arabert_baseline_style_task1_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: 0.5100
• Qwk: 0.6698
• Mse: 0.5053

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	0.3333	2	5.4651	-0.0419	5.4241
No log	0.6667	4	2.3074	0.1693	2.2743
No log	1.0	6	1.2115	0.1561	1.2001
No log	1.3333	8	0.9959	0.4358	0.9923
No log	1.6667	10	0.7290	0.4921	0.7317
No log	2.0	12	0.7145	0.5587	0.7193
No log	2.3333	14	0.7329	0.5435	0.7351
No log	2.6667	16	0.7694	0.5219	0.7660
No log	3.0	18	0.9233	0.4358	0.9210
No log	3.3333	20	0.8596	0.4516	0.8545
No log	3.6667	22	0.7445	0.5743	0.7322
No log	4.0	24	0.7760	0.4773	0.7614
No log	4.3333	26	0.6783	0.5743	0.6684
No log	4.6667	28	0.7836	0.5152	0.7834
No log	5.0	30	0.7387	0.5188	0.7401
No log	5.3333	32	0.5679	0.5743	0.5656
No log	5.6667	34	0.5183	0.5743	0.5136
No log	6.0	36	0.5055	0.5743	0.5017
No log	6.3333	38	0.5518	0.5188	0.5511
No log	6.6667	40	0.6558	0.5188	0.6585
No log	7.0	42	0.6773	0.6025	0.6812
No log	7.3333	44	0.6216	0.6025	0.6242
No log	7.6667	46	0.5382	0.6698	0.5375
No log	8.0	48	0.5083	0.6698	0.5056
No log	8.3333	50	0.5045	0.6698	0.5004
No log	8.6667	52	0.5059	0.7151	0.5009
No log	9.0	54	0.5070	0.7151	0.5018
No log	9.3333	56	0.5069	0.7151	0.5016
No log	9.6667	58	0.5086	0.7151	0.5037
No log	10.0	60	0.5100	0.6698	0.5053

Framework versions

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