ArabicNewSplits6_FineTuningAraBERT_run1_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: 0.8963
• Qwk: 0.1736
• Mse: 0.8963
• Rmse: 0.9467

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	3.3350	-0.0258	3.3350	1.8262
No log	0.16	4	1.7722	-0.0101	1.7722	1.3312
No log	0.24	6	1.2167	0.0255	1.2167	1.1030
No log	0.32	8	1.6673	0.0390	1.6673	1.2912
No log	0.4	10	2.3463	0.0793	2.3463	1.5318
No log	0.48	12	1.1652	0.0	1.1652	1.0794
No log	0.56	14	0.6792	0.1724	0.6792	0.8242
No log	0.64	16	0.5965	0.0	0.5965	0.7723
No log	0.72	18	0.5670	0.0	0.5670	0.7530
No log	0.8	20	0.5668	-0.0081	0.5668	0.7529
No log	0.88	22	0.6409	0.0	0.6409	0.8005
No log	0.96	24	0.6173	0.0222	0.6173	0.7857
No log	1.04	26	0.7052	0.0409	0.7052	0.8397
No log	1.12	28	0.7815	0.0145	0.7815	0.8840
No log	1.2	30	0.6089	0.2418	0.6089	0.7803
No log	1.28	32	0.5907	0.1385	0.5907	0.7686
No log	1.3600	34	0.7185	0.2485	0.7185	0.8476
No log	1.44	36	0.6760	0.2000	0.6760	0.8222
No log	1.52	38	0.5871	0.2308	0.5871	0.7662
No log	1.6	40	0.9571	0.1290	0.9571	0.9783
No log	1.6800	42	0.7634	0.1238	0.7634	0.8737
No log	1.76	44	0.6060	0.2889	0.6060	0.7785
No log	1.8400	46	0.6206	0.2516	0.6206	0.7878
No log	1.92	48	0.6676	0.0983	0.6676	0.8171
No log	2.0	50	0.6673	0.2821	0.6673	0.8169
No log	2.08	52	0.8527	-0.0595	0.8527	0.9234
No log	2.16	54	0.8476	-0.0909	0.8476	0.9206
No log	2.24	56	0.5771	0.2632	0.5771	0.7597
No log	2.32	58	0.6328	0.0108	0.6328	0.7955
No log	2.4	60	0.6786	0.0359	0.6786	0.8238
No log	2.48	62	0.7128	0.0680	0.7128	0.8443
No log	2.56	64	0.5723	0.4556	0.5723	0.7565
No log	2.64	66	0.7020	0.1746	0.7020	0.8378
No log	2.7200	68	0.6541	0.3224	0.6541	0.8088
No log	2.8	70	0.6134	0.2184	0.6134	0.7832
No log	2.88	72	0.7513	0.0578	0.7513	0.8668
No log	2.96	74	0.6110	0.2258	0.6110	0.7816
No log	3.04	76	0.6326	0.2967	0.6326	0.7953
No log	3.12	78	0.7566	0.1429	0.7566	0.8698
No log	3.2	80	0.5921	0.4409	0.5921	0.7695
No log	3.2800	82	0.5527	0.3478	0.5527	0.7434
No log	3.36	84	0.5819	0.3478	0.5819	0.7628
No log	3.44	86	0.8029	0.0654	0.8029	0.8960
No log	3.52	88	1.1760	0.0178	1.1760	1.0844
No log	3.6	90	1.0855	0.0790	1.0855	1.0419
No log	3.68	92	0.7676	0.3188	0.7676	0.8761
No log	3.76	94	0.6331	0.3299	0.6331	0.7957
No log	3.84	96	0.6438	0.3462	0.6438	0.8023
No log	3.92	98	0.9316	0.1278	0.9316	0.9652
No log	4.0	100	1.0614	0.0831	1.0614	1.0303
No log	4.08	102	0.8179	0.1525	0.8179	0.9044
No log	4.16	104	0.6224	0.3951	0.6224	0.7889
No log	4.24	106	0.6481	0.4010	0.6481	0.8051
No log	4.32	108	0.8782	0.1220	0.8782	0.9371
No log	4.4	110	1.0105	0.1014	1.0105	1.0052
No log	4.48	112	1.0695	0.1014	1.0695	1.0342
No log	4.5600	114	0.9170	0.1939	0.9170	0.9576
No log	4.64	116	0.9819	0.0882	0.9819	0.9909
No log	4.72	118	0.9113	0.2569	0.9113	0.9546
No log	4.8	120	0.8898	0.2903	0.8898	0.9433
No log	4.88	122	1.0287	0.1241	1.0287	1.0143
No log	4.96	124	0.9667	0.2126	0.9667	0.9832
