arabert_baseline_development_task6_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: 0.7890
• Qwk: 0.4503
• Mse: 0.7890

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	1.5011	0.2195	1.5011
No log	1.0	4	1.0235	0.3558	1.0235
No log	1.5	6	1.1605	0.3913	1.1605
No log	2.0	8	1.0269	0.3558	1.0269
No log	2.5	10	0.9070	0.3558	0.9070
No log	3.0	12	0.8423	0.3913	0.8423
No log	3.5	14	0.7993	0.3558	0.7993
No log	4.0	16	0.7964	0.4740	0.7964
No log	4.5	18	0.8073	0.4740	0.8073
No log	5.0	20	0.8740	0.6091	0.8740
No log	5.5	22	0.8159	0.6056	0.8159
No log	6.0	24	0.8260	0.6056	0.8260
No log	6.5	26	0.8604	0.6056	0.8604
No log	7.0	28	0.8853	0.6056	0.8853
No log	7.5	30	0.8139	0.5172	0.8139
No log	8.0	32	0.7491	0.4503	0.7491
No log	8.5	34	0.7564	0.4503	0.7564
No log	9.0	36	0.7842	0.4503	0.7842
No log	9.5	38	0.7831	0.4503	0.7831
No log	10.0	40	0.7890	0.4503	0.7890

Framework versions

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