ArabicNewSplits6_FineTuningAraBERT_run3_AugV5_k1_task1_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: 0.7776
• Qwk: 0.5621
• Mse: 0.7776
• Rmse: 0.8818

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	0.25	2	5.4237	-0.0126	5.4237	2.3289
No log	0.5	4	3.0585	0.0908	3.0585	1.7489
No log	0.75	6	1.7385	0.1255	1.7385	1.3185
No log	1.0	8	1.1790	0.2362	1.1790	1.0858
No log	1.25	10	1.3043	0.1578	1.3043	1.1421
No log	1.5	12	1.3931	0.1301	1.3931	1.1803
No log	1.75	14	1.3789	0.1290	1.3789	1.1743
No log	2.0	16	1.3012	0.1715	1.3012	1.1407
No log	2.25	18	1.3019	0.1748	1.3019	1.1410
No log	2.5	20	1.2357	0.2141	1.2357	1.1116
No log	2.75	22	1.2045	0.2396	1.2045	1.0975
No log	3.0	24	1.1729	0.1839	1.1729	1.0830
No log	3.25	26	1.1284	0.2258	1.1284	1.0623
No log	3.5	28	1.1133	0.2049	1.1133	1.0551
No log	3.75	30	1.0995	0.1837	1.0995	1.0486
No log	4.0	32	1.0663	0.2537	1.0663	1.0326
No log	4.25	34	1.0297	0.2315	1.0297	1.0147
No log	4.5	36	1.0239	0.3035	1.0239	1.0119
No log	4.75	38	0.9992	0.3470	0.9992	0.9996
No log	5.0	40	0.9663	0.3609	0.9663	0.9830
No log	5.25	42	0.9604	0.4247	0.9604	0.9800
No log	5.5	44	0.9367	0.4751	0.9367	0.9678
No log	5.75	46	0.8828	0.4727	0.8828	0.9396
No log	6.0	48	0.8632	0.4564	0.8632	0.9291
No log	6.25	50	0.8867	0.4633	0.8867	0.9416
No log	6.5	52	0.8952	0.4394	0.8952	0.9462
No log	6.75	54	0.8654	0.4633	0.8654	0.9303
No log	7.0	56	0.8256	0.4654	0.8256	0.9086
No log	7.25	58	0.7991	0.4841	0.7991	0.8939
No log	7.5	60	0.7876	0.5058	0.7876	0.8875
No log	7.75	62	0.7948	0.5271	0.7948	0.8915
No log	8.0	64	0.8171	0.4777	0.8171	0.9039
No log	8.25	66	0.8463	0.4782	0.8463	0.9199
No log	8.5	68	0.8408	0.4776	0.8408	0.9169
No log	8.75	70	0.8277	0.4975	0.8277	0.9098
No log	9.0	72	0.8082	0.5192	0.8082	0.8990
No log	9.25	74	0.7930	0.5586	0.7930	0.8905
No log	9.5	76	0.7851	0.5572	0.7851	0.8861
No log	9.75	78	0.7797	0.5572	0.7797	0.8830
No log	10.0	80	0.7776	0.5621	0.7776	0.8818

Framework versions

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