No log	5.04	126	0.7501	0.3514	0.7501	0.8661
No log	5.12	128	0.8012	0.1861	0.8012	0.8951
No log	5.2	130	0.8178	0.1930	0.8178	0.9043
No log	5.28	132	1.0378	0.1331	1.0378	1.0187
No log	5.36	134	1.1549	0.0530	1.1549	1.0747
No log	5.44	136	1.0429	0.0769	1.0429	1.0212
No log	5.52	138	0.7929	0.1928	0.7929	0.8905
No log	5.6	140	0.7470	0.2830	0.7470	0.8643
No log	5.68	142	0.8859	0.1795	0.8859	0.9412
No log	5.76	144	1.1328	0.1111	1.1328	1.0644
No log	5.84	146	1.0759	0.0791	1.0759	1.0372
No log	5.92	148	0.8187	0.1705	0.8187	0.9048
No log	6.0	150	0.7301	0.3433	0.7301	0.8545
No log	6.08	152	0.6831	0.3548	0.6831	0.8265
No log	6.16	154	0.7160	0.3433	0.7160	0.8461
No log	6.24	156	0.8565	0.1392	0.8565	0.9255
No log	6.32	158	0.8876	0.1169	0.8876	0.9421
No log	6.4	160	0.7593	0.3103	0.7593	0.8714
No log	6.48	162	0.7757	0.2692	0.7757	0.8808
No log	6.5600	164	0.8268	0.2364	0.8268	0.9093
No log	6.64	166	0.7779	0.2692	0.7779	0.8820
No log	6.72	168	0.8085	0.2294	0.8085	0.8992
No log	6.8	170	0.8454	0.2146	0.8454	0.9195
No log	6.88	172	0.9151	0.0598	0.9151	0.9566
No log	6.96	174	0.8822	0.0569	0.8822	0.9393
No log	7.04	176	0.7859	0.2830	0.7859	0.8865
No log	7.12	178	0.7581	0.2300	0.7581	0.8707
No log	7.2	180	0.7590	0.2692	0.7590	0.8712
No log	7.28	182	0.7170	0.2692	0.7170	0.8468
No log	7.36	184	0.7227	0.2692	0.7227	0.8501
No log	7.44	186	0.7160	0.2692	0.7160	0.8461
No log	7.52	188	0.7300	0.2780	0.7300	0.8544
No log	7.6	190	0.8455	0.1799	0.8455	0.9195
No log	7.68	192	0.8673	0.1148	0.8673	0.9313
No log	7.76	194	0.7691	0.2830	0.7691	0.8770
No log	7.84	196	0.6987	0.2986	0.6987	0.8359
No log	7.92	198	0.6945	0.3301	0.6945	0.8334
No log	8.0	200	0.7564	0.2579	0.7564	0.8697
No log	8.08	202	0.8433	0.1261	0.8433	0.9183
No log	8.16	204	0.9240	0.1197	0.9240	0.9612
No log	8.24	206	0.9431	0.1524	0.9431	0.9712
No log	8.32	208	0.8929	0.2066	0.8929	0.9449
No log	8.4	210	0.8194	0.1930	0.8194	0.9052
No log	8.48	212	0.7849	0.2857	0.7849	0.8859
No log	8.56	214	0.7856	0.2857	0.7856	0.8863
No log	8.64	216	0.8527	0.1931	0.8527	0.9234
No log	8.72	218	0.9015	0.1385	0.9015	0.9494
No log	8.8	220	0.9844	0.1765	0.9844	0.9922
No log	8.88	222	1.0516	0.1254	1.0516	1.0255
No log	8.96	224	1.0236	0.1524	1.0236	1.0117
No log	9.04	226	0.9811	0.1765	0.9811	0.9905
No log	9.12	228	0.8925	0.0720	0.8925	0.9447
No log	9.2	230	0.8111	0.2838	0.8111	0.9006
No log	9.28	232	0.7786	0.2877	0.7786	0.8824
No log	9.36	234	0.7815	0.2877	0.7815	0.8840
No log	9.44	236	0.8107	0.2364	0.8107	0.9004
No log	9.52	238	0.8441	0.2000	0.8441	0.9188
No log	9.6	240	0.8898	0.2068	0.8898	0.9433
No log	9.68	242	0.9103	0.1756	0.9103	0.9541
No log	9.76	244	0.9111	0.1756	0.9111	0.9545
No log	9.84	246	0.9030	0.1756	0.9030	0.9503
No log	9.92	248	0.8976	0.1736	0.8976	0.9474
No log	10.0	250	0.8963	0.1736	0.8963	0.9467

Framework versions

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