ArabicNewSplits6_WithDuplicationsForScore5_FineTuningAraBERT_run2_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: 0.8679
• Qwk: 0.264
• Mse: 0.8679
• Rmse: 0.9316

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	3.0809	-0.0126	3.0809	1.7552
No log	0.3077	4	1.5858	-0.0327	1.5858	1.2593
No log	0.4615	6	1.0707	0.0038	1.0707	1.0348
No log	0.6154	8	0.6138	0.1605	0.6138	0.7835
No log	0.7692	10	0.5849	0.0	0.5849	0.7648
No log	0.9231	12	0.5909	0.0	0.5909	0.7687
No log	1.0769	14	0.6527	0.3455	0.6527	0.8079
No log	1.2308	16	0.8075	0.1392	0.8075	0.8986
No log	1.3846	18	0.7048	0.1765	0.7048	0.8395
No log	1.5385	20	0.6644	0.1828	0.6644	0.8151
No log	1.6923	22	0.5867	-0.0303	0.5867	0.7660
No log	1.8462	24	0.5860	0.0	0.5860	0.7655
No log	2.0	26	0.5985	-0.0303	0.5985	0.7736
No log	2.1538	28	0.8705	0.1416	0.8705	0.9330
No log	2.3077	30	1.6979	0.0258	1.6979	1.3030
No log	2.4615	32	1.1933	0.1111	1.1933	1.0924
No log	2.6154	34	0.6598	0.1813	0.6598	0.8123
No log	2.7692	36	0.6070	0.0145	0.6070	0.7791
No log	2.9231	38	0.6143	0.0	0.6143	0.7838
No log	3.0769	40	0.5967	0.0145	0.5967	0.7725
No log	3.2308	42	0.5955	-0.0370	0.5955	0.7717
No log	3.3846	44	0.5796	-0.0370	0.5796	0.7613
No log	3.5385	46	0.5740	0.2970	0.5740	0.7576
No log	3.6923	48	0.5572	0.2704	0.5572	0.7464
No log	3.8462	50	0.5548	0.2418	0.5548	0.7449
No log	4.0	52	0.5915	0.1899	0.5915	0.7691
No log	4.1538	54	0.6786	0.2821	0.6786	0.8238
No log	4.3077	56	0.7196	0.2390	0.7196	0.8483
No log	4.4615	58	0.7411	0.2618	0.7411	0.8609
No log	4.6154	60	0.7798	0.2941	0.7798	0.8831
No log	4.7692	62	0.7853	0.2922	0.7853	0.8862
No log	4.9231	64	0.7960	0.2569	0.7960	0.8922
No log	5.0769	66	0.7930	0.2000	0.7930	0.8905
No log	5.2308	68	0.8164	0.2593	0.8164	0.9035
No log	5.3846	70	0.8895	0.2126	0.8895	0.9432
No log	5.5385	72	0.6733	0.3488	0.6733	0.8206
No log	5.6923	74	0.5797	0.4112	0.5797	0.7614
No log	5.8462	76	0.7058	0.3818	0.7058	0.8401
No log	6.0	78	1.1104	0.1709	1.1104	1.0537
No log	6.1538	80	1.2836	0.0933	1.2836	1.1330
No log	6.3077	82	1.0089	0.2061	1.0089	1.0044
No log	6.4615	84	0.8262	0.3496	0.8262	0.9090
No log	6.6154	86	0.6483	0.4182	0.6483	0.8051
No log	6.7692	88	0.6546	0.4489	0.6546	0.8091
No log	6.9231	90	0.8422	0.3000	0.8422	0.9177
No log	7.0769	92	1.1845	0.2107	1.1845	1.0883
No log	7.2308	94	1.1575	0.2388	1.1575	1.0759
No log	7.3846	96	0.9914	0.2302	0.9914	0.9957
No log	7.5385	98	0.8034	0.3391	0.8034	0.8963
No log	7.6923	100	0.6999	0.3247	0.6999	0.8366
No log	7.8462	102	0.7128	0.3422	0.7128	0.8443
No log	8.0	104	0.8792	0.2314	0.8792	0.9376
No log	8.1538	106	0.9688	0.2124	0.9688	0.9843
No log	8.3077	108	0.9054	0.2381	0.9054	0.9515
No log	8.4615	110	0.8654	0.2389	0.8654	0.9302
No log	8.6154	112	0.7779	0.3684	0.7779	0.8820
No log	8.7692	114	0.7185	0.3363	0.7185	0.8477
No log	8.9231	116	0.7140	0.3363	0.7140	0.8450
No log	9.0769	118	0.7740	0.3684	0.7740	0.8798
No log	9.2308	120	0.8557	0.2569	0.8557	0.9251
No log	9.3846	122	0.8894	0.264	0.8894	0.9431
No log	9.5385	124	0.8908	0.264	0.8908	0.9438
No log	9.6923	126	0.8676	0.264	0.8676	0.9315
No log	9.8462	128	0.8641	0.2569	0.8641	0.9296
No log	10.0	130	0.8679	0.264	0.8679	0.9316

Framework versions

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