arabert_baseline_relevance_task3_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: 0.4533
• Qwk: 0.0
• Mse: 0.4662

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	0.6667	2	0.5670	0.0237	0.5557
No log	1.3333	4	0.2239	0.0	0.2180
No log	2.0	6	0.2912	0.0	0.2878
No log	2.6667	8	0.3225	0.0	0.3205
No log	3.3333	10	0.3274	0.0	0.3276
No log	4.0	12	0.3260	0.0	0.3295
No log	4.6667	14	0.3269	0.0	0.3323
No log	5.3333	16	0.3627	0.0	0.3697
No log	6.0	18	0.3851	0.0	0.3932
No log	6.6667	20	0.3452	0.0	0.3543
No log	7.3333	22	0.3589	0.0	0.3692
No log	8.0	24	0.3714	0.0	0.3825
No log	8.6667	26	0.4113	0.0	0.4233
No log	9.3333	28	0.4400	0.0	0.4526
No log	10.0	30	0.4533	0.0	0.4662

Framework versions

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