ArabicNewSplits4_WithDuplicationsForScore5_FineTuningAraBERT_run1_AugV5_k4_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: 0.7448
• Qwk: 0.2372
• Mse: 0.7448
• Rmse: 0.8630

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	0.1	2	3.2078	-0.0028	3.2078	1.7910
No log	0.2	4	1.5451	0.0255	1.5451	1.2430
No log	0.3	6	1.0837	0.0335	1.0837	1.0410
No log	0.4	8	0.7208	0.1398	0.7208	0.8490
No log	0.5	10	0.5880	0.0909	0.5880	0.7668
No log	0.6	12	0.6068	0.0534	0.6068	0.7790
No log	0.7	14	0.6055	0.1278	0.6055	0.7782
No log	0.8	16	0.5965	0.0222	0.5965	0.7723
No log	0.9	18	0.6167	0.0897	0.6167	0.7853
No log	1.0	20	0.8301	0.0476	0.8301	0.9111
No log	1.1	22	0.7052	0.0429	0.7052	0.8398
No log	1.2	24	0.9406	0.0857	0.9406	0.9699
No log	1.3	26	0.9606	0.1333	0.9606	0.9801
No log	1.4	28	0.8483	0.1030	0.8483	0.9210
No log	1.5	30	0.6657	0.0625	0.6657	0.8159
No log	1.6	32	0.7859	0.1765	0.7859	0.8865
No log	1.7	34	1.5105	0.0833	1.5105	1.2290
No log	1.8	36	1.3564	0.0833	1.3564	1.1646
No log	1.9	38	0.7619	0.1919	0.7619	0.8729
No log	2.0	40	0.5993	0.0365	0.5993	0.7741
No log	2.1	42	0.6835	0.0638	0.6835	0.8268
No log	2.2	44	0.7053	0.1233	0.7053	0.8398
No log	2.3	46	0.6824	0.1667	0.6824	0.8261
No log	2.4	48	0.6650	0.0452	0.6650	0.8155
No log	2.5	50	0.7971	0.1414	0.7971	0.8928
No log	2.6	52	0.6954	0.1628	0.6954	0.8339
No log	2.7	54	0.6811	0.1565	0.6811	0.8253
No log	2.8	56	0.6948	0.1467	0.6948	0.8336
No log	2.9	58	0.7502	0.2169	0.7502	0.8662
No log	3.0	60	0.6620	0.1373	0.6620	0.8136
No log	3.1	62	0.6828	0.1351	0.6828	0.8263
No log	3.2	64	0.7646	0.0899	0.7646	0.8744
No log	3.3	66	0.8254	0.0531	0.8254	0.9085
No log	3.4	68	0.8442	0.0495	0.8442	0.9188
No log	3.5	70	0.8740	0.0359	0.8740	0.9349
No log	3.6	72	0.9018	0.0189	0.9018	0.9496
No log	3.7	74	0.9268	0.1131	0.9268	0.9627
No log	3.8	76	1.0730	0.0367	1.0730	1.0359
No log	3.9	78	1.6611	0.0327	1.6611	1.2889
No log	4.0	80	1.7461	0.0096	1.7461	1.3214
No log	4.1	82	1.1547	0.1161	1.1547	1.0745
No log	4.2	84	0.8773	0.0493	0.8773	0.9366
No log	4.3	86	0.8552	0.0707	0.8552	0.9248
No log	4.4	88	1.1661	0.0938	1.1661	1.0798
No log	4.5	90	1.6476	0.0809	1.6476	1.2836
No log	4.6	92	1.4166	0.0704	1.4166	1.1902
No log	4.7	94	0.8426	0.1304	0.8426	0.9180
No log	4.8	96	0.7782	0.0980	0.7782	0.8821
No log	4.9	98	0.7502	0.1230	0.7502	0.8662
No log	5.0	100	0.8006	0.2086	0.8006	0.8948
No log	5.1	102	1.2550	0.0365	1.2550	1.1203
No log	5.2	104	1.6616	0.0881	1.6616	1.2890
No log	5.3	106	1.5159	0.1068	1.5159	1.2312
No log	5.4	108	0.9728	0.1790	0.9728	0.9863
No log	5.5	110	0.7998	0.2850	0.7998	0.8943
No log	5.6	112	0.7470	0.2811	0.7470	0.8643
No log	5.7	114	0.7517	0.2593	0.7517	0.8670
No log	5.8	116	0.8336	0.1790	0.8336	0.9130
No log	5.9	118	0.8887	0.1718	0.8887	0.9427
No log	6.0	120	0.8925	0.2356	0.8925	0.9447
No log	6.1	122	0.9367	0.1351	0.9367	0.9678
No log	6.2	124	0.9827	0.1366	0.9827	0.9913
No log	6.3	126	0.9027	0.1579	0.9027	0.9501
No log	6.4	128	0.8716	0.2667	0.8716	0.9336
No log	6.5	130	0.9416	0.1652	0.9416	0.9703
No log	6.6	132	0.9379	0.1652	0.9379	0.9685
No log	6.7	134	0.9617	0.1074	0.9617	0.9807
No log	6.8	136	0.8503	0.2320	0.8503	0.9221
No log	6.9	138	0.7571	0.3004	0.7571	0.8701
No log	7.0	140	0.7832	0.2775	0.7832	0.8850
No log	7.1	142	0.8964	0.1441	0.8964	0.9468
No log	7.2	144	0.8296	0.1652	0.8296	0.9109
No log	7.3	146	0.7365	0.3455	0.7365	0.8582
No log	7.4	148	0.7629	0.2857	0.7629	0.8735
No log	7.5	150	0.8673	0.2000	0.8673	0.9313
No log	7.6	152	0.8516	0.1660	0.8516	0.9228
No log	7.7	154	0.8878	0.1724	0.8878	0.9422
No log	7.8	156	0.8569	0.1724	0.8569	0.9257
No log	7.9	158	0.7572	0.25	0.7572	0.8702
No log	8.0	160	0.7601	0.2579	0.7601	0.8718
No log	8.1	162	0.8578	0.1724	0.8578	0.9262
No log	8.2	164	1.0199	0.1746	1.0199	1.0099
No log	8.3	166	1.0130	0.1486	1.0130	1.0065
No log	8.4	168	0.8936	0.1441	0.8936	0.9453
No log	8.5	170	0.8044	0.1724	0.8044	0.8969
No log	8.6	172	0.7369	0.3846	0.7369	0.8584
No log	8.7	174	0.7253	0.3846	0.7253	0.8516
No log	8.8	176	0.7371	0.3171	0.7371	0.8585
No log	8.9	178	0.7665	0.1781	0.7665	0.8755
No log	9.0	180	0.7725	0.1781	0.7725	0.8789
No log	9.1	182	0.7667	0.1712	0.7667	0.8756
No log	9.2	184	0.7450	0.2762	0.7450	0.8631
No log	9.3	186	0.7197	0.4286	0.7197	0.8483
No log	9.4	188	0.7204	0.4286	0.7204	0.8488
No log	9.5	190	0.7270	0.4286	0.7270	0.8527
No log	9.6	192	0.7397	0.3427	0.7397	0.8601
No log	9.7	194	0.7374	0.3427	0.7374	0.8587
No log	9.8	196	0.7419	0.3427	0.7419	0.8613
No log	9.9	198	0.7447	0.2372	0.7447	0.8629
No log	10.0	200	0.7448	0.2372	0.7448	0.8630

Framework versions

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