Arabic_FineTuningAraBERT_AugV0_k3_task3_organization_fold0

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: 0.9297
• Qwk: 0.0
• Mse: 0.9297
• Rmse: 0.9642

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: 8
• eval_batch_size: 8
• 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	Rmse
No log	0.1429	2	4.5347	0.0	4.5347	2.1295
No log	0.2857	4	2.6030	-0.0722	2.6030	1.6134
No log	0.4286	6	1.1663	0.0833	1.1663	1.0800
No log	0.5714	8	0.8668	0.0	0.8668	0.9310
No log	0.7143	10	0.8909	0.0	0.8909	0.9439
No log	0.8571	12	0.8354	0.0	0.8354	0.9140
No log	1.0	14	0.9813	0.0	0.9813	0.9906
No log	1.1429	16	1.2730	-0.2073	1.2730	1.1283
No log	1.2857	18	1.0339	0.0	1.0339	1.0168
No log	1.4286	20	0.8617	0.0	0.8617	0.9283
No log	1.5714	22	0.8184	0.0	0.8184	0.9047
No log	1.7143	24	0.8540	0.0	0.8540	0.9241
No log	1.8571	26	0.8464	0.0	0.8464	0.9200
No log	2.0	28	0.9249	0.0	0.9249	0.9617
No log	2.1429	30	0.9747	0.0	0.9747	0.9873
No log	2.2857	32	0.9842	0.0	0.9842	0.9921
No log	2.4286	34	0.9170	0.0	0.9170	0.9576
No log	2.5714	36	0.9076	0.0	0.9076	0.9527
No log	2.7143	38	0.9634	0.0	0.9634	0.9815
No log	2.8571	40	0.9098	0.0	0.9098	0.9539
No log	3.0	42	0.8946	0.0	0.8946	0.9458
No log	3.1429	44	0.9477	0.0	0.9477	0.9735
No log	3.2857	46	0.9997	-0.1786	0.9997	0.9999
No log	3.4286	48	1.0116	-0.1786	1.0116	1.0058
No log	3.5714	50	0.9601	0.0	0.9601	0.9798
No log	3.7143	52	0.9033	0.0	0.9033	0.9504
No log	3.8571	54	0.9188	0.0	0.9188	0.9585
No log	4.0	56	0.9413	0.0	0.9413	0.9702
No log	4.1429	58	0.9330	0.0	0.9330	0.9659
No log	4.2857	60	1.0394	-0.3841	1.0394	1.0195
No log	4.4286	62	1.0055	-0.375	1.0055	1.0027
No log	4.5714	64	0.9360	0.0	0.9360	0.9675
No log	4.7143	66	0.9226	0.0	0.9226	0.9605
No log	4.8571	68	0.9119	0.0	0.9119	0.9550
No log	5.0	70	0.9285	0.0	0.9285	0.9636
No log	5.1429	72	1.0412	-0.1440	1.0412	1.0204
No log	5.2857	74	1.1266	-0.2073	1.1266	1.0614
No log	5.4286	76	1.0449	-0.1440	1.0449	1.0222
No log	5.5714	78	0.9719	0.0	0.9719	0.9859
No log	5.7143	80	0.9105	0.0	0.9105	0.9542
No log	5.8571	82	0.9039	0.0	0.9039	0.9507
No log	6.0	84	0.8996	0.0	0.8996	0.9485
No log	6.1429	86	0.9236	0.0	0.9236	0.9611
No log	6.2857	88	0.9518	0.0	0.9518	0.9756
No log	6.4286	90	0.9742	0.2080	0.9742	0.9870
No log	6.5714	92	0.9879	-0.1159	0.9879	0.9939
No log	6.7143	94	0.9745	0.2080	0.9745	0.9871
No log	6.8571	96	0.9271	0.0	0.9271	0.9629
No log	7.0	98	0.8800	0.1852	0.8800	0.9381
No log	7.1429	100	0.8898	0.1852	0.8898	0.9433
No log	7.2857	102	0.8958	0.1852	0.8958	0.9465
No log	7.4286	104	0.8901	0.1852	0.8901	0.9435
No log	7.5714	106	0.8958	0.0	0.8958	0.9465
No log	7.7143	108	0.9220	0.0	0.9220	0.9602
No log	7.8571	110	0.9667	0.2143	0.9667	0.9832
No log	8.0	112	1.0376	0.0530	1.0376	1.0186
No log	8.1429	114	1.0667	-0.0565	1.0667	1.0328
No log	8.2857	116	1.0296	0.0530	1.0296	1.0147
No log	8.4286	118	0.9699	0.2143	0.9699	0.9849
No log	8.5714	120	0.9232	0.0	0.9232	0.9608
No log	8.7143	122	0.9074	0.1852	0.9074	0.9526
No log	8.8571	124	0.9049	0.1852	0.9049	0.9513
No log	9.0	126	0.9022	0.1852	0.9022	0.9498
No log	9.1429	128	0.9019	0.1852	0.9019	0.9497
No log	9.2857	130	0.9042	0.1852	0.9042	0.9509
No log	9.4286	132	0.9093	0.1852	0.9093	0.9536
No log	9.5714	134	0.9162	0.1852	0.9162	0.9572
No log	9.7143	136	0.9233	0.0	0.9233	0.9609
No log	9.8571	138	0.9276	0.0	0.9276	0.9631
No log	10.0	140	0.9297	0.0	0.9297	0.9642

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu118
• Datasets 2.21.0
• Tokenizers 0.19.1