arabert_baseline_style_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: 0.5805
• Qwk: 0.6441
• Mse: 0.5805

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	1.2965	0.2222	1.2965
No log	1.0	4	0.8067	0.1439	0.8067
No log	1.5	6	0.7795	0.3684	0.7795
No log	2.0	8	0.7284	0.3684	0.7284
No log	2.5	10	0.6398	0.3259	0.6398
No log	3.0	12	0.6052	0.5926	0.6052
No log	3.5	14	0.5965	0.5746	0.5965
No log	4.0	16	1.0642	0.6351	1.0642
No log	4.5	18	1.1268	0.5882	1.1268
No log	5.0	20	1.0448	0.6351	1.0448
No log	5.5	22	0.7295	0.5862	0.7295
No log	6.0	24	0.5209	0.5193	0.5209
No log	6.5	26	0.5470	0.4093	0.5470
No log	7.0	28	0.4932	0.4936	0.4932
No log	7.5	30	0.4781	0.6940	0.4781
No log	8.0	32	0.5670	0.6441	0.5670
No log	8.5	34	0.6172	0.6631	0.6172
No log	9.0	36	0.6178	0.6631	0.6178
No log	9.5	38	0.5895	0.6631	0.5895
No log	10.0	40	0.5805	0.6441	0.5805

Framework versions

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