arabert_cross_relevance_task1_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: 0.4917
• Qwk: -0.0345
• Mse: 0.4917

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	0.5402	0.0504	0.5402
No log	0.2353	4	0.5522	0.1129	0.5522
No log	0.3529	6	0.4760	0.1129	0.4760
No log	0.4706	8	0.2772	0.0	0.2772
No log	0.5882	10	0.2693	0.0	0.2693
No log	0.7059	12	0.2659	-0.0235	0.2659
No log	0.8235	14	0.3218	-0.0448	0.3218
No log	0.9412	16	0.3400	-0.1260	0.3400
No log	1.0588	18	0.3031	-0.0764	0.3031
No log	1.1765	20	0.3006	-0.0235	0.3006
No log	1.2941	22	0.3269	0.0	0.3269
No log	1.4118	24	0.3005	0.0	0.3005
No log	1.5294	26	0.2883	-0.0374	0.2883
No log	1.6471	28	0.3040	-0.0764	0.3040
No log	1.7647	30	0.2964	-0.0374	0.2964
No log	1.8824	32	0.2940	-0.0235	0.2940
No log	2.0	34	0.2905	-0.0473	0.2905
No log	2.1176	36	0.2906	-0.0473	0.2906
No log	2.2353	38	0.2973	0.0	0.2973
No log	2.3529	40	0.3635	0.0	0.3635
No log	2.4706	42	0.4742	-0.0268	0.4742
No log	2.5882	44	0.4561	-0.0185	0.4561
No log	2.7059	46	0.3327	0.0	0.3327
No log	2.8235	48	0.2840	0.0	0.2840
No log	2.9412	50	0.2878	-0.0235	0.2878
No log	3.0588	52	0.2902	-0.0235	0.2902
No log	3.1765	54	0.2821	-0.0473	0.2821
No log	3.2941	56	0.2813	0.0	0.2813
No log	3.4118	58	0.3125	0.0	0.3125
No log	3.5294	60	0.3580	0.0	0.3580
No log	3.6471	62	0.3412	0.0	0.3412
No log	3.7647	64	0.3136	0.0	0.3136
No log	3.8824	66	0.2971	-0.0235	0.2971
No log	4.0	68	0.2918	-0.0135	0.2918
No log	4.1176	70	0.2894	-0.0235	0.2894
No log	4.2353	72	0.2973	0.0	0.2973
No log	4.3529	74	0.3060	0.0	0.3060
No log	4.4706	76	0.3066	0.0	0.3066
No log	4.5882	78	0.3086	0.0	0.3086
No log	4.7059	80	0.3207	0.0	0.3207
No log	4.8235	82	0.3228	0.0	0.3228
No log	4.9412	84	0.3232	0.0	0.3232
No log	5.0588	86	0.3306	0.0	0.3306
No log	5.1765	88	0.3322	0.0	0.3322
No log	5.2941	90	0.3595	0.0	0.3595
No log	5.4118	92	0.3659	0.0	0.3659
No log	5.5294	94	0.3968	0.0	0.3968
No log	5.6471	96	0.4325	-0.0185	0.4325
No log	5.7647	98	0.4240	-0.0185	0.4240
No log	5.8824	100	0.3886	-0.0096	0.3886
No log	6.0	102	0.3688	-0.0096	0.3688
No log	6.1176	104	0.3669	-0.0096	0.3669
No log	6.2353	106	0.3824	-0.0096	0.3824
No log	6.3529	108	0.3765	-0.0323	0.3765
No log	6.4706	110	0.3534	-0.0235	0.3534
No log	6.5882	112	0.3724	-0.0235	0.3724
No log	6.7059	114	0.4165	-0.0323	0.4165
No log	6.8235	116	0.4596	-0.0533	0.4596
No log	6.9412	118	0.4708	-0.0591	0.4708
No log	7.0588	120	0.4502	-0.0405	0.4502
No log	7.1765	122	0.4102	-0.0323	0.4102
No log	7.2941	124	0.4035	-0.0323	0.4035
No log	7.4118	126	0.4253	-0.0323	0.4253
No log	7.5294	128	0.4225	-0.0323	0.4225
No log	7.6471	130	0.4086	-0.0323	0.4086
No log	7.7647	132	0.4280	-0.0260	0.4280
No log	7.8824	134	0.4752	-0.0233	0.4752
No log	8.0	136	0.4963	-0.0193	0.4963
No log	8.1176	138	0.5295	-0.0104	0.5295
No log	8.2353	140	0.5401	-0.0104	0.5401
No log	8.3529	142	0.5096	-0.0193	0.5096
No log	8.4706	144	0.4677	-0.0279	0.4677
No log	8.5882	146	0.4416	-0.0135	0.4416
No log	8.7059	148	0.4452	-0.0135	0.4452
No log	8.8235	150	0.4725	-0.0279	0.4725
No log	8.9412	152	0.4991	-0.0233	0.4991
No log	9.0588	154	0.5151	-0.0251	0.5151
No log	9.1765	156	0.5160	-0.0251	0.5160
No log	9.2941	158	0.5242	-0.0251	0.5242
No log	9.4118	160	0.5189	-0.0251	0.5189
No log	9.5294	162	0.5080	-0.0193	0.5080
No log	9.6471	164	0.4972	-0.0233	0.4972
No log	9.7647	166	0.4929	-0.0167	0.4929
No log	9.8824	168	0.4916	-0.0345	0.4916
No log	10.0	170	0.4917	-0.0345	0.4917

Framework versions

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