ArabicNewSplits4_WithDuplicationsForScore5_FineTuningAraBERT_run2_AugV5_k3_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: 1.0328
• Qwk: 0.1642
• Mse: 1.0328
• Rmse: 1.0163

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	0.125	2	3.3159	-0.0227	3.3159	1.8210
No log	0.25	4	1.6913	-0.0070	1.6913	1.3005
No log	0.375	6	1.8024	0.0374	1.8024	1.3426
No log	0.5	8	0.9080	0.1718	0.9080	0.9529
No log	0.625	10	0.5996	0.0388	0.5996	0.7744
No log	0.75	12	0.6387	-0.0732	0.6387	0.7992
No log	0.875	14	0.5943	0.0	0.5943	0.7709
No log	1.0	16	0.7112	0.0164	0.7112	0.8433
No log	1.125	18	1.0124	0.0947	1.0124	1.0062
No log	1.25	20	0.7972	0.1238	0.7972	0.8929
No log	1.375	22	0.6313	-0.0909	0.6313	0.7946
No log	1.5	24	0.6036	-0.0159	0.6036	0.7769
No log	1.625	26	0.6379	-0.0081	0.6379	0.7987
No log	1.75	28	0.7582	0.0	0.7582	0.8708
No log	1.875	30	0.7252	0.0500	0.7252	0.8516
No log	2.0	32	1.0343	0.1417	1.0343	1.0170
No log	2.125	34	1.4572	0.0270	1.4572	1.2071
No log	2.25	36	0.9572	0.1652	0.9572	0.9784
No log	2.375	38	0.8041	-0.0612	0.8041	0.8967
No log	2.5	40	0.9649	-0.1589	0.9649	0.9823
No log	2.625	42	0.8933	-0.0110	0.8933	0.9452
No log	2.75	44	0.7702	0.0838	0.7702	0.8776
No log	2.875	46	1.0450	0.1562	1.0450	1.0222
No log	3.0	48	1.0477	0.1621	1.0477	1.0236
No log	3.125	50	0.8029	0.2239	0.8029	0.8960
No log	3.25	52	0.9038	-0.0679	0.9038	0.9507
No log	3.375	54	1.0109	0.0	1.0109	1.0055
No log	3.5	56	0.9513	-0.1004	0.9513	0.9753
No log	3.625	58	1.0204	0.0279	1.0204	1.0102
No log	3.75	60	1.3325	0.0141	1.3325	1.1544
No log	3.875	62	1.3734	0.0141	1.3734	1.1719
No log	4.0	64	1.0512	0.0745	1.0512	1.0253
No log	4.125	66	0.8772	0.1736	0.8772	0.9366
No log	4.25	68	0.9995	0.1803	0.9995	0.9997
No log	4.375	70	1.2347	0.0625	1.2347	1.1112
No log	4.5	72	1.5833	0.1233	1.5833	1.2583
No log	4.625	74	1.6077	0.1900	1.6077	1.2679
No log	4.75	76	1.1020	0.1331	1.1020	1.0498
No log	4.875	78	0.9202	0.2771	0.9202	0.9593
No log	5.0	80	0.9036	0.2627	0.9036	0.9506
No log	5.125	82	0.8128	0.3247	0.8128	0.9015
No log	5.25	84	1.2873	0.1429	1.2873	1.1346
No log	5.375	86	1.7899	0.1496	1.7899	1.3379
No log	5.5	88	1.6550	0.1097	1.6550	1.2865
No log	5.625	90	1.1294	0.1128	1.1294	1.0627
No log	5.75	92	0.7064	0.2986	0.7064	0.8405
No log	5.875	94	0.7812	0.3306	0.7812	0.8839
No log	6.0	96	0.7578	0.3036	0.7578	0.8705
No log	6.125	98	0.6867	0.2963	0.6867	0.8287
No log	6.25	100	0.9551	0.1333	0.9551	0.9773
No log	6.375	102	1.3773	0.1601	1.3773	1.1736
No log	6.5	104	1.4379	0.1572	1.4379	1.1991
No log	6.625	106	1.1783	0.1471	1.1783	1.0855
No log	6.75	108	0.8736	0.1928	0.8736	0.9347
No log	6.875	110	0.7875	0.2536	0.7875	0.8874
No log	7.0	112	0.7409	0.2605	0.7409	0.8608
No log	7.125	114	0.7791	0.2605	0.7791	0.8827
No log	7.25	116	0.8687	0.2199	0.8687	0.9320
No log	7.375	118	1.1443	0.1886	1.1443	1.0697
No log	7.5	120	1.6665	0.1591	1.6665	1.2909
No log	7.625	122	1.9510	0.0824	1.9510	1.3968
No log	7.75	124	1.9432	0.0826	1.9432	1.3940
No log	7.875	126	1.7346	0.1534	1.7346	1.3170
No log	8.0	128	1.3980	0.2152	1.3980	1.1824
No log	8.125	130	1.0814	0.1655	1.0814	1.0399
No log	8.25	132	0.9614	0.2000	0.9614	0.9805
No log	8.375	134	0.9384	0.2424	0.9384	0.9687
No log	8.5	136	0.9148	0.2180	0.9148	0.9565
No log	8.625	138	0.9172	0.2195	0.9172	0.9577
No log	8.75	140	0.9989	0.1818	0.9989	0.9994
No log	8.875	142	1.0920	0.2000	1.0920	1.0450
No log	9.0	144	1.1518	0.2119	1.1518	1.0732
No log	9.125	146	1.1822	0.1938	1.1822	1.0873
No log	9.25	148	1.1950	0.1882	1.1950	1.0932
No log	9.375	150	1.1834	0.1938	1.1834	1.0878
No log	9.5	152	1.1538	0.1938	1.1538	1.0742
No log	9.625	154	1.1085	0.1937	1.1085	1.0528
No log	9.75	156	1.0666	0.1698	1.0666	1.0328
No log	9.875	158	1.0414	0.1642	1.0414	1.0205
No log	10.0	160	1.0328	0.1642	1.0328	1.0163

Framework versions

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