ArabicNewSplits6_FineTuningAraBERT_run3_AugV5_k5_task3_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: 0.9088
• Qwk: 0.1790
• Mse: 0.9088
• Rmse: 0.9533

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	3.3872	0.0026	3.3872	1.8404
No log	0.16	4	1.7993	-0.0070	1.7993	1.3414
No log	0.24	6	1.2147	0.0588	1.2147	1.1021
No log	0.32	8	0.7356	0.1416	0.7356	0.8577
No log	0.4	10	0.6227	0.1304	0.6227	0.7891
No log	0.48	12	0.6534	0.0815	0.6534	0.8083
No log	0.56	14	0.6920	0.1373	0.6920	0.8319
No log	0.64	16	0.7350	0.1323	0.7350	0.8573
No log	0.72	18	0.7085	0.0877	0.7085	0.8417
No log	0.8	20	0.6613	0.0	0.6613	0.8132
No log	0.88	22	0.6234	0.0071	0.6234	0.7896
No log	0.96	24	0.6264	-0.0081	0.6264	0.7914
No log	1.04	26	0.6281	-0.0853	0.6281	0.7925
No log	1.12	28	0.7102	-0.0222	0.7102	0.8427
No log	1.2	30	0.8220	0.1351	0.8220	0.9067
No log	1.28	32	0.7936	0.1930	0.7936	0.8908
No log	1.3600	34	0.6148	0.2381	0.6148	0.7841
No log	1.44	36	0.6252	0.0720	0.6252	0.7907
No log	1.52	38	0.6023	0.0720	0.6023	0.7761
No log	1.6	40	0.5197	-0.0303	0.5197	0.7209
No log	1.6800	42	0.5246	0.2201	0.5246	0.7243
No log	1.76	44	0.5722	0.2184	0.5722	0.7564
No log	1.8400	46	0.5219	0.1579	0.5219	0.7224
No log	1.92	48	0.6583	0.2381	0.6583	0.8113
No log	2.0	50	1.1257	0.0069	1.1257	1.0610
No log	2.08	52	1.1725	0.0102	1.1725	1.0828
No log	2.16	54	0.9033	0.2188	0.9033	0.9504
No log	2.24	56	0.5960	0.1304	0.5960	0.7720
No log	2.32	58	0.8722	0.2137	0.8722	0.9339
No log	2.4	60	0.9445	0.1276	0.9445	0.9718
No log	2.48	62	0.7353	0.2157	0.7353	0.8575
No log	2.56	64	0.5989	0.0256	0.5989	0.7739
No log	2.64	66	0.6218	-0.0127	0.6218	0.7885
No log	2.7200	68	0.5868	0.0256	0.5868	0.7660
No log	2.8	70	0.5950	0.2289	0.5950	0.7714
No log	2.88	72	0.5883	0.2527	0.5883	0.7670
No log	2.96	74	0.7004	0.2079	0.7004	0.8369
No log	3.04	76	0.7479	0.1402	0.7479	0.8648
No log	3.12	78	0.6851	0.1230	0.6851	0.8277
No log	3.2	80	0.6233	0.2457	0.6233	0.7895
No log	3.2800	82	0.6247	0.2179	0.6247	0.7904
No log	3.36	84	0.7588	0.2150	0.7588	0.8711
No log	3.44	86	0.7546	0.2153	0.7546	0.8687
No log	3.52	88	0.6423	0.2000	0.6423	0.8015
No log	3.6	90	0.6347	0.2370	0.6347	0.7967
No log	3.68	92	0.6277	0.2527	0.6277	0.7922
No log	3.76	94	0.8327	0.2068	0.8327	0.9126
No log	3.84	96	0.9809	0.1815	0.9809	0.9904
No log	3.92	98	0.9608	0.2615	0.9608	0.9802
No log	4.0	100	1.1073	0.2061	1.1073	1.0523
No log	4.08	102	1.4608	0.1746	1.4608	1.2086
No log	4.16	104	1.2485	0.1678	1.2485	1.1174
No log	4.24	106	0.8060	0.3929	0.8060	0.8978
No log	4.32	108	0.6928	0.3524	0.6928	0.8323
No log	4.4	110	0.7744	0.3171	0.7744	0.8800
No log	4.48	112	0.8828	0.2333	0.8828	0.9396
No log	4.5600	114	0.9625	0.1698	0.9625	0.9811
No log	4.64	116	0.9359	0.2604	0.9359	0.9674
No log	4.72	118	0.6973	0.3021	0.6973	0.8350
No log	4.8	120	0.6168	0.3917	0.6168	0.7854
No log	4.88	122	0.6039	0.3607	0.6039	0.7771
