arabert_cross_organization_task4_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: 0.5671
• Qwk: 0.6088
• Mse: 0.5665

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	0.1111	2	1.6087	0.0446	1.6095
No log	0.2222	4	1.0873	0.0325	1.0875
No log	0.3333	6	0.9052	0.3977	0.9045
No log	0.4444	8	0.7546	0.3836	0.7542
No log	0.5556	10	0.5626	0.4920	0.5622
No log	0.6667	12	0.5049	0.6220	0.5048
No log	0.7778	14	0.6046	0.5251	0.6033
No log	0.8889	16	0.4861	0.6302	0.4861
No log	1.0	18	0.6306	0.6759	0.6318
No log	1.1111	20	0.5794	0.7289	0.5803
No log	1.2222	22	0.4351	0.6154	0.4348
No log	1.3333	24	0.5139	0.5149	0.5133
No log	1.4444	26	0.4487	0.5967	0.4481
No log	1.5556	28	0.4505	0.7663	0.4507
No log	1.6667	30	0.4917	0.7702	0.4921
No log	1.7778	32	0.4432	0.6859	0.4426
No log	1.8889	34	0.5427	0.5673	0.5415
No log	2.0	36	0.5133	0.5536	0.5122
No log	2.1111	38	0.4171	0.6515	0.4166
No log	2.2222	40	0.4265	0.6951	0.4260
No log	2.3333	42	0.4487	0.6401	0.4480
No log	2.4444	44	0.5093	0.5826	0.5082
No log	2.5556	46	0.5193	0.5750	0.5182
No log	2.6667	48	0.4705	0.6472	0.4699
No log	2.7778	50	0.4668	0.7101	0.4666
No log	2.8889	52	0.4574	0.7064	0.4572
No log	3.0	54	0.4490	0.6031	0.4485
No log	3.1111	56	0.5005	0.5434	0.4998
No log	3.2222	58	0.4728	0.5714	0.4721
No log	3.3333	60	0.4444	0.6493	0.4440
No log	3.4444	62	0.4764	0.6914	0.4760
No log	3.5556	64	0.5124	0.6726	0.5118
No log	3.6667	66	0.5636	0.5952	0.5627
No log	3.7778	68	0.5479	0.5977	0.5469
No log	3.8889	70	0.5130	0.6191	0.5121
No log	4.0	72	0.4884	0.6461	0.4879
No log	4.1111	74	0.4839	0.5980	0.4832
No log	4.2222	76	0.5315	0.5457	0.5305
No log	4.3333	78	0.5336	0.5465	0.5326
No log	4.4444	80	0.4907	0.5701	0.4901
No log	4.5556	82	0.4836	0.6726	0.4834
No log	4.6667	84	0.5058	0.6880	0.5058
No log	4.7778	86	0.5131	0.6355	0.5127
No log	4.8889	88	0.6336	0.5424	0.6325
No log	5.0	90	0.7217	0.5167	0.7203
No log	5.1111	92	0.6484	0.5394	0.6473
No log	5.2222	94	0.5239	0.6047	0.5234
No log	5.3333	96	0.4960	0.6591	0.4957
No log	5.4444	98	0.4958	0.6725	0.4956
No log	5.5556	100	0.5131	0.6097	0.5126
No log	5.6667	102	0.5789	0.5640	0.5779
No log	5.7778	104	0.6393	0.5598	0.6380
No log	5.8889	106	0.6013	0.5640	0.6003
No log	6.0	108	0.5576	0.6085	0.5569
No log	6.1111	110	0.5394	0.6682	0.5389
No log	6.2222	112	0.5429	0.6569	0.5424
No log	6.3333	114	0.5841	0.5873	0.5832
No log	6.4444	116	0.6113	0.5642	0.6102
No log	6.5556	118	0.5828	0.5631	0.5818
No log	6.6667	120	0.5283	0.6190	0.5277
No log	6.7778	122	0.5137	0.6474	0.5132
No log	6.8889	124	0.5253	0.6135	0.5246
No log	7.0	126	0.5588	0.5693	0.5579
No log	7.1111	128	0.5741	0.5644	0.5732
No log	7.2222	130	0.5814	0.5627	0.5805
No log	7.3333	132	0.5510	0.5687	0.5502
No log	7.4444	134	0.5316	0.6227	0.5311
No log	7.5556	136	0.5345	0.6034	0.5339
No log	7.6667	138	0.5500	0.5709	0.5493
No log	7.7778	140	0.5557	0.5748	0.5551
No log	7.8889	142	0.5534	0.5855	0.5528
No log	8.0	144	0.5613	0.5748	0.5606
No log	8.1111	146	0.5663	0.5748	0.5657
No log	8.2222	148	0.5748	0.5709	0.5741
No log	8.3333	150	0.5887	0.5693	0.5880
No log	8.4444	152	0.5802	0.5858	0.5796
No log	8.5556	154	0.5690	0.6232	0.5685
No log	8.6667	156	0.5680	0.6356	0.5676
No log	8.7778	158	0.5770	0.6074	0.5764
No log	8.8889	160	0.5943	0.5640	0.5935
No log	9.0	162	0.6147	0.5503	0.6138
No log	9.1111	164	0.6207	0.5539	0.6198
No log	9.2222	166	0.6153	0.5503	0.6144
No log	9.3333	168	0.5994	0.5602	0.5986
No log	9.4444	170	0.5838	0.5640	0.5830
No log	9.5556	172	0.5750	0.5795	0.5744
No log	9.6667	174	0.5688	0.5982	0.5682
No log	9.7778	176	0.5665	0.6054	0.5659
No log	9.8889	178	0.5665	0.6054	0.5659
No log	10.0	180	0.5671	0.6088	0.5665

Framework versions

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