ArabicNewSplits6_FineTuningAraBERT_run1_AugV5_k1_task5_organization

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: 1.1757
• Qwk: 0.5591
• Mse: 1.1757
• Rmse: 1.0843

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: 8
• eval_batch_size: 8
• 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	Rmse
No log	0.3333	2	2.2751	0.0754	2.2751	1.5083
No log	0.6667	4	1.4631	0.1916	1.4631	1.2096
No log	1.0	6	1.2862	0.2251	1.2862	1.1341
No log	1.3333	8	1.4077	0.3051	1.4077	1.1865
No log	1.6667	10	1.5598	0.3233	1.5598	1.2489
No log	2.0	12	1.7420	0.3061	1.7420	1.3198
No log	2.3333	14	1.8863	0.3024	1.8863	1.3734
No log	2.6667	16	1.8811	0.3115	1.8811	1.3715
No log	3.0	18	1.6201	0.3958	1.6201	1.2728
No log	3.3333	20	1.2105	0.4758	1.2105	1.1002
No log	3.6667	22	1.1342	0.5052	1.1342	1.0650
No log	4.0	24	1.1248	0.5248	1.1248	1.0605
No log	4.3333	26	1.1648	0.5133	1.1648	1.0793
No log	4.6667	28	1.1437	0.5277	1.1437	1.0694
No log	5.0	30	1.1393	0.5345	1.1393	1.0674
No log	5.3333	32	1.1861	0.5230	1.1861	1.0891
No log	5.6667	34	1.1782	0.5161	1.1782	1.0855
No log	6.0	36	1.0884	0.5549	1.0884	1.0433
No log	6.3333	38	0.9953	0.5658	0.9953	0.9976
No log	6.6667	40	0.9034	0.5783	0.9034	0.9505
No log	7.0	42	0.8922	0.5728	0.8922	0.9445
No log	7.3333	44	0.9335	0.5862	0.9335	0.9662
No log	7.6667	46	1.0067	0.5483	1.0067	1.0033
No log	8.0	48	1.1065	0.5591	1.1065	1.0519
No log	8.3333	50	1.1425	0.5591	1.1425	1.0689
No log	8.6667	52	1.1581	0.5591	1.1581	1.0762
No log	9.0	54	1.1645	0.5591	1.1645	1.0791
No log	9.3333	56	1.1748	0.5591	1.1748	1.0839
No log	9.6667	58	1.1789	0.5591	1.1789	1.0857
No log	10.0	60	1.1757	0.5591	1.1757	1.0843

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu118
• Datasets 2.21.0
• Tokenizers 0.19.1