arabert_cross_relevance_task7_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: 0.4851
• Qwk: 0.0152
• Mse: 0.4851

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	0.4468	0.0241	0.4468
No log	0.2353	4	0.5293	0.1111	0.5293
No log	0.3529	6	0.4293	0.0657	0.4293
No log	0.4706	8	0.2782	0.0	0.2782
No log	0.5882	10	0.2775	0.0	0.2775
No log	0.7059	12	0.3091	0.0172	0.3091
No log	0.8235	14	0.3313	-0.0714	0.3313
No log	0.9412	16	0.3259	0.0101	0.3259
No log	1.0588	18	0.3382	0.0	0.3382
No log	1.1765	20	0.3574	-0.0135	0.3574
No log	1.2941	22	0.3580	-0.0274	0.3580
No log	1.4118	24	0.4301	0.0	0.4301
No log	1.5294	26	0.4619	0.0	0.4619
No log	1.6471	28	0.3492	0.0	0.3492
No log	1.7647	30	0.3168	-0.0417	0.3168
No log	1.8824	32	0.3350	-0.0496	0.3350
No log	2.0	34	0.2965	0.0203	0.2965
No log	2.1176	36	0.2856	0.0	0.2856
No log	2.2353	38	0.2924	0.0	0.2924
No log	2.3529	40	0.2917	0.0	0.2917
No log	2.4706	42	0.2928	0.0	0.2928
No log	2.5882	44	0.2979	-0.0616	0.2979
No log	2.7059	46	0.3126	0.0	0.3126
No log	2.8235	48	0.3738	0.0	0.3738
No log	2.9412	50	0.3664	0.0	0.3664
No log	3.0588	52	0.3323	0.0	0.3323
No log	3.1765	54	0.3388	0.0	0.3388
No log	3.2941	56	0.3765	0.0	0.3765
No log	3.4118	58	0.3896	0.0	0.3896
No log	3.5294	60	0.3597	0.0	0.3597
No log	3.6471	62	0.3248	0.0	0.3248
No log	3.7647	64	0.3130	-0.0473	0.3130
No log	3.8824	66	0.3039	0.0	0.3039
No log	4.0	68	0.3036	-0.0235	0.3036
No log	4.1176	70	0.3005	0.0	0.3005
No log	4.2353	72	0.3729	0.0	0.3729
No log	4.3529	74	0.4274	0.0	0.4274
No log	4.4706	76	0.4149	0.0	0.4149
No log	4.5882	78	0.3636	0.0	0.3636
No log	4.7059	80	0.3435	0.0	0.3435
No log	4.8235	82	0.3605	0.0	0.3605
No log	4.9412	84	0.4118	0.0	0.4118
No log	5.0588	86	0.4373	0.0	0.4373
No log	5.1765	88	0.3877	0.0	0.3877
No log	5.2941	90	0.3152	0.0	0.3152
No log	5.4118	92	0.3033	0.0	0.3033
No log	5.5294	94	0.3085	0.0	0.3085
No log	5.6471	96	0.3451	0.0	0.3451
No log	5.7647	98	0.4554	0.0	0.4554
No log	5.8824	100	0.5290	0.0230	0.5290
No log	6.0	102	0.5208	0.0230	0.5208
No log	6.1176	104	0.4384	0.0	0.4384
No log	6.2353	106	0.3592	0.0	0.3592
No log	6.3529	108	0.3270	-0.0235	0.3270
No log	6.4706	110	0.3341	-0.0235	0.3341
No log	6.5882	112	0.3501	0.0	0.3501
No log	6.7059	114	0.4010	0.0224	0.4010
No log	6.8235	116	0.4289	0.0224	0.4289
No log	6.9412	118	0.4617	0.0432	0.4617
No log	7.0588	120	0.4638	0.0432	0.4638
No log	7.1765	122	0.4444	0.0432	0.4444
No log	7.2941	124	0.4625	0.0120	0.4625
No log	7.4118	126	0.5041	0.0139	0.5041
No log	7.5294	128	0.5111	0.0139	0.5111
No log	7.6471	130	0.5193	-0.0026	0.5193
No log	7.7647	132	0.4942	0.0139	0.4942
No log	7.8824	134	0.4487	0.0432	0.4487
No log	8.0	136	0.3941	0.0	0.3941
No log	8.1176	138	0.3656	0.0	0.3656
No log	8.2353	140	0.3670	0.0	0.3670
No log	8.3529	142	0.3813	0.0	0.3813
No log	8.4706	144	0.4173	0.0432	0.4173
No log	8.5882	146	0.4510	0.0230	0.4510
No log	8.7059	148	0.4637	0.0230	0.4637
No log	8.8235	150	0.4865	-0.0026	0.4865
No log	8.9412	152	0.5054	-0.0101	0.5054
No log	9.0588	154	0.5000	0.0074	0.5000
No log	9.1765	156	0.4847	-0.0026	0.4847
No log	9.2941	158	0.4728	-0.0026	0.4728
No log	9.4118	160	0.4652	0.0054	0.4652
No log	9.5294	162	0.4641	0.0054	0.4641
No log	9.6471	164	0.4713	-0.0026	0.4713
No log	9.7647	166	0.4801	0.0152	0.4801
No log	9.8824	168	0.4839	0.0152	0.4839
No log	10.0	170	0.4851	0.0152	0.4851

Framework versions

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