arabert_cross_organization_task6_fold0

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: 0.7289
• Qwk: 0.5912
• Mse: 0.7281

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: 64
• eval_batch_size: 64
• 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
No log	0.125	2	3.7981	0.0861	3.7944
No log	0.25	4	1.8461	0.1823	1.8441
No log	0.375	6	1.1492	0.2329	1.1494
No log	0.5	8	1.1066	0.2992	1.1058
No log	0.625	10	1.6977	0.2447	1.6961
No log	0.75	12	1.1183	0.3816	1.1176
No log	0.875	14	0.8750	0.4339	0.8743
No log	1.0	16	0.8766	0.4674	0.8758
No log	1.125	18	0.8999	0.4970	0.8993
No log	1.25	20	1.0556	0.4407	1.0551
No log	1.375	22	0.9067	0.5062	0.9066
No log	1.5	24	0.7184	0.6052	0.7185
No log	1.625	26	0.6931	0.6242	0.6932
No log	1.75	28	0.7936	0.5514	0.7935
No log	1.875	30	0.8377	0.5377	0.8373
No log	2.0	32	0.8011	0.5791	0.8005
No log	2.125	34	0.7939	0.5753	0.7931
No log	2.25	36	0.8366	0.5598	0.8358
No log	2.375	38	0.8212	0.5795	0.8205
No log	2.5	40	0.8212	0.5632	0.8208
No log	2.625	42	0.7563	0.5916	0.7561
No log	2.75	44	0.6997	0.6115	0.6996
No log	2.875	46	0.6942	0.6175	0.6940
No log	3.0	48	0.7142	0.6105	0.7140
No log	3.125	50	0.7317	0.6078	0.7315
No log	3.25	52	0.7975	0.5612	0.7973
No log	3.375	54	0.7273	0.6054	0.7271
No log	3.5	56	0.6944	0.6134	0.6942
No log	3.625	58	0.7265	0.6116	0.7261
No log	3.75	60	0.7407	0.5990	0.7403
No log	3.875	62	0.7605	0.5691	0.7603
No log	4.0	64	0.8547	0.5212	0.8545
No log	4.125	66	0.8228	0.5516	0.8226
No log	4.25	68	0.7513	0.5927	0.7511
No log	4.375	70	0.7420	0.6018	0.7418
No log	4.5	72	0.7641	0.5709	0.7638
No log	4.625	74	0.7963	0.5551	0.7959
No log	4.75	76	0.7971	0.5599	0.7966
No log	4.875	78	0.7809	0.5579	0.7805
No log	5.0	80	0.7156	0.5743	0.7153
No log	5.125	82	0.6952	0.5899	0.6949
No log	5.25	84	0.6808	0.5961	0.6805
No log	5.375	86	0.6954	0.5871	0.6949
No log	5.5	88	0.7218	0.5816	0.7212
No log	5.625	90	0.7876	0.5645	0.7868
No log	5.75	92	0.7595	0.5754	0.7586
No log	5.875	94	0.7368	0.6061	0.7359
No log	6.0	96	0.7572	0.6099	0.7562
No log	6.125	98	0.7439	0.5964	0.7429
No log	6.25	100	0.7256	0.6034	0.7248
No log	6.375	102	0.7947	0.5680	0.7940
No log	6.5	104	0.8265	0.5668	0.8259
No log	6.625	106	0.7903	0.5673	0.7898
No log	6.75	108	0.7488	0.5824	0.7483
No log	6.875	110	0.7186	0.5990	0.7181
No log	7.0	112	0.7043	0.6168	0.7037
No log	7.125	114	0.6952	0.6292	0.6946
No log	7.25	116	0.6989	0.6175	0.6983
No log	7.375	118	0.7371	0.5881	0.7365
No log	7.5	120	0.7318	0.5884	0.7312
No log	7.625	122	0.7079	0.6065	0.7074
No log	7.75	124	0.7068	0.6169	0.7062
No log	7.875	126	0.7229	0.6054	0.7222
No log	8.0	128	0.7270	0.6014	0.7262
No log	8.125	130	0.7290	0.5959	0.7282
No log	8.25	132	0.7324	0.5958	0.7316
No log	8.375	134	0.7328	0.5947	0.7320
No log	8.5	136	0.7398	0.5911	0.7390
No log	8.625	138	0.7320	0.5901	0.7312
No log	8.75	140	0.7358	0.5861	0.7349
No log	8.875	142	0.7295	0.5902	0.7286
No log	9.0	144	0.7119	0.5893	0.7110
No log	9.125	146	0.7008	0.6031	0.7000
No log	9.25	148	0.6991	0.6031	0.6984
No log	9.375	150	0.7049	0.5906	0.7041
No log	9.5	152	0.7156	0.5915	0.7148
No log	9.625	154	0.7220	0.5914	0.7212
No log	9.75	156	0.7265	0.5923	0.7257
No log	9.875	158	0.7278	0.5923	0.7270
No log	10.0	160	0.7289	0.5912	0.7281

Framework versions

• Transformers 4.44.0
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1