ArabicNewSplits6_FineTuningAraBERT_run3_AugV5_k3_task5_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: 0.7357
• Qwk: 0.6549
• Mse: 0.7357
• Rmse: 0.8577

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	2.2520	0.0250	2.2520	1.5007
No log	0.3077	4	1.5061	0.1915	1.5061	1.2272
No log	0.4615	6	1.3795	0.1989	1.3795	1.1745
No log	0.6154	8	1.3783	0.2438	1.3783	1.1740
No log	0.7692	10	1.5196	0.3111	1.5196	1.2327
No log	0.9231	12	1.6249	0.3439	1.6249	1.2747
No log	1.0769	14	1.4826	0.3791	1.4826	1.2176
No log	1.2308	16	1.2894	0.3823	1.2894	1.1355
No log	1.3846	18	1.2640	0.3823	1.2640	1.1243
No log	1.5385	20	1.5182	0.3706	1.5182	1.2321
No log	1.6923	22	1.7873	0.2237	1.7873	1.3369
No log	1.8462	24	1.7017	0.2636	1.7017	1.3045
No log	2.0	26	1.3279	0.4333	1.3279	1.1523
No log	2.1538	28	1.0668	0.3165	1.0668	1.0329
No log	2.3077	30	1.0991	0.3483	1.0991	1.0484
No log	2.4615	32	1.0666	0.3347	1.0666	1.0328
No log	2.6154	34	1.0788	0.3809	1.0788	1.0386
No log	2.7692	36	1.1557	0.4210	1.1557	1.0750
No log	2.9231	38	1.1541	0.4308	1.1541	1.0743
No log	3.0769	40	1.1556	0.4588	1.1556	1.0750
No log	3.2308	42	1.0765	0.4530	1.0765	1.0375
No log	3.3846	44	1.0202	0.4171	1.0202	1.0100
No log	3.5385	46	0.9643	0.4663	0.9643	0.9820
No log	3.6923	48	0.9439	0.4731	0.9439	0.9716
No log	3.8462	50	0.9467	0.4484	0.9467	0.9730
No log	4.0	52	0.9799	0.4917	0.9799	0.9899
No log	4.1538	54	0.9807	0.5550	0.9807	0.9903
No log	4.3077	56	0.9045	0.5939	0.9045	0.9510
No log	4.4615	58	0.8461	0.5575	0.8461	0.9198
No log	4.6154	60	0.8105	0.5828	0.8105	0.9003
No log	4.7692	62	0.8119	0.6121	0.8119	0.9011
No log	4.9231	64	0.8026	0.6249	0.8026	0.8959
No log	5.0769	66	0.8056	0.6151	0.8056	0.8975
No log	5.2308	68	0.7975	0.6168	0.7975	0.8930
No log	5.3846	70	0.7892	0.6280	0.7892	0.8883
No log	5.5385	72	0.8620	0.5767	0.8620	0.9285
No log	5.6923	74	0.8903	0.5807	0.8903	0.9436
No log	5.8462	76	0.8730	0.5993	0.8730	0.9343
No log	6.0	78	0.8870	0.6013	0.8870	0.9418
No log	6.1538	80	0.9549	0.6056	0.9549	0.9772
No log	6.3077	82	0.9149	0.6165	0.9149	0.9565
No log	6.4615	84	0.7948	0.6383	0.7948	0.8915
No log	6.6154	86	0.7438	0.6627	0.7438	0.8625
No log	6.7692	88	0.7391	0.6795	0.7391	0.8597
No log	6.9231	90	0.7789	0.6386	0.7789	0.8826
No log	7.0769	92	0.9008	0.6369	0.9008	0.9491
No log	7.2308	94	0.9325	0.6343	0.9325	0.9657
No log	7.3846	96	0.8526	0.6334	0.8526	0.9234
No log	7.5385	98	0.7636	0.6496	0.7636	0.8738
No log	7.6923	100	0.7290	0.6818	0.7290	0.8538
No log	7.8462	102	0.7179	0.6793	0.7179	0.8473
No log	8.0	104	0.7130	0.6793	0.7130	0.8444
No log	8.1538	106	0.7115	0.7052	0.7115	0.8435
No log	8.3077	108	0.7192	0.6960	0.7192	0.8480
No log	8.4615	110	0.7389	0.6645	0.7389	0.8596
No log	8.6154	112	0.7520	0.6776	0.7520	0.8672
No log	8.7692	114	0.7562	0.6776	0.7562	0.8696
No log	8.9231	116	0.7449	0.6758	0.7449	0.8631
No log	9.0769	118	0.7416	0.6549	0.7416	0.8612
No log	9.2308	120	0.7416	0.6549	0.7416	0.8612
No log	9.3846	122	0.7312	0.6567	0.7312	0.8551
No log	9.5385	124	0.7295	0.6567	0.7295	0.8541
No log	9.6923	126	0.7295	0.6567	0.7295	0.8541
No log	9.8462	128	0.7327	0.6567	0.7327	0.8560
No log	10.0	130	0.7357	0.6549	0.7357	0.8577

Framework versions

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