arabert_cross_organization_task3_fold6

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: 0.6744
• Qwk: 0.5374
• Mse: 0.6727

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	2.0484	0.0727	2.0464
No log	0.2353	4	1.0229	0.2634	1.0209
No log	0.3529	6	1.0565	0.3738	1.0563
No log	0.4706	8	0.8346	0.5570	0.8340
No log	0.5882	10	0.7369	0.3583	0.7363
No log	0.7059	12	0.6299	0.3918	0.6296
No log	0.8235	14	0.5123	0.5734	0.5122
No log	0.9412	16	0.4760	0.6338	0.4759
No log	1.0588	18	0.4963	0.5827	0.4956
No log	1.1765	20	0.5522	0.5578	0.5507
No log	1.2941	22	0.5052	0.6935	0.5043
No log	1.4118	24	0.4977	0.6903	0.4967
No log	1.5294	26	0.5405	0.5739	0.5389
No log	1.6471	28	0.5155	0.5351	0.5142
No log	1.7647	30	0.4597	0.6960	0.4592
No log	1.8824	32	0.5096	0.7559	0.5095
No log	2.0	34	0.4755	0.7285	0.4750
No log	2.1176	36	0.4913	0.5583	0.4902
No log	2.2353	38	0.5379	0.5453	0.5365
No log	2.3529	40	0.4972	0.5887	0.4960
No log	2.4706	42	0.4876	0.6288	0.4866
No log	2.5882	44	0.4834	0.6174	0.4824
No log	2.7059	46	0.4806	0.6244	0.4797
No log	2.8235	48	0.5129	0.5775	0.5117
No log	2.9412	50	0.5356	0.5568	0.5343
No log	3.0588	52	0.5206	0.5719	0.5195
No log	3.1765	54	0.4824	0.6328	0.4817
No log	3.2941	56	0.4890	0.6817	0.4886
No log	3.4118	58	0.4872	0.6139	0.4865
No log	3.5294	60	0.5427	0.5620	0.5416
No log	3.6471	62	0.5604	0.5617	0.5591
No log	3.7647	64	0.5186	0.5723	0.5175
No log	3.8824	66	0.5145	0.5822	0.5135
No log	4.0	68	0.5162	0.5904	0.5152
No log	4.1176	70	0.5201	0.5923	0.5191
No log	4.2353	72	0.5476	0.5839	0.5464
No log	4.3529	74	0.5528	0.5930	0.5517
No log	4.4706	76	0.5558	0.5946	0.5547
No log	4.5882	78	0.5638	0.5843	0.5626
No log	4.7059	80	0.5691	0.5607	0.5679
No log	4.8235	82	0.6117	0.5231	0.6104
No log	4.9412	84	0.5964	0.5445	0.5952
No log	5.0588	86	0.5870	0.5441	0.5858
No log	5.1765	88	0.5544	0.5918	0.5534
No log	5.2941	90	0.5807	0.5683	0.5796
No log	5.4118	92	0.6445	0.5409	0.6429
No log	5.5294	94	0.6914	0.5268	0.6896
No log	5.6471	96	0.6566	0.5829	0.6550
No log	5.7647	98	0.6232	0.5841	0.6218
No log	5.8824	100	0.6460	0.5480	0.6445
No log	6.0	102	0.7138	0.5185	0.7121
No log	6.1176	104	0.7136	0.4997	0.7120
No log	6.2353	106	0.6311	0.5377	0.6297
No log	6.3529	108	0.5676	0.5748	0.5666
No log	6.4706	110	0.5579	0.5804	0.5571
No log	6.5882	112	0.5941	0.5534	0.5929
No log	6.7059	114	0.6495	0.5405	0.6479
No log	6.8235	116	0.6811	0.5303	0.6793
No log	6.9412	118	0.6469	0.5504	0.6453
No log	7.0588	120	0.6031	0.5779	0.6018
No log	7.1765	122	0.6073	0.5800	0.6059
No log	7.2941	124	0.6215	0.5636	0.6200
No log	7.4118	126	0.6881	0.5394	0.6862
No log	7.5294	128	0.7415	0.5163	0.7395
No log	7.6471	130	0.7372	0.5194	0.7352
No log	7.7647	132	0.6878	0.5221	0.6860
No log	7.8824	134	0.6484	0.5397	0.6467
No log	8.0	136	0.6265	0.5443	0.6250
No log	8.1176	138	0.6341	0.5416	0.6325
No log	8.2353	140	0.6558	0.5401	0.6541
No log	8.3529	142	0.6638	0.5263	0.6621
No log	8.4706	144	0.6606	0.5378	0.6589
No log	8.5882	146	0.6408	0.5416	0.6392
No log	8.7059	148	0.6283	0.5514	0.6267
No log	8.8235	150	0.6390	0.5462	0.6374
No log	8.9412	152	0.6538	0.5432	0.6522
No log	9.0588	154	0.6607	0.5432	0.6590
No log	9.1765	156	0.6707	0.5374	0.6690
No log	9.2941	158	0.6864	0.5207	0.6846
No log	9.4118	160	0.6949	0.5207	0.6931
No log	9.5294	162	0.6950	0.5207	0.6932
No log	9.6471	164	0.6884	0.5221	0.6866
No log	9.7647	166	0.6794	0.5336	0.6777
No log	9.8824	168	0.6756	0.5374	0.6739
No log	10.0	170	0.6744	0.5374	0.6727

Framework versions

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