arabert_cross_organization_task7_fold3

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: 1.2477
• Qwk: 0.0397
• Mse: 1.2477

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: 64
• eval_batch_size: 64
• 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
No log	0.1176	2	4.6630	-0.0214	4.6630
No log	0.2353	4	1.9131	-0.0296	1.9131
No log	0.3529	6	1.4779	0.0333	1.4779
No log	0.4706	8	1.3898	-0.0031	1.3898
No log	0.5882	10	1.1632	-0.0182	1.1632
No log	0.7059	12	1.2046	0.0046	1.2046
No log	0.8235	14	1.1480	-0.0723	1.1480
No log	0.9412	16	1.1436	-0.0110	1.1436
No log	1.0588	18	1.1619	-0.0110	1.1619
No log	1.1765	20	1.1482	-0.0110	1.1482
No log	1.2941	22	1.1202	-0.0110	1.1202
No log	1.4118	24	1.1170	0.0062	1.1170
No log	1.5294	26	1.1174	0.0413	1.1174
No log	1.6471	28	1.1486	-0.0248	1.1486
No log	1.7647	30	1.2697	0.0	1.2697
No log	1.8824	32	1.2498	-0.0147	1.2498
No log	2.0	34	1.1692	-0.0248	1.1692
No log	2.1176	36	1.1913	-0.0203	1.1913
No log	2.2353	38	1.2360	-0.0255	1.2360
No log	2.3529	40	1.1562	-0.0046	1.1562
No log	2.4706	42	1.1715	-0.0074	1.1715
No log	2.5882	44	1.2562	-0.0147	1.2562
No log	2.7059	46	1.1883	-0.0039	1.1883
No log	2.8235	48	1.1458	0.0518	1.1458
No log	2.9412	50	1.2475	-0.0289	1.2475
No log	3.0588	52	1.3234	0.0059	1.3234
No log	3.1765	54	1.1838	-0.0002	1.1838
No log	3.2941	56	1.1405	0.0753	1.1405
No log	3.4118	58	1.1817	0.0686	1.1817
No log	3.5294	60	1.1663	0.0753	1.1663
No log	3.6471	62	1.1574	0.0944	1.1574
No log	3.7647	64	1.1895	0.0474	1.1895
No log	3.8824	66	1.1771	0.0160	1.1771
No log	4.0	68	1.1811	0.0725	1.1811
No log	4.1176	70	1.2361	0.0469	1.2361
No log	4.2353	72	1.2386	0.0650	1.2386
No log	4.3529	74	1.1740	0.1183	1.1740
No log	4.4706	76	1.1578	0.1142	1.1578
No log	4.5882	78	1.1478	0.0927	1.1478
No log	4.7059	80	1.1556	0.0890	1.1556
No log	4.8235	82	1.1743	0.0682	1.1743
No log	4.9412	84	1.1621	0.0682	1.1621
No log	5.0588	86	1.1329	0.0963	1.1329
No log	5.1765	88	1.1322	0.1113	1.1322
No log	5.2941	90	1.1440	0.1234	1.1440
No log	5.4118	92	1.2118	0.0376	1.2118
No log	5.5294	94	1.2862	0.0713	1.2862
No log	5.6471	96	1.3217	0.0599	1.3217
No log	5.7647	98	1.2121	0.0624	1.2121
No log	5.8824	100	1.1592	0.1501	1.1592
No log	6.0	102	1.1912	0.0385	1.1912
No log	6.1176	104	1.1673	0.0828	1.1673
No log	6.2353	106	1.1644	0.0448	1.1644
No log	6.3529	108	1.2438	0.0629	1.2438
No log	6.4706	110	1.3083	0.0622	1.3083
No log	6.5882	112	1.2970	0.0470	1.2970
No log	6.7059	114	1.2160	0.0295	1.2160
No log	6.8235	116	1.1682	0.0881	1.1682
No log	6.9412	118	1.1659	0.0898	1.1659
No log	7.0588	120	1.1694	0.1029	1.1694
No log	7.1765	122	1.1813	0.0829	1.1813
No log	7.2941	124	1.1840	0.0847	1.1840
No log	7.4118	126	1.1941	0.0847	1.1941
No log	7.5294	128	1.2296	0.0915	1.2296
No log	7.6471	130	1.2731	0.0561	1.2731
No log	7.7647	132	1.2863	0.0680	1.2863
No log	7.8824	134	1.2644	0.0391	1.2644
No log	8.0	136	1.2471	0.0771	1.2471
No log	8.1176	138	1.2423	0.0789	1.2423
No log	8.2353	140	1.2401	0.0596	1.2401
No log	8.3529	142	1.2475	0.0596	1.2475
No log	8.4706	144	1.2539	0.0703	1.2539
No log	8.5882	146	1.2593	0.0794	1.2593
No log	8.7059	148	1.2770	0.0551	1.2770
No log	8.8235	150	1.2937	0.0858	1.2937
No log	8.9412	152	1.2970	0.0840	1.2970
No log	9.0588	154	1.3005	0.0657	1.3005
No log	9.1765	156	1.2860	0.0657	1.2860
No log	9.2941	158	1.2799	0.0355	1.2799
No log	9.4118	160	1.2697	0.0407	1.2697
No log	9.5294	162	1.2616	0.0653	1.2616
No log	9.6471	164	1.2564	0.0374	1.2564
No log	9.7647	166	1.2527	0.0397	1.2527
No log	9.8824	168	1.2493	0.0397	1.2493
No log	10.0	170	1.2477	0.0397	1.2477

Framework versions

• Transformers 4.44.0
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1