arabert_cross_organization_task5_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: 0.6701
• Qwk: 0.5451
• Mse: 0.6691

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	2.7122	0.0168	2.7155
No log	0.25	4	1.4877	0.0689	1.4872
No log	0.375	6	0.8699	0.3929	0.8691
No log	0.5	8	0.8864	0.4400	0.8851
No log	0.625	10	0.6806	0.4650	0.6795
No log	0.75	12	0.5592	0.6237	0.5585
No log	0.875	14	0.5290	0.6160	0.5281
No log	1.0	16	0.6313	0.5082	0.6296
No log	1.125	18	0.5309	0.6193	0.5297
No log	1.25	20	0.5510	0.7458	0.5512
No log	1.375	22	0.5012	0.7370	0.5009
No log	1.5	24	0.5238	0.5768	0.5223
No log	1.625	26	0.4818	0.6624	0.4807
No log	1.75	28	0.4667	0.6909	0.4659
No log	1.875	30	0.4599	0.6935	0.4592
No log	2.0	32	0.4866	0.6274	0.4855
No log	2.125	34	0.5237	0.5988	0.5225
No log	2.25	36	0.5515	0.6989	0.5510
No log	2.375	38	0.5435	0.6886	0.5430
No log	2.5	40	0.5180	0.6261	0.5171
No log	2.625	42	0.4674	0.7076	0.4669
No log	2.75	44	0.4765	0.7329	0.4764
No log	2.875	46	0.4676	0.7156	0.4672
No log	3.0	48	0.5297	0.5702	0.5287
No log	3.125	50	0.5444	0.5777	0.5434
No log	3.25	52	0.5202	0.6120	0.5195
No log	3.375	54	0.5330	0.6250	0.5324
No log	3.5	56	0.5282	0.6303	0.5276
No log	3.625	58	0.5263	0.6051	0.5256
No log	3.75	60	0.4900	0.6534	0.4896
No log	3.875	62	0.4962	0.7418	0.4960
No log	4.0	64	0.5088	0.6350	0.5084
No log	4.125	66	0.6180	0.5459	0.6170
No log	4.25	68	0.7619	0.4764	0.7604
No log	4.375	70	0.6897	0.5533	0.6885
No log	4.5	72	0.5780	0.6397	0.5775
No log	4.625	74	0.5272	0.6578	0.5269
No log	4.75	76	0.5132	0.6359	0.5126
No log	4.875	78	0.5879	0.5243	0.5869
No log	5.0	80	0.5987	0.5203	0.5977
No log	5.125	82	0.5609	0.5675	0.5602
No log	5.25	84	0.5591	0.6400	0.5588
No log	5.375	86	0.5977	0.6016	0.5971
No log	5.5	88	0.6286	0.5753	0.6277
No log	5.625	90	0.6304	0.5617	0.6293
No log	5.75	92	0.6330	0.5239	0.6318
No log	5.875	94	0.5530	0.5293	0.5520
No log	6.0	96	0.4931	0.6645	0.4927
No log	6.125	98	0.5014	0.7126	0.5012
No log	6.25	100	0.5246	0.6717	0.5242
No log	6.375	102	0.6077	0.5609	0.6068
No log	6.5	104	0.7359	0.5061	0.7344
No log	6.625	106	0.7769	0.5024	0.7753
No log	6.75	108	0.7401	0.5143	0.7386
No log	6.875	110	0.6620	0.5573	0.6609
No log	7.0	112	0.6087	0.5571	0.6078
No log	7.125	114	0.5917	0.5262	0.5907
No log	7.25	116	0.5659	0.5230	0.5650
No log	7.375	118	0.5449	0.5886	0.5442
No log	7.5	120	0.5472	0.6127	0.5467
No log	7.625	122	0.5711	0.6000	0.5706
No log	7.75	124	0.6023	0.5779	0.6016
No log	7.875	126	0.6306	0.5660	0.6297
No log	8.0	128	0.6471	0.5684	0.6461
No log	8.125	130	0.6631	0.5684	0.6621
No log	8.25	132	0.6608	0.5684	0.6598
No log	8.375	134	0.6696	0.5678	0.6686
No log	8.5	136	0.6628	0.5569	0.6618
No log	8.625	138	0.6415	0.5560	0.6405
No log	8.75	140	0.6188	0.5664	0.6180
No log	8.875	142	0.5990	0.5782	0.5984
No log	9.0	144	0.5948	0.6002	0.5943
No log	9.125	146	0.5994	0.5935	0.5989
No log	9.25	148	0.6090	0.5803	0.6084
No log	9.375	150	0.6188	0.5734	0.6182
No log	9.5	152	0.6329	0.5705	0.6322
No log	9.625	154	0.6491	0.5642	0.6482
No log	9.75	156	0.6638	0.5451	0.6628
No log	9.875	158	0.6687	0.5451	0.6678
No log	10.0	160	0.6701	0.5451	0.6691

Framework versions

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