	0.4170
No log	7.8889	142	0.4493	0.0224	0.4493
No log	8.0	144	0.4891	0.0432	0.4891
No log	8.1111	146	0.4976	0.0230	0.4976
No log	8.2222	148	0.4661	0.0432	0.4661
No log	8.3333	150	0.4184	0.0224	0.4184
No log	8.4444	152	0.3950	0.0	0.3950
No log	8.5556	154	0.3931	0.0	0.3931
No log	8.6667	156	0.4033	0.0224	0.4033
No log	8.7778	158	0.4222	0.0224	0.4222
No log	8.8889	160	0.4526	0.0432	0.4526
No log	9.0	162	0.4915	0.0432	0.4915
No log	9.1111	164	0.5166	0.0054	0.5166
No log	9.2222	166	0.5171	-0.0026	0.5171
No log	9.3333	168	0.5083	0.0139	0.5083
No log	9.4444	170	0.4931	0.0327	0.4931
No log	9.5556	172	0.4776	0.0432	0.4776
No log	9.6667	174	0.4695	0.0432	0.4695
No log	9.7778	176	0.4633	0.0432	0.4633
No log	9.8889	178	0.4623	0.0432	0.4623
No log	10.0	180	0.4618	0.0432	0.4618

Framework versions

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