arabert_baseline_grammar_task8_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: 0.5087
• Qwk: 0.6725
• Mse: 0.5087

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	2.2317	0.0	2.2317
No log	1.0	4	1.4866	0.2297	1.4866
No log	1.5	6	1.0270	0.5643	1.0270
No log	2.0	8	1.1388	0.4074	1.1388
No log	2.5	10	1.0946	0.4074	1.0946
No log	3.0	12	0.5591	0.6159	0.5591
No log	3.5	14	0.5042	0.6216	0.5042
No log	4.0	16	0.4874	0.6585	0.4874
No log	4.5	18	0.5862	0.6053	0.5862
No log	5.0	20	0.7098	0.5549	0.7098
No log	5.5	22	0.6879	0.5549	0.6879
No log	6.0	24	0.5641	0.6053	0.5641
No log	6.5	26	0.5895	0.6725	0.5895
No log	7.0	28	0.6417	0.6211	0.6417
No log	7.5	30	0.7537	0.6211	0.7537
No log	8.0	32	0.6776	0.6211	0.6776
No log	8.5	34	0.6081	0.6211	0.6081
No log	9.0	36	0.5508	0.6725	0.5508
No log	9.5	38	0.5104	0.6725	0.5104
No log	10.0	40	0.5087	0.6725	0.5087

Framework versions

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