ArabicNewSplits6_FineTuningAraBERT_run2_AugV5_k2_task2_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: 0.7333
• Qwk: 0.5405
• Mse: 0.7333
• Rmse: 0.8563

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	3.8411	-0.0075	3.8411	1.9599
No log	0.3077	4	2.0204	0.0682	2.0204	1.4214
No log	0.4615	6	1.2544	0.0951	1.2544	1.1200
No log	0.6154	8	0.9788	-0.0115	0.9788	0.9893
No log	0.7692	10	0.8849	0.1077	0.8849	0.9407
No log	0.9231	12	0.7830	0.0698	0.7830	0.8849
No log	1.0769	14	0.7069	0.1809	0.7069	0.8408
No log	1.2308	16	0.7165	0.1786	0.7165	0.8464
No log	1.3846	18	0.9541	0.1709	0.9541	0.9768
No log	1.5385	20	1.1009	0.1160	1.1009	1.0493
No log	1.6923	22	1.0328	0.1408	1.0328	1.0163
No log	1.8462	24	0.8405	0.2545	0.8405	0.9168
No log	2.0	26	0.6651	0.3000	0.6651	0.8156
No log	2.1538	28	0.6104	0.4154	0.6104	0.7813
No log	2.3077	30	0.6237	0.3761	0.6237	0.7898
No log	2.4615	32	0.6557	0.2636	0.6557	0.8097
No log	2.6154	34	0.6597	0.2636	0.6597	0.8122
No log	2.7692	36	0.7236	0.3266	0.7236	0.8507
No log	2.9231	38	0.8948	0.2975	0.8948	0.9459
No log	3.0769	40	0.9210	0.3157	0.9210	0.9597
No log	3.2308	42	0.7093	0.4461	0.7093	0.8422
No log	3.3846	44	0.6529	0.5408	0.6529	0.8080
No log	3.5385	46	0.7139	0.4840	0.7139	0.8449
No log	3.6923	48	0.7797	0.4169	0.7797	0.8830
No log	3.8462	50	0.7339	0.3581	0.7339	0.8567
No log	4.0	52	0.6078	0.4499	0.6078	0.7796
No log	4.1538	54	0.5813	0.5064	0.5813	0.7625
No log	4.3077	56	0.8562	0.4117	0.8562	0.9253
No log	4.4615	58	1.3180	0.2792	1.3180	1.1480
No log	4.6154	60	1.4946	0.2529	1.4946	1.2225
No log	4.7692	62	1.3781	0.2669	1.3781	1.1739
No log	4.9231	64	1.0734	0.3488	1.0734	1.0360
No log	5.0769	66	0.7788	0.4388	0.7788	0.8825
No log	5.2308	68	0.6144	0.4295	0.6144	0.7838
No log	5.3846	70	0.5476	0.4824	0.5476	0.7400
No log	5.5385	72	0.5565	0.4871	0.5565	0.7460
No log	5.6923	74	0.5776	0.4785	0.5776	0.7600
No log	5.8462	76	0.6050	0.4120	0.6050	0.7778
No log	6.0	78	0.6439	0.4246	0.6439	0.8024
No log	6.1538	80	0.6866	0.4359	0.6866	0.8286
No log	6.3077	82	0.7575	0.4710	0.7575	0.8703
No log	6.4615	84	0.7875	0.4460	0.7875	0.8874
No log	6.6154	86	0.7572	0.4982	0.7572	0.8702
No log	6.7692	88	0.6862	0.4871	0.6862	0.8284
No log	6.9231	90	0.6544	0.4756	0.6544	0.8090
No log	7.0769	92	0.6447	0.4750	0.6447	0.8029
No log	7.2308	94	0.6758	0.4750	0.6758	0.8221
No log	7.3846	96	0.7186	0.5276	0.7186	0.8477
No log	7.5385	98	0.7640	0.5089	0.7640	0.8741
No log	7.6923	100	0.7901	0.5104	0.7901	0.8889
No log	7.8462	102	0.7761	0.5246	0.7761	0.8809
No log	8.0	104	0.7393	0.5350	0.7393	0.8598
No log	8.1538	106	0.7308	0.5516	0.7308	0.8548
No log	8.3077	108	0.7179	0.5558	0.7179	0.8473
No log	8.4615	110	0.7194	0.5570	0.7194	0.8481
No log	8.6154	112	0.7161	0.5516	0.7161	0.8462
No log	8.7692	114	0.7135	0.5503	0.7135	0.8447
No log	8.9231	116	0.7075	0.5317	0.7075	0.8411
No log	9.0769	118	0.7069	0.5450	0.7069	0.8408
No log	9.2308	120	0.7109	0.5516	0.7109	0.8432
No log	9.3846	122	0.7196	0.5609	0.7196	0.8483
No log	9.5385	124	0.7255	0.5405	0.7255	0.8517
No log	9.6923	126	0.7298	0.5405	0.7298	0.8543
No log	9.8462	128	0.7320	0.5405	0.7320	0.8556
No log	10.0	130	0.7333	0.5405	0.7333	0.8563

Framework versions

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