ArabicNewSplits6_FineTuningAraBERT_run3_AugV5_k6_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: 0.9686
• Qwk: 0.6243
• Mse: 0.9686
• Rmse: 0.9842

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	0.08	2	2.3493	0.0370	2.3493	1.5328
No log	0.16	4	1.5551	0.1802	1.5551	1.2470
No log	0.24	6	1.6260	0.0161	1.6260	1.2751
No log	0.32	8	1.7163	0.0050	1.7163	1.3101
No log	0.4	10	1.5886	0.2037	1.5886	1.2604
No log	0.48	12	1.4815	0.1949	1.4815	1.2172
No log	0.56	14	1.3441	0.1093	1.3441	1.1594
No log	0.64	16	1.3351	0.0999	1.3351	1.1555
No log	0.72	18	1.3321	0.1170	1.3321	1.1542
No log	0.8	20	1.3881	0.2159	1.3881	1.1782
No log	0.88	22	1.4184	0.3413	1.4184	1.1910
No log	0.96	24	1.3928	0.3325	1.3928	1.1802
No log	1.04	26	1.2830	0.2979	1.2830	1.1327
No log	1.12	28	1.2244	0.3538	1.2244	1.1065
No log	1.2	30	1.2486	0.3873	1.2486	1.1174
No log	1.28	32	1.3487	0.4069	1.3487	1.1613
No log	1.3600	34	1.4322	0.4090	1.4322	1.1967
No log	1.44	36	1.4281	0.3805	1.4281	1.1950
No log	1.52	38	1.3573	0.2768	1.3573	1.1650
No log	1.6	40	1.3254	0.1364	1.3254	1.1513
No log	1.6800	42	1.3595	0.1891	1.3595	1.1660
No log	1.76	44	1.3296	0.1897	1.3296	1.1531
No log	1.8400	46	1.2899	0.1522	1.2899	1.1358
No log	1.92	48	1.2966	0.1704	1.2966	1.1387
No log	2.0	50	1.3443	0.3051	1.3443	1.1594
No log	2.08	52	1.4733	0.3791	1.4733	1.2138
No log	2.16	54	1.6648	0.3289	1.6648	1.2903
No log	2.24	56	1.7885	0.3085	1.7885	1.3373
No log	2.32	58	1.6929	0.3219	1.6929	1.3011
No log	2.4	60	1.4630	0.3898	1.4630	1.2095
No log	2.48	62	1.2450	0.3681	1.2450	1.1158
No log	2.56	64	1.1192	0.3999	1.1192	1.0579
No log	2.64	66	1.0737	0.4038	1.0737	1.0362
No log	2.7200	68	1.0466	0.4365	1.0466	1.0231
No log	2.8	70	1.0287	0.4857	1.0287	1.0143
No log	2.88	72	0.9831	0.4652	0.9831	0.9915
No log	2.96	74	0.9508	0.4881	0.9508	0.9751
No log	3.04	76	0.9547	0.4948	0.9547	0.9771
No log	3.12	78	1.0516	0.5279	1.0516	1.0255
No log	3.2	80	1.1540	0.5426	1.1540	1.0742
No log	3.2800	82	1.1962	0.5290	1.1962	1.0937
No log	3.36	84	1.2387	0.5226	1.2387	1.1129
No log	3.44	86	1.2126	0.5239	1.2126	1.1012
No log	3.52	88	1.1418	0.5317	1.1418	1.0686
No log	3.6	90	1.1152	0.5405	1.1152	1.0560
No log	3.68	92	1.1724	0.5487	1.1724	1.0828
No log	3.76	94	1.1657	0.5219	1.1657	1.0797
No log	3.84	96	1.0374	0.5827	1.0374	1.0185
No log	3.92	98	0.8888	0.6608	0.8888	0.9428
No log	4.0	100	0.7992	0.6471	0.7992	0.8940
No log	4.08	102	0.7700	0.6480	0.7700	0.8775
No log	4.16	104	0.7662	0.6616	0.7662	0.8753
No log	4.24	106	0.7692	0.6621	0.7692	0.8770
No log	4.32	108	0.8309	0.6859	0.8309	0.9115
No log	4.4	110	1.0234	0.6121	1.0234	1.0116
No log	4.48	112	1.2095	0.5809	1.2095	1.0998
No log	4.5600	114	1.3919	0.5513	1.3919	1.1798
No log	4.64	116	1.4710	0.5456	1.4710	1.2129
No log	4.72	118	1.4604	0.5226	1.4604	1.2085
No log	4.8	120	1.3758	0.5394	1.3758	1.1729
No log	4.88	122	1.3076	0.5343	1.3076	1.1435
No log	4.96	124	1.2204	0.5290	1.2204	1.1047
