ArabicNewSplits5_FineTuningAraBERT_run1_AugV5_k2_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: 0.9955
• Qwk: 0.5628
• Mse: 0.9955
• Rmse: 0.9977

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	0.1538	2	2.3340	0.0239	2.3340	1.5277
No log	0.3077	4	1.5660	0.1626	1.5660	1.2514
No log	0.4615	6	1.3957	0.1733	1.3957	1.1814
No log	0.6154	8	1.3934	0.1431	1.3934	1.1804
No log	0.7692	10	1.4455	0.2037	1.4455	1.2023
No log	0.9231	12	1.4430	0.3615	1.4430	1.2013
No log	1.0769	14	1.3241	0.2111	1.3241	1.1507
No log	1.2308	16	1.2938	0.1857	1.2938	1.1374
No log	1.3846	18	1.2459	0.2065	1.2459	1.1162
No log	1.5385	20	1.2525	0.3311	1.2525	1.1191
No log	1.6923	22	1.3500	0.3772	1.3500	1.1619
No log	1.8462	24	1.2912	0.3867	1.2912	1.1363
No log	2.0	26	1.1539	0.2488	1.1539	1.0742
No log	2.1538	28	1.1282	0.3289	1.1282	1.0622
No log	2.3077	30	1.1093	0.3647	1.1093	1.0532
No log	2.4615	32	1.0772	0.3857	1.0772	1.0379
No log	2.6154	34	1.1012	0.3306	1.1012	1.0494
No log	2.7692	36	1.1784	0.3925	1.1784	1.0856
No log	2.9231	38	1.2863	0.4563	1.2863	1.1342
No log	3.0769	40	1.2390	0.3928	1.2390	1.1131
No log	3.2308	42	1.2090	0.3988	1.2090	1.0995
No log	3.3846	44	1.2211	0.4251	1.2211	1.1050
No log	3.5385	46	1.2530	0.4436	1.2530	1.1194
No log	3.6923	48	1.2175	0.4354	1.2175	1.1034
No log	3.8462	50	1.2157	0.4443	1.2157	1.1026
No log	4.0	52	1.1911	0.4859	1.1911	1.0914
No log	4.1538	54	1.1271	0.4865	1.1271	1.0616
No log	4.3077	56	1.1143	0.5196	1.1143	1.0556
No log	4.4615	58	1.0647	0.5065	1.0647	1.0318
No log	4.6154	60	0.9777	0.5325	0.9777	0.9888
No log	4.7692	62	0.8584	0.5783	0.8584	0.9265
No log	4.9231	64	0.8560	0.6025	0.8560	0.9252
No log	5.0769	66	0.9153	0.5824	0.9153	0.9567
No log	5.2308	68	0.9862	0.5397	0.9862	0.9931
No log	5.3846	70	0.9592	0.5438	0.9592	0.9794
No log	5.5385	72	0.9504	0.5758	0.9504	0.9749
No log	5.6923	74	0.8845	0.5994	0.8845	0.9405
No log	5.8462	76	0.8677	0.5852	0.8677	0.9315
No log	6.0	78	0.9170	0.6113	0.9170	0.9576
No log	6.1538	80	1.0059	0.5985	1.0059	1.0029
No log	6.3077	82	0.9865	0.6101	0.9865	0.9932
No log	6.4615	84	0.8991	0.6236	0.8991	0.9482
No log	6.6154	86	0.8650	0.6094	0.8650	0.9301
No log	6.7692	88	0.8586	0.6266	0.8586	0.9266
No log	6.9231	90	0.9127	0.6375	0.9127	0.9553
No log	7.0769	92	0.9775	0.5816	0.9775	0.9887
No log	7.2308	94	1.0493	0.5714	1.0493	1.0243
No log	7.3846	96	1.1051	0.5658	1.1051	1.0513
No log	7.5385	98	1.1066	0.5703	1.1066	1.0519
No log	7.6923	100	1.0728	0.5792	1.0728	1.0358
No log	7.8462	102	1.0382	0.5802	1.0382	1.0189
No log	8.0	104	1.0285	0.5802	1.0285	1.0142
No log	8.1538	106	1.0019	0.5792	1.0019	1.0010
No log	8.3077	108	0.9958	0.5792	0.9958	0.9979
No log	8.4615	110	0.9830	0.5628	0.9830	0.9915
No log	8.6154	112	0.9869	0.5628	0.9869	0.9934
No log	8.7692	114	1.0006	0.5664	1.0006	1.0003
No log	8.9231	116	1.0036	0.5664	1.0036	1.0018
No log	9.0769	118	1.0195	0.5642	1.0195	1.0097
No log	9.2308	120	1.0369	0.5736	1.0369	1.0183
No log	9.3846	122	1.0336	0.5631	1.0336	1.0167
No log	9.5385	124	1.0125	0.5642	1.0125	1.0062
No log	9.6923	126	1.0019	0.5688	1.0019	1.0010
No log	9.8462	128	0.9986	0.5664	0.9986	0.9993
No log	10.0	130	0.9955	0.5628	0.9955	0.9977

Framework versions

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