arabert_baseline_grammar_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: 0.4892
• Qwk: 0.6051
• Mse: 0.4892

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	1.0890	0.2687	1.0890
No log	1.0	4	0.7915	0.3000	0.7915
No log	1.5	6	0.8650	0.3529	0.8650
No log	2.0	8	0.7927	0.3529	0.7927
No log	2.5	10	0.6821	0.2602	0.6821
No log	3.0	12	0.6081	0.4793	0.6081
No log	3.5	14	0.6266	0.5882	0.6266
No log	4.0	16	0.7668	0.5191	0.7668
No log	4.5	18	0.7382	0.6164	0.7382
No log	5.0	20	0.5722	0.7651	0.5722
No log	5.5	22	0.5444	0.6755	0.5444
No log	6.0	24	0.5298	0.6755	0.5298
No log	6.5	26	0.5351	0.7651	0.5351
No log	7.0	28	0.5998	0.6957	0.5998
No log	7.5	30	0.5874	0.6957	0.5874
No log	8.0	32	0.5369	0.6216	0.5369
No log	8.5	34	0.4961	0.6480	0.4961
No log	9.0	36	0.4879	0.6051	0.4879
No log	9.5	38	0.4895	0.6051	0.4895
No log	10.0	40	0.4892	0.6051	0.4892

Framework versions

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