arabert_baseline_relevance_task8_fold1

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: 0.2901
• Qwk: 0.4096
• Mse: 0.2901

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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse
No log	0.5	2	0.6071	-0.1667	0.6071
No log	1.0	4	0.6292	-0.0548	0.6292
No log	1.5	6	0.5112	0.3467	0.5112
No log	2.0	8	0.2941	0.2000	0.2941
No log	2.5	10	0.2595	0.3171	0.2595
No log	3.0	12	0.3011	0.3171	0.3011
No log	3.5	14	0.2934	0.3171	0.2934
No log	4.0	16	0.2489	0.3171	0.2489
No log	4.5	18	0.2716	0.4096	0.2716
No log	5.0	20	0.3200	0.3099	0.3200
No log	5.5	22	0.3416	0.2000	0.3416
No log	6.0	24	0.3237	0.4096	0.3237
No log	6.5	26	0.2949	0.4096	0.2949
No log	7.0	28	0.2821	0.4096	0.2821
No log	7.5	30	0.2766	0.4096	0.2766
No log	8.0	32	0.2780	0.4096	0.2780
No log	8.5	34	0.2807	0.4096	0.2807
No log	9.0	36	0.2841	0.4096	0.2841
No log	9.5	38	0.2889	0.4096	0.2889
No log	10.0	40	0.2901	0.4096	0.2901

Framework versions

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