ArabicNewSplits6_WithDuplicationsForScore5_FineTuningAraBERT_run1_AugV5_k2_task3_organization

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: 0.7854
• Qwk: 0.2212
• Mse: 0.7854
• Rmse: 0.8862

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	0.1538	2	3.3768	-0.0149	3.3768	1.8376
No log	0.3077	4	1.6870	-0.0370	1.6870	1.2988
No log	0.4615	6	0.9515	0.0431	0.9515	0.9754
No log	0.6154	8	1.2418	0.1235	1.2418	1.1144
No log	0.7692	10	0.5584	0.2109	0.5584	0.7472
No log	0.9231	12	0.6467	0.0080	0.6467	0.8042
No log	1.0769	14	0.5480	-0.0081	0.5480	0.7403
No log	1.2308	16	0.5609	0.1467	0.5609	0.7489
No log	1.3846	18	0.5740	0.2000	0.5740	0.7576
No log	1.5385	20	0.7608	0.1841	0.7608	0.8722
No log	1.6923	22	1.5922	0.0602	1.5922	1.2618
No log	1.8462	24	1.0920	0.0539	1.0920	1.0450
No log	2.0	26	0.7193	0.0504	0.7193	0.8481
No log	2.1538	28	0.9516	0.0588	0.9516	0.9755
No log	2.3077	30	0.8720	0.1443	0.8720	0.9338
No log	2.4615	32	0.6561	0.0952	0.6561	0.8100
No log	2.6154	34	1.3252	0.0949	1.3252	1.1512
No log	2.7692	36	1.4020	0.1037	1.4020	1.1840
No log	2.9231	38	0.8484	0.0901	0.8484	0.9211
No log	3.0769	40	0.6063	0.2727	0.6063	0.7787
No log	3.2308	42	0.7571	0.1818	0.7571	0.8701
No log	3.3846	44	0.6869	0.1781	0.6869	0.8288
No log	3.5385	46	0.6329	0.1243	0.6329	0.7956
No log	3.6923	48	0.9547	0.0769	0.9547	0.9771
No log	3.8462	50	1.0709	0.1145	1.0709	1.0348
No log	4.0	52	0.8680	0.2000	0.8680	0.9317
No log	4.1538	54	0.6352	0.2340	0.6352	0.7970
No log	4.3077	56	0.6073	0.2273	0.6073	0.7793
No log	4.4615	58	0.5994	0.2410	0.5994	0.7742
No log	4.6154	60	0.6570	0.2593	0.6570	0.8106
No log	4.7692	62	0.6430	0.1915	0.6430	0.8019
No log	4.9231	64	0.6551	0.1915	0.6551	0.8094
No log	5.0769	66	0.8920	0.2074	0.8920	0.9445
No log	5.2308	68	1.0565	0.1746	1.0565	1.0278
No log	5.3846	70	0.8772	0.1858	0.8772	0.9366
No log	5.5385	72	0.7255	0.1759	0.7255	0.8518
No log	5.6923	74	0.7127	0.2709	0.7127	0.8442
No log	5.8462	76	0.7194	0.1841	0.7194	0.8482
No log	6.0	78	0.8399	0.1790	0.8399	0.9165
No log	6.1538	80	0.8917	0.1930	0.8917	0.9443
No log	6.3077	82	0.8389	0.1858	0.8389	0.9159
No log	6.4615	84	0.7383	0.2222	0.7383	0.8593
No log	6.6154	86	0.7182	0.2897	0.7182	0.8474
No log	6.7692	88	0.7489	0.2744	0.7489	0.8654
No log	6.9231	90	0.8302	0.1864	0.8302	0.9112
No log	7.0769	92	0.8890	0.1864	0.8890	0.9428
No log	7.2308	94	0.8688	0.1864	0.8688	0.9321
No log	7.3846	96	0.8869	0.2381	0.8869	0.9418
No log	7.5385	98	0.7685	0.3247	0.7685	0.8766
No log	7.6923	100	0.7031	0.1855	0.7031	0.8385
No log	7.8462	102	0.6972	0.2500	0.6972	0.8350
No log	8.0	104	0.6948	0.1927	0.6948	0.8335
No log	8.1538	106	0.7496	0.3067	0.7496	0.8658
No log	8.3077	108	0.9063	0.2074	0.9063	0.9520
No log	8.4615	110	0.9971	0.1938	0.9971	0.9986
No log	8.6154	112	0.9544	0.2072	0.9544	0.9769
No log	8.7692	114	0.8399	0.2075	0.8399	0.9165
No log	8.9231	116	0.7535	0.2381	0.7535	0.8680
No log	9.0769	118	0.7114	0.3488	0.7114	0.8435
No log	9.2308	120	0.7125	0.3143	0.7125	0.8441
No log	9.3846	122	0.7276	0.3143	0.7276	0.8530
No log	9.5385	124	0.7511	0.3208	0.7511	0.8666
No log	9.6923	126	0.7680	0.2759	0.7680	0.8763
No log	9.8462	128	0.7777	0.2838	0.7777	0.8819
No log	10.0	130	0.7854	0.2212	0.7854	0.8862

Framework versions

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