arabert_cross_organization_task2_fold6

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: 0.8091
• Qwk: 0.4688
• Mse: 0.8081

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	2.7795	0.0105	2.7827
No log	0.25	4	1.5291	0.1173	1.5287
No log	0.375	6	0.8451	0.3969	0.8442
No log	0.5	8	0.7864	0.4495	0.7852
No log	0.625	10	0.6936	0.4833	0.6924
No log	0.75	12	1.1360	0.3318	1.1343
No log	0.875	14	1.0124	0.3651	1.0104
No log	1.0	16	0.6020	0.5728	0.6007
No log	1.125	18	0.7313	0.4817	0.7294
No log	1.25	20	0.6700	0.4955	0.6684
No log	1.375	22	0.4911	0.6467	0.4906
No log	1.5	24	0.4996	0.7310	0.4995
No log	1.625	26	0.4835	0.5735	0.4828
No log	1.75	28	0.6109	0.4913	0.6097
No log	1.875	30	0.5349	0.5702	0.5337
No log	2.0	32	0.5033	0.6710	0.5027
No log	2.125	34	0.5412	0.6126	0.5400
No log	2.25	36	0.7223	0.4624	0.7201
No log	2.375	38	0.7359	0.4653	0.7337
No log	2.5	40	0.5466	0.5891	0.5455
No log	2.625	42	0.5036	0.7135	0.5034
No log	2.75	44	0.4797	0.6820	0.4794
No log	2.875	46	0.5194	0.5891	0.5185
No log	3.0	48	0.5890	0.5385	0.5878
No log	3.125	50	0.5541	0.5617	0.5529
No log	3.25	52	0.5096	0.6247	0.5088
No log	3.375	54	0.5168	0.6305	0.5160
No log	3.5	56	0.5860	0.5436	0.5848
No log	3.625	58	0.7175	0.4990	0.7160
No log	3.75	60	0.6365	0.5253	0.6352
No log	3.875	62	0.5233	0.6176	0.5225
No log	4.0	64	0.5179	0.6032	0.5171
No log	4.125	66	0.5802	0.5490	0.5792
No log	4.25	68	0.6555	0.5187	0.6543
No log	4.375	70	0.6681	0.5214	0.6669
No log	4.5	72	0.5977	0.5522	0.5967
No log	4.625	74	0.6327	0.5265	0.6317
No log	4.75	76	0.7255	0.4790	0.7244
No log	4.875	78	0.6478	0.5237	0.6469
No log	5.0	80	0.6738	0.4846	0.6728
No log	5.125	82	0.7006	0.4808	0.6996
No log	5.25	84	0.6997	0.4980	0.6987
No log	5.375	86	0.7244	0.5042	0.7234
No log	5.5	88	0.8734	0.4504	0.8720
No log	5.625	90	0.9975	0.3988	0.9959
No log	5.75	92	0.8921	0.4437	0.8908
No log	5.875	94	0.6714	0.5075	0.6707
No log	6.0	96	0.6012	0.5740	0.6007
No log	6.125	98	0.6310	0.5373	0.6304
No log	6.25	100	0.7371	0.4660	0.7361
No log	6.375	102	0.7404	0.4660	0.7394
No log	6.5	104	0.6821	0.5187	0.6812
No log	6.625	106	0.7033	0.5011	0.7023
No log	6.75	108	0.8055	0.4525	0.8043
No log	6.875	110	0.9021	0.4325	0.9007
No log	7.0	112	0.8396	0.4487	0.8384
No log	7.125	114	0.7100	0.5011	0.7090
No log	7.25	116	0.6902	0.5001	0.6894
No log	7.375	118	0.7441	0.4773	0.7431
No log	7.5	120	0.8389	0.4453	0.8379
No log	7.625	122	0.8787	0.4279	0.8776
No log	7.75	124	0.8597	0.4412	0.8587
No log	7.875	126	0.7923	0.4723	0.7914
No log	8.0	128	0.7502	0.4956	0.7493
No log	8.125	130	0.7529	0.4956	0.7521
No log	8.25	132	0.8227	0.4737	0.8217
No log	8.375	134	0.9122	0.4331	0.9110
No log	8.5	136	0.9081	0.4331	0.9069
No log	8.625	138	0.8495	0.4547	0.8484
No log	8.75	140	0.8359	0.4581	0.8348
No log	8.875	142	0.8222	0.4649	0.8211
No log	9.0	144	0.8247	0.4649	0.8237
No log	9.125	146	0.8487	0.4615	0.8476
No log	9.25	148	0.8495	0.4581	0.8485
No log	9.375	150	0.8322	0.4649	0.8311
No log	9.5	152	0.8155	0.4688	0.8145
No log	9.625	154	0.7982	0.4774	0.7972
No log	9.75	156	0.7978	0.4774	0.7968
No log	9.875	158	0.8054	0.4774	0.8044
No log	10.0	160	0.8091	0.4688	0.8081

Framework versions

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