No log	5.04	126	1.1880	0.5626	1.1880	1.0899
No log	5.12	128	1.1493	0.5818	1.1493	1.0721
No log	5.2	130	1.0096	0.6320	1.0096	1.0048
No log	5.28	132	0.9188	0.6421	0.9188	0.9585
No log	5.36	134	0.8962	0.6444	0.8962	0.9467
No log	5.44	136	0.8967	0.6578	0.8967	0.9469
No log	5.52	138	0.8971	0.6753	0.8971	0.9471
No log	5.6	140	0.8943	0.6651	0.8943	0.9457
No log	5.68	142	0.9062	0.6437	0.9062	0.9520
No log	5.76	144	0.9470	0.6478	0.9470	0.9731
No log	5.84	146	0.9849	0.6416	0.9849	0.9924
No log	5.92	148	1.0016	0.6393	1.0016	1.0008
No log	6.0	150	1.0410	0.6405	1.0410	1.0203
No log	6.08	152	1.0452	0.6359	1.0452	1.0223
No log	6.16	154	1.0508	0.6251	1.0508	1.0251
No log	6.24	156	1.0041	0.5947	1.0041	1.0021
No log	6.32	158	0.9746	0.6099	0.9746	0.9872
No log	6.4	160	0.9729	0.5870	0.9729	0.9863
No log	6.48	162	1.0027	0.5644	1.0027	1.0013
No log	6.5600	164	1.0319	0.5807	1.0319	1.0158
No log	6.64	166	1.0675	0.5746	1.0675	1.0332
No log	6.72	168	1.1025	0.5428	1.1025	1.0500
No log	6.8	170	1.1161	0.5496	1.1161	1.0565
No log	6.88	172	1.0783	0.5681	1.0783	1.0384
No log	6.96	174	1.0211	0.5602	1.0211	1.0105
No log	7.04	176	0.9904	0.5746	0.9904	0.9952
No log	7.12	178	0.9840	0.5830	0.9840	0.9919
No log	7.2	180	1.0033	0.5780	1.0033	1.0016
No log	7.28	182	0.9997	0.5838	0.9997	0.9998
No log	7.36	184	1.0196	0.5688	1.0196	1.0098
No log	7.44	186	1.0552	0.5659	1.0552	1.0272
No log	7.52	188	1.0936	0.5726	1.0936	1.0458
No log	7.6	190	1.0777	0.5671	1.0777	1.0381
No log	7.68	192	1.0737	0.5671	1.0737	1.0362
No log	7.76	194	1.0575	0.5833	1.0575	1.0283
No log	7.84	196	1.0281	0.5768	1.0281	1.0140
No log	7.92	198	1.0080	0.5803	1.0080	1.0040
No log	8.0	200	1.0034	0.5712	1.0034	1.0017
No log	8.08	202	0.9963	0.5723	0.9963	0.9982
No log	8.16	204	0.9926	0.5723	0.9926	0.9963
No log	8.24	206	1.0054	0.5712	1.0054	1.0027
No log	8.32	208	1.0053	0.5803	1.0053	1.0026
No log	8.4	210	0.9911	0.5928	0.9911	0.9955
No log	8.48	212	0.9766	0.6207	0.9766	0.9882
No log	8.56	214	0.9653	0.6300	0.9653	0.9825
No log	8.64	216	0.9485	0.6465	0.9485	0.9739
No log	8.72	218	0.9328	0.6480	0.9328	0.9658
No log	8.8	220	0.9373	0.6421	0.9373	0.9681
No log	8.88	222	0.9389	0.6421	0.9389	0.9689
No log	8.96	224	0.9343	0.6421	0.9343	0.9666
No log	9.04	226	0.9218	0.6421	0.9218	0.9601
No log	9.12	228	0.9147	0.6374	0.9147	0.9564
No log	9.2	230	0.9120	0.6374	0.9120	0.9550
No log	9.28	232	0.9140	0.6295	0.9140	0.9560
No log	9.36	234	0.9157	0.6280	0.9157	0.9569
No log	9.44	236	0.9226	0.6266	0.9226	0.9605
No log	9.52	238	0.9281	0.6266	0.9281	0.9634
No log	9.6	240	0.9337	0.6330	0.9337	0.9663
No log	9.68	242	0.9406	0.6207	0.9406	0.9698
No log	9.76	244	0.9503	0.6207	0.9503	0.9748
No log	9.84	246	0.9608	0.6286	0.9608	0.9802
No log	9.92	248	0.9670	0.6243	0.9670	0.9834
No log	10.0	250	0.9686	0.6243	0.9686	0.9842

Framework versions

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