arabert_cross_relevance_task2_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: 0.6695
• Qwk: 0.0252
• Mse: 0.6695

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	0.3996	0.0319	0.3996
No log	0.2353	4	0.4069	0.0909	0.4069
No log	0.3529	6	0.3579	-0.0167	0.3579
No log	0.4706	8	0.3710	0.0	0.3710
No log	0.5882	10	0.3188	-0.0235	0.3188
No log	0.7059	12	0.3518	-0.1194	0.3518
No log	0.8235	14	0.3275	-0.0616	0.3275
No log	0.9412	16	0.3104	0.0	0.3104
No log	1.0588	18	0.3535	0.0	0.3535
No log	1.1765	20	0.3752	0.0	0.3752
No log	1.2941	22	0.3373	0.0	0.3373
No log	1.4118	24	0.3102	0.0	0.3102
No log	1.5294	26	0.2963	0.0	0.2963
No log	1.6471	28	0.2934	0.0	0.2934
No log	1.7647	30	0.2889	0.0	0.2889
No log	1.8824	32	0.2810	0.0	0.2810
No log	2.0	34	0.2866	-0.0235	0.2866
No log	2.1176	36	0.3061	0.0068	0.3061
No log	2.2353	38	0.3392	-0.0433	0.3392
No log	2.3529	40	0.3020	-0.0034	0.3020
No log	2.4706	42	0.2698	-0.0235	0.2698
No log	2.5882	44	0.2696	-0.0235	0.2696
No log	2.7059	46	0.2745	-0.0235	0.2745
No log	2.8235	48	0.2757	-0.0235	0.2757
No log	2.9412	50	0.3094	0.0	0.3094
No log	3.0588	52	0.3116	0.0	0.3116
No log	3.1765	54	0.2780	0.0	0.2780
No log	3.2941	56	0.2758	-0.0235	0.2758
No log	3.4118	58	0.2767	-0.0235	0.2767
No log	3.5294	60	0.2717	-0.0235	0.2717
No log	3.6471	62	0.2957	0.0	0.2957
No log	3.7647	64	0.3009	0.0	0.3009
No log	3.8824	66	0.2803	0.0	0.2803
No log	4.0	68	0.2754	-0.0235	0.2754
No log	4.1176	70	0.2758	-0.0235	0.2758
No log	4.2353	72	0.2948	-0.0235	0.2948
No log	4.3529	74	0.3207	0.0	0.3207
No log	4.4706	76	0.3204	0.0	0.3204
No log	4.5882	78	0.2864	-0.0235	0.2864
No log	4.7059	80	0.2756	-0.0235	0.2756
No log	4.8235	82	0.2778	-0.0235	0.2778
No log	4.9412	84	0.2766	-0.0235	0.2766
No log	5.0588	86	0.2988	-0.0235	0.2988
No log	5.1765	88	0.3889	0.0224	0.3889
No log	5.2941	90	0.4631	0.0123	0.4631
No log	5.4118	92	0.4091	0.0224	0.4091
No log	5.5294	94	0.3390	-0.0235	0.3390
No log	5.6471	96	0.3069	-0.0235	0.3069
No log	5.7647	98	0.3133	-0.0235	0.3133
No log	5.8824	100	0.3527	-0.0235	0.3527
No log	6.0	102	0.3823	-0.0235	0.3823
No log	6.1176	104	0.3926	-0.0096	0.3926
No log	6.2353	106	0.3681	-0.0235	0.3681
No log	6.3529	108	0.3415	-0.0235	0.3415
No log	6.4706	110	0.3362	-0.0235	0.3362
No log	6.5882	112	0.3928	-0.0323	0.3928
No log	6.7059	114	0.4304	-0.0093	0.4304
No log	6.8235	116	0.4880	0.0152	0.4880
No log	6.9412	118	0.5316	-0.0043	0.5316
No log	7.0588	120	0.5155	0.0090	0.5155
No log	7.1765	122	0.4655	0.0135	0.4655
No log	7.2941	124	0.4459	0.0029	0.4459
No log	7.4118	126	0.4302	-0.0180	0.4302
No log	7.5294	128	0.4389	-0.0260	0.4389
No log	7.6471	130	0.4613	-0.0025	0.4613
No log	7.7647	132	0.4740	0.0311	0.4740
No log	7.8824	134	0.4735	0.0311	0.4735
No log	8.0	136	0.4930	0.0088	0.4930
No log	8.1176	138	0.5329	0.0102	0.5329
No log	8.2353	140	0.5907	-0.0105	0.5907
No log	8.3529	142	0.6354	-0.0082	0.6354
No log	8.4706	144	0.6447	-0.0082	0.6447
No log	8.5882	146	0.6437	0.0032	0.6437
No log	8.7059	148	0.6301	-0.0201	0.6301
No log	8.8235	150	0.6096	-0.0201	0.6096
No log	8.9412	152	0.6030	-0.0201	0.6030
No log	9.0588	154	0.6322	-0.0082	0.6322
No log	9.1765	156	0.6675	0.0142	0.6675
No log	9.2941	158	0.7002	0.0252	0.7002
No log	9.4118	160	0.7136	0.0455	0.7136
No log	9.5294	162	0.7118	0.0455	0.7118
No log	9.6471	164	0.7001	0.0355	0.7001
No log	9.7647	166	0.6860	0.0252	0.6860
No log	9.8824	168	0.6738	0.0252	0.6738
No log	10.0	170	0.6695	0.0252	0.6695

Framework versions

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