arabert_baseline_mechanics_task4_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: 0.4306
• Qwk: 0.1429
• Mse: 0.4306

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	1.0	2	4.1119	0.0	4.1119
No log	2.0	4	2.8498	0.0	2.8498
No log	3.0	6	2.1097	-0.0270	2.1097
No log	4.0	8	1.5588	0.0	1.5588
No log	5.0	10	0.8885	0.0	0.8885
No log	6.0	12	0.6177	0.0	0.6177
No log	7.0	14	0.5062	-0.4286	0.5062
No log	8.0	16	0.4578	0.0	0.4578
No log	9.0	18	0.4347	0.0909	0.4347
No log	10.0	20	0.4306	0.1429	0.4306

Framework versions

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