ArabicNewSplits6_WithDuplicationsForScore5_FineTuningAraBERT_run2_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: 1.2126
• Qwk: 0.5990
• Mse: 1.2126
• Rmse: 1.1012

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	2.2049	0.0523	2.2049	1.4849
No log	0.6667	4	1.4421	0.2062	1.4421	1.2009
No log	1.0	6	1.4432	0.1239	1.4432	1.2014
No log	1.3333	8	1.6837	0.3015	1.6837	1.2976
No log	1.6667	10	2.0500	0.2336	2.0500	1.4318
No log	2.0	12	2.2219	0.1702	2.2219	1.4906
No log	2.3333	14	1.9707	0.2710	1.9707	1.4038
No log	2.6667	16	1.6829	0.3751	1.6829	1.2972
No log	3.0	18	1.2477	0.4713	1.2477	1.1170
No log	3.3333	20	1.1112	0.5267	1.1112	1.0541
No log	3.6667	22	1.2117	0.5409	1.2117	1.1008
No log	4.0	24	1.4098	0.5526	1.4098	1.1873
No log	4.3333	26	1.4175	0.5474	1.4175	1.1906
No log	4.6667	28	1.2567	0.5568	1.2567	1.1210
No log	5.0	30	1.0613	0.5710	1.0613	1.0302
No log	5.3333	32	0.9790	0.5766	0.9790	0.9895
No log	5.6667	34	1.0823	0.6013	1.0823	1.0404
No log	6.0	36	1.1279	0.6000	1.1279	1.0620
No log	6.3333	38	1.1179	0.6035	1.1179	1.0573
No log	6.6667	40	1.2520	0.6144	1.2520	1.1189
No log	7.0	42	1.1995	0.6293	1.1995	1.0952
No log	7.3333	44	1.1625	0.6331	1.1625	1.0782
No log	7.6667	46	1.1039	0.6263	1.1039	1.0506
No log	8.0	48	1.0191	0.6228	1.0191	1.0095
No log	8.3333	50	1.0057	0.6166	1.0057	1.0028
No log	8.6667	52	1.0395	0.6185	1.0395	1.0196
No log	9.0	54	1.0830	0.6334	1.0830	1.0407
No log	9.3333	56	1.1535	0.5903	1.1535	1.0740
No log	9.6667	58	1.2103	0.5990	1.2103	1.1001
No log	10.0	60	1.2126	0.5990	1.2126	1.1012

Framework versions

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