ArabicNewSplits6_FineTuningAraBERT_run2_AugV5_k3_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: 0.7724
• Qwk: 0.2646
• Mse: 0.7724
• Rmse: 0.8789

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	0.125	2	3.0876	0.0162	3.0876	1.7572
No log	0.25	4	1.5205	0.0210	1.5205	1.2331
No log	0.375	6	0.9349	0.0551	0.9349	0.9669
No log	0.5	8	0.7495	0.0918	0.7495	0.8657
No log	0.625	10	0.8371	0.0631	0.8371	0.9149
No log	0.75	12	0.7385	0.1111	0.7385	0.8594
No log	0.875	14	0.5999	0.0815	0.5999	0.7745
No log	1.0	16	0.6784	0.2350	0.6784	0.8236
No log	1.125	18	0.7568	0.2000	0.7568	0.8699
No log	1.25	20	0.6052	0.2222	0.6052	0.7780
No log	1.375	22	0.6104	0.2222	0.6104	0.7813
No log	1.5	24	0.6395	0.1282	0.6395	0.7997
No log	1.625	26	0.5864	0.1329	0.5864	0.7658
No log	1.75	28	0.5883	0.0725	0.5883	0.7670
No log	1.875	30	0.6065	0.1111	0.6065	0.7788
No log	2.0	32	0.8248	0.2744	0.8248	0.9082
No log	2.125	34	0.7368	0.1667	0.7368	0.8584
No log	2.25	36	0.6238	0.0815	0.6238	0.7898
No log	2.375	38	0.6442	-0.0435	0.6442	0.8026
No log	2.5	40	0.6526	0.1020	0.6526	0.8079
No log	2.625	42	0.6802	0.2749	0.6802	0.8248
No log	2.75	44	0.6389	0.1111	0.6389	0.7993
No log	2.875	46	0.5955	0.0303	0.5955	0.7717
No log	3.0	48	0.5888	0.0222	0.5888	0.7673
No log	3.125	50	0.6088	0.1282	0.6088	0.7803
No log	3.25	52	0.5988	0.0886	0.5988	0.7738
No log	3.375	54	0.6105	0.1707	0.6105	0.7814
No log	3.5	56	0.6595	0.2421	0.6595	0.8121
No log	3.625	58	0.6886	0.2727	0.6886	0.8298
No log	3.75	60	0.6641	0.1667	0.6641	0.8149
No log	3.875	62	0.8291	0.2000	0.8291	0.9106
No log	4.0	64	0.7989	0.2300	0.7989	0.8938
No log	4.125	66	0.7124	0.2762	0.7124	0.8440
No log	4.25	68	1.2798	0.1475	1.2798	1.1313
No log	4.375	70	1.1942	0.1523	1.1942	1.0928
No log	4.5	72	0.7125	0.2381	0.7125	0.8441
No log	4.625	74	0.9853	0.2320	0.9853	0.9926
No log	4.75	76	1.0188	0.1704	1.0188	1.0094
No log	4.875	78	0.7531	0.1759	0.7531	0.8678
No log	5.0	80	0.6711	0.2549	0.6711	0.8192
No log	5.125	82	0.8807	0.3306	0.8807	0.9385
No log	5.25	84	0.9129	0.2829	0.9129	0.9555
No log	5.375	86	0.7164	0.2442	0.7164	0.8464
No log	5.5	88	0.7397	0.1841	0.7397	0.8601
No log	5.625	90	0.9423	0.136	0.9423	0.9707
No log	5.75	92	0.8404	0.2000	0.8404	0.9167
No log	5.875	94	0.6311	0.2323	0.6311	0.7944
No log	6.0	96	0.6897	0.3524	0.6897	0.8305
No log	6.125	98	0.6644	0.3208	0.6644	0.8151
No log	6.25	100	0.6267	0.2692	0.6267	0.7917
No log	6.375	102	0.8165	0.2381	0.8165	0.9036
No log	6.5	104	0.8442	0.2070	0.8442	0.9188
No log	6.625	106	0.7264	0.2075	0.7264	0.8523
No log	6.75	108	0.6997	0.2308	0.6997	0.8365
No log	6.875	110	0.6898	0.2919	0.6898	0.8305
No log	7.0	112	0.7012	0.3208	0.7012	0.8374
No log	7.125	114	0.6888	0.3208	0.6888	0.8299
No log	7.25	116	0.7299	0.2372	0.7299	0.8544
No log	7.375	118	0.7473	0.2372	0.7473	0.8645
No log	7.5	120	0.8384	0.1861	0.8384	0.9156
No log	7.625	122	0.8254	0.2212	0.8254	0.9085
No log	7.75	124	0.7516	0.2523	0.7516	0.8669
No log	7.875	126	0.6728	0.2300	0.6728	0.8203
No log	8.0	128	0.6366	0.4074	0.6366	0.7979
No log	8.125	130	0.6395	0.4074	0.6395	0.7997
No log	8.25	132	0.6666	0.2692	0.6666	0.8164
No log	8.375	134	0.7637	0.2593	0.7637	0.8739
No log	8.5	136	0.8543	0.1867	0.8543	0.9243
No log	8.625	138	0.8706	0.1867	0.8706	0.9330
No log	8.75	140	0.8245	0.2479	0.8245	0.9080
No log	8.875	142	0.7541	0.2646	0.7541	0.8684
No log	9.0	144	0.7211	0.2857	0.7211	0.8492
No log	9.125	146	0.7185	0.3153	0.7185	0.8476
No log	9.25	148	0.7180	0.2511	0.7180	0.8473
No log	9.375	150	0.7137	0.2857	0.7137	0.8448
No log	9.5	152	0.7285	0.3188	0.7285	0.8535
No log	9.625	154	0.7479	0.2920	0.7479	0.8648
No log	9.75	156	0.7653	0.2646	0.7653	0.8748
No log	9.875	158	0.7693	0.2646	0.7693	0.8771
No log	10.0	160	0.7724	0.2646	0.7724	0.8789

Framework versions

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