arabert_baseline_development_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: 0.2965
• Qwk: 0.5749
• Mse: 0.2965

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	0.9461	0.2857	0.9461
No log	1.0	4	0.6077	0.6164	0.6077
No log	1.5	6	0.4804	0.5646	0.4804
No log	2.0	8	1.7805	0.1870	1.7805
No log	2.5	10	0.8599	0.3772	0.8599
No log	3.0	12	0.2911	0.6316	0.2911
No log	3.5	14	0.2592	0.6316	0.2592
No log	4.0	16	0.2544	0.6316	0.2544
No log	4.5	18	0.2769	0.6316	0.2769
No log	5.0	20	0.3520	0.5749	0.3520
No log	5.5	22	0.4464	0.5556	0.4464
No log	6.0	24	0.4041	0.5749	0.4041
No log	6.5	26	0.3621	0.6719	0.3621
No log	7.0	28	0.3337	0.7094	0.3337
No log	7.5	30	0.2849	0.7094	0.2849
No log	8.0	32	0.2691	0.7342	0.2691
No log	8.5	34	0.2750	0.6719	0.2750
No log	9.0	36	0.2882	0.5749	0.2882
No log	9.5	38	0.2930	0.5749	0.2930
No log	10.0	40	0.2965	0.5749	0.2965

Framework versions

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