ArabicNewSplits6_WithDuplicationsForScore5_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: 0.6757
• Qwk: 0.6888
• Mse: 0.6757
• Rmse: 0.8220

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	0.2857	2	2.2291	0.0408	2.2291	1.4930
No log	0.5714	4	1.4107	0.2161	1.4107	1.1877
No log	0.8571	6	1.2038	0.2634	1.2038	1.0972
No log	1.1429	8	1.2473	0.3538	1.2473	1.1168
No log	1.4286	10	1.3863	0.3907	1.3863	1.1774
No log	1.7143	12	1.4115	0.4301	1.4115	1.1881
No log	2.0	14	1.3178	0.4440	1.3178	1.1480
No log	2.2857	16	1.0888	0.4789	1.0888	1.0435
No log	2.5714	18	1.0108	0.4844	1.0108	1.0054
No log	2.8571	20	0.9046	0.5436	0.9046	0.9511
No log	3.1429	22	0.8340	0.5915	0.8340	0.9133
No log	3.4286	24	0.8197	0.6108	0.8197	0.9053
No log	3.7143	26	0.8901	0.6103	0.8901	0.9434
No log	4.0	28	0.7990	0.6280	0.7990	0.8939
No log	4.2857	30	0.7188	0.6644	0.7188	0.8478
No log	4.5714	32	0.6735	0.6770	0.6735	0.8207
No log	4.8571	34	0.6502	0.6915	0.6502	0.8063
No log	5.1429	36	0.6319	0.6791	0.6319	0.7949
No log	5.4286	38	0.6192	0.7041	0.6192	0.7869
No log	5.7143	40	0.6921	0.6867	0.6921	0.8319
No log	6.0	42	0.8217	0.6529	0.8217	0.9065
No log	6.2857	44	0.9014	0.6748	0.9014	0.9494
No log	6.5714	46	0.8613	0.6706	0.8613	0.9281
No log	6.8571	48	0.7102	0.6848	0.7102	0.8427
No log	7.1429	50	0.6102	0.7004	0.6102	0.7811
No log	7.4286	52	0.6139	0.6970	0.6139	0.7835
No log	7.7143	54	0.6091	0.7107	0.6091	0.7804
No log	8.0	56	0.6125	0.7004	0.6125	0.7826
No log	8.2857	58	0.6598	0.6832	0.6598	0.8123
No log	8.5714	60	0.6985	0.6911	0.6985	0.8358
No log	8.8571	62	0.6971	0.6911	0.6971	0.8349
No log	9.1429	64	0.6968	0.6911	0.6968	0.8348
No log	9.4286	66	0.6901	0.6976	0.6901	0.8307
No log	9.7143	68	0.6817	0.6888	0.6817	0.8257
No log	10.0	70	0.6757	0.6888	0.6757	0.8220

Framework versions

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