arabert_baseline_grammar_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: 0.2845
• Qwk: 0.7701
• Mse: 0.2845

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	1.4505	0.2222	1.4505
No log	1.0	4	0.9573	0.4455	0.9573
No log	1.5	6	0.5247	0.5852	0.5247
No log	2.0	8	0.9251	0.5783	0.9251
No log	2.5	10	0.3991	0.6688	0.3991
No log	3.0	12	0.2870	0.8323	0.2870
No log	3.5	14	0.2995	0.8323	0.2995
No log	4.0	16	0.3141	0.8062	0.3141
No log	4.5	18	0.3722	0.6994	0.3722
No log	5.0	20	0.3507	0.6994	0.3507
No log	5.5	22	0.3458	0.6087	0.3458
No log	6.0	24	0.2616	0.8146	0.2616
No log	6.5	26	0.2314	0.8350	0.2314
No log	7.0	28	0.2388	0.8350	0.2388
No log	7.5	30	0.2311	0.8627	0.2311
No log	8.0	32	0.2383	0.8889	0.2383
No log	8.5	34	0.2552	0.8627	0.2552
No log	9.0	36	0.2780	0.7785	0.2780
No log	9.5	38	0.2870	0.7701	0.2870
No log	10.0	40	0.2845	0.7701	0.2845

Framework versions

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