No log	4.96	124	0.5950	0.3333	0.5950	0.7714
No log	5.04	126	0.6302	0.4093	0.6302	0.7939
No log	5.12	128	0.9153	0.2432	0.9153	0.9567
No log	5.2	130	1.0260	0.1880	1.0260	1.0129
No log	5.28	132	0.7631	0.1811	0.7631	0.8736
No log	5.36	134	0.5236	0.3797	0.5236	0.7236
No log	5.44	136	0.5360	0.3333	0.5360	0.7321
No log	5.52	138	0.6518	0.3571	0.6518	0.8073
No log	5.6	140	1.0122	0.2117	1.0122	1.0061
No log	5.68	142	1.2657	0.2107	1.2657	1.1250
No log	5.76	144	1.1952	0.1837	1.1952	1.0933
No log	5.84	146	0.8045	0.3000	0.8045	0.8969
No log	5.92	148	0.5986	0.3488	0.5986	0.7737
No log	6.0	150	0.6020	0.3488	0.6020	0.7759
No log	6.08	152	0.7802	0.2356	0.7802	0.8833
No log	6.16	154	1.1998	0.1831	1.1998	1.0954
No log	6.24	156	1.1639	0.1831	1.1639	1.0788
No log	6.32	158	0.8233	0.2423	0.8233	0.9074
No log	6.4	160	0.6395	0.3267	0.6395	0.7997
No log	6.48	162	0.6197	0.28	0.6197	0.7872
No log	6.5600	164	0.6929	0.2323	0.6929	0.8324
No log	6.64	166	0.8094	0.2074	0.8094	0.8997
No log	6.72	168	0.8983	0.2069	0.8983	0.9478
No log	6.8	170	0.8588	0.2432	0.8588	0.9267
No log	6.88	172	0.8420	0.1765	0.8420	0.9176
No log	6.96	174	0.8347	0.1765	0.8347	0.9136
No log	7.04	176	0.8139	0.2000	0.8139	0.9022
No log	7.12	178	0.7661	0.2233	0.7661	0.8753
No log	7.2	180	0.8665	0.2074	0.8665	0.9309
No log	7.28	182	1.1062	0.1506	1.1062	1.0518
No log	7.36	184	1.3659	0.1608	1.3659	1.1687
No log	7.44	186	1.3378	0.1608	1.3378	1.1566
No log	7.52	188	1.1216	0.1515	1.1216	1.0591
No log	7.6	190	0.8646	0.1416	0.8646	0.9298
No log	7.68	192	0.7690	0.2315	0.7690	0.8769
No log	7.76	194	0.7407	0.2315	0.7407	0.8606
No log	7.84	196	0.7892	0.2308	0.7892	0.8884
No log	7.92	198	0.9043	0.1790	0.9043	0.9509
No log	8.0	200	1.0429	0.1235	1.0429	1.0212
No log	8.08	202	1.0401	0.1235	1.0401	1.0198
No log	8.16	204	0.9624	0.1464	0.9624	0.9810
No log	8.24	206	0.8869	0.1416	0.8869	0.9417
No log	8.32	208	0.8372	0.2153	0.8372	0.9150
No log	8.4	210	0.8403	0.2153	0.8403	0.9167
No log	8.48	212	0.8314	0.2153	0.8314	0.9118
No log	8.56	214	0.8359	0.2153	0.8359	0.9143
No log	8.64	216	0.8246	0.2157	0.8246	0.9081
No log	8.72	218	0.8347	0.2077	0.8347	0.9136
No log	8.8	220	0.8324	0.2077	0.8324	0.9124
No log	8.88	222	0.8764	0.1776	0.8764	0.9362
No log	8.96	224	0.8851	0.1776	0.8851	0.9408
No log	9.04	226	0.8651	0.1776	0.8651	0.9301
No log	9.12	228	0.8401	0.2153	0.8401	0.9166
No log	9.2	230	0.8281	0.2744	0.8281	0.9100
No log	9.28	232	0.8253	0.2744	0.8253	0.9085
No log	9.36	234	0.8516	0.2153	0.8516	0.9228
No log	9.44	236	0.8890	0.2146	0.8890	0.9429
No log	9.52	238	0.9129	0.1795	0.9129	0.9555
No log	9.6	240	0.9328	0.1795	0.9328	0.9658
No log	9.68	242	0.9290	0.1795	0.9290	0.9638
No log	9.76	244	0.9231	0.1795	0.9231	0.9608
No log	9.84	246	0.9177	0.1795	0.9177	0.9580
No log	9.92	248	0.9107	0.1790	0.9107	0.9543
No log	10.0	250	0.9088	0.1790	0.9088	0.9533

Framework versions

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