arabert_cross_organization_task3_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: 0.6414
• Qwk: 0.6254
• Mse: 0.6411

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	3.6434	0.0780	3.6404
No log	0.25	4	1.6551	0.1755	1.6543
No log	0.375	6	0.9488	0.4106	0.9484
No log	0.5	8	1.0719	0.3366	1.0716
No log	0.625	10	1.0207	0.3808	1.0205
No log	0.75	12	1.0654	0.3920	1.0651
No log	0.875	14	1.0180	0.4457	1.0177
No log	1.0	16	0.8222	0.5205	0.8220
No log	1.125	18	0.7688	0.5547	0.7686
No log	1.25	20	0.7543	0.5667	0.7542
No log	1.375	22	0.7022	0.5869	0.7023
No log	1.5	24	0.6869	0.5910	0.6872
No log	1.625	26	0.6389	0.6198	0.6392
No log	1.75	28	0.6739	0.6023	0.6742
No log	1.875	30	0.8507	0.5267	0.8506
No log	2.0	32	0.7005	0.6030	0.7005
No log	2.125	34	0.6617	0.6277	0.6616
No log	2.25	36	0.6952	0.6193	0.6949
No log	2.375	38	0.7256	0.5870	0.7252
No log	2.5	40	0.6299	0.6350	0.6298
No log	2.625	42	0.6176	0.6371	0.6175
No log	2.75	44	0.6444	0.6384	0.6442
No log	2.875	46	0.6872	0.5979	0.6869
No log	3.0	48	0.8213	0.5261	0.8207
No log	3.125	50	0.7530	0.5473	0.7526
No log	3.25	52	0.6049	0.6395	0.6048
No log	3.375	54	0.6171	0.6335	0.6172
No log	3.5	56	0.5849	0.6419	0.5850
No log	3.625	58	0.6694	0.5941	0.6691
No log	3.75	60	0.7309	0.5351	0.7304
No log	3.875	62	0.7275	0.5529	0.7270
No log	4.0	64	0.6510	0.5973	0.6507
No log	4.125	66	0.6155	0.6262	0.6153
No log	4.25	68	0.6374	0.5996	0.6372
No log	4.375	70	0.6653	0.5931	0.6651
No log	4.5	72	0.6605	0.6196	0.6604
No log	4.625	74	0.6569	0.6417	0.6569
No log	4.75	76	0.6686	0.6063	0.6685
No log	4.875	78	0.7418	0.5794	0.7412
No log	5.0	80	0.7296	0.5729	0.7290
No log	5.125	82	0.6713	0.6226	0.6708
No log	5.25	84	0.6820	0.6063	0.6815
No log	5.375	86	0.6422	0.6148	0.6418
No log	5.5	88	0.6309	0.6239	0.6306
No log	5.625	90	0.6305	0.6251	0.6303
No log	5.75	92	0.6548	0.6222	0.6545
No log	5.875	94	0.6437	0.6155	0.6434
No log	6.0	96	0.6279	0.6238	0.6276
No log	6.125	98	0.6403	0.6193	0.6399
No log	6.25	100	0.6589	0.6169	0.6584
No log	6.375	102	0.6903	0.5964	0.6897
No log	6.5	104	0.6913	0.5965	0.6908
No log	6.625	106	0.6504	0.6147	0.6500
No log	6.75	108	0.6328	0.6279	0.6326
No log	6.875	110	0.6266	0.6408	0.6266
No log	7.0	112	0.6369	0.6376	0.6368
No log	7.125	114	0.6952	0.6018	0.6949
No log	7.25	116	0.7459	0.5938	0.7454
No log	7.375	118	0.7167	0.5933	0.7163
No log	7.5	120	0.6903	0.5986	0.6900
No log	7.625	122	0.6830	0.5957	0.6826
No log	7.75	124	0.6864	0.6038	0.6859
No log	7.875	126	0.6719	0.6101	0.6715
No log	8.0	128	0.6397	0.6237	0.6394
No log	8.125	130	0.6323	0.6296	0.6321
No log	8.25	132	0.6430	0.6264	0.6428
No log	8.375	134	0.6710	0.6229	0.6706
No log	8.5	136	0.6705	0.6229	0.6702
No log	8.625	138	0.6535	0.6235	0.6533
No log	8.75	140	0.6507	0.6235	0.6505
No log	8.875	142	0.6514	0.6225	0.6511
No log	9.0	144	0.6704	0.6186	0.6700
No log	9.125	146	0.6775	0.6242	0.6771
No log	9.25	148	0.6775	0.6183	0.6771
No log	9.375	150	0.6708	0.6190	0.6704
No log	9.5	152	0.6600	0.6231	0.6596
No log	9.625	154	0.6487	0.6220	0.6483
No log	9.75	156	0.6435	0.6254	0.6432
No log	9.875	158	0.6413	0.6266	0.6410
No log	10.0	160	0.6414	0.6254	0.6411

Framework versions

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