arabert_baseline_mechanics_task6_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: 0.7645
• Qwk: 0.6286
• Mse: 0.7645

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	1.8889	0.0476	1.8889
No log	1.0	4	1.1278	0.4828	1.1278
No log	1.5	6	1.1361	0.3077	1.1361
No log	2.0	8	1.2627	0.1818	1.2627
No log	2.5	10	1.5946	0.25	1.5946
No log	3.0	12	1.0845	0.3429	1.0845
No log	3.5	14	0.8934	0.3333	0.8934
No log	4.0	16	0.9166	0.5333	0.9166
No log	4.5	18	0.9074	0.5333	0.9074
No log	5.0	20	0.8620	0.6061	0.8620
No log	5.5	22	0.8171	0.5294	0.8171
No log	6.0	24	0.7859	0.5294	0.7859
No log	6.5	26	0.8244	0.6111	0.8244
No log	7.0	28	0.8510	0.5946	0.8510
No log	7.5	30	0.8300	0.5946	0.8300
No log	8.0	32	0.7975	0.6286	0.7975
No log	8.5	34	0.7763	0.6286	0.7763
No log	9.0	36	0.7703	0.6286	0.7703
No log	9.5	38	0.7657	0.6286	0.7657
No log	10.0	40	0.7645	0.6286	0.7645

Framework versions

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