arabert_cross_vocabulary_task6_fold3

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: 0.9261
• Qwk: 0.0
• Mse: 0.9261

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse
No log	0.0290	2	6.5303	0.0019	6.5303
No log	0.0580	4	3.1144	0.0013	3.1144
No log	0.0870	6	1.8229	0.0085	1.8229
No log	0.1159	8	1.2398	0.0	1.2398
No log	0.1449	10	0.7589	0.0496	0.7589
No log	0.1739	12	0.7859	0.0071	0.7859
No log	0.2029	14	0.8223	-0.0684	0.8223
No log	0.2319	16	0.8121	-0.0353	0.8121
No log	0.2609	18	0.7513	-0.0578	0.7513
No log	0.2899	20	0.7698	0.0309	0.7698
No log	0.3188	22	0.7570	-0.0286	0.7570
No log	0.3478	24	0.7494	-0.0743	0.7494
No log	0.3768	26	0.7478	0.0	0.7478
No log	0.4058	28	0.7555	0.0	0.7555
No log	0.4348	30	0.7954	0.0	0.7954
No log	0.4638	32	0.8212	0.0	0.8212
No log	0.4928	34	0.8769	0.0	0.8769
No log	0.5217	36	0.9076	0.0	0.9076
No log	0.5507	38	0.8669	0.0	0.8669
No log	0.5797	40	0.7835	-0.0530	0.7835
No log	0.6087	42	0.8154	-0.0118	0.8154
No log	0.6377	44	1.0142	0.0093	1.0142
No log	0.6667	46	1.0534	-0.0026	1.0534
No log	0.6957	48	0.9329	0.0381	0.9329
No log	0.7246	50	0.8143	0.0225	0.8143
No log	0.7536	52	0.7703	-0.0860	0.7703
No log	0.7826	54	0.7862	0.0	0.7862
No log	0.8116	56	0.8271	0.0	0.8271
No log	0.8406	58	0.8682	0.0	0.8682
No log	0.8696	60	0.9099	0.0	0.9099
No log	0.8986	62	0.9266	0.0	0.9266
No log	0.9275	64	0.9343	0.0	0.9343
No log	0.9565	66	0.9325	0.0	0.9325
No log	0.9855	68	0.9261	0.0	0.9261

Framework versions

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