arabert_cross_organization_task1_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: 0.5605
• Qwk: 0.5728
• Mse: 0.5597

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	0.1333	2	2.6067	0.0127	2.6095
No log	0.2667	4	1.3449	0.1252	1.3444
No log	0.4	6	0.8578	0.3924	0.8566
No log	0.5333	8	0.9106	0.4859	0.9097
No log	0.6667	10	0.8748	0.2420	0.8739
No log	0.8	12	0.7294	0.3596	0.7288
No log	0.9333	14	0.5690	0.6415	0.5689
No log	1.0667	16	0.5434	0.5530	0.5434
No log	1.2	18	0.5471	0.5275	0.5466
No log	1.3333	20	0.5550	0.5931	0.5542
No log	1.4667	22	0.6050	0.5930	0.6038
No log	1.6	24	0.5418	0.6626	0.5409
No log	1.7333	26	0.5460	0.5657	0.5451
No log	1.8667	28	0.5935	0.4982	0.5926
No log	2.0	30	0.5427	0.5075	0.5420
No log	2.1333	32	0.4531	0.6187	0.4527
No log	2.2667	34	0.4419	0.6495	0.4414
No log	2.4	36	0.4419	0.6978	0.4415
No log	2.5333	38	0.4760	0.6296	0.4752
No log	2.6667	40	0.6868	0.4906	0.6857
No log	2.8	42	0.7124	0.4820	0.7111
No log	2.9333	44	0.5222	0.5761	0.5212
No log	3.0667	46	0.4482	0.7104	0.4476
No log	3.2	48	0.4458	0.6877	0.4451
No log	3.3333	50	0.4617	0.6149	0.4608
No log	3.4667	52	0.4966	0.5426	0.4956
No log	3.6	54	0.4450	0.6238	0.4441
No log	3.7333	56	0.4262	0.6657	0.4254
No log	3.8667	58	0.4293	0.6880	0.4287
No log	4.0	60	0.4457	0.6871	0.4450
No log	4.1333	62	0.4834	0.6153	0.4825
No log	4.2667	64	0.4860	0.6114	0.4851
No log	4.4	66	0.4688	0.6743	0.4681
No log	4.5333	68	0.4821	0.6484	0.4814
No log	4.6667	70	0.5849	0.5461	0.5839
No log	4.8	72	0.6875	0.4677	0.6865
No log	4.9333	74	0.6204	0.5060	0.6195
No log	5.0667	76	0.4953	0.6000	0.4944
No log	5.2	78	0.4472	0.6794	0.4465
No log	5.3333	80	0.4447	0.6714	0.4438
No log	5.4667	82	0.5001	0.5886	0.4990
No log	5.6	84	0.5921	0.5277	0.5908
No log	5.7333	86	0.6040	0.5105	0.6027
No log	5.8667	88	0.5520	0.5654	0.5509
No log	6.0	90	0.4986	0.6296	0.4977
No log	6.1333	92	0.4842	0.6769	0.4835
No log	6.2667	94	0.4892	0.6687	0.4885
No log	6.4	96	0.5316	0.5863	0.5305
No log	6.5333	98	0.6215	0.5289	0.6203
No log	6.6667	100	0.6345	0.5159	0.6334
No log	6.8	102	0.6012	0.5377	0.6002
No log	6.9333	104	0.5566	0.5601	0.5558
No log	7.0667	106	0.5366	0.5820	0.5359
No log	7.2	108	0.5404	0.5733	0.5396
No log	7.3333	110	0.5546	0.5683	0.5539
No log	7.4667	112	0.5313	0.5733	0.5306
No log	7.6	114	0.5360	0.5716	0.5353
No log	7.7333	116	0.5598	0.5575	0.5589
No log	7.8667	118	0.5884	0.5500	0.5875
No log	8.0	120	0.6065	0.5413	0.6055
No log	8.1333	122	0.6118	0.5428	0.6108
No log	8.2667	124	0.6220	0.5539	0.6210
No log	8.4	126	0.6154	0.5499	0.6145
No log	8.5333	128	0.6035	0.5508	0.6025
No log	8.6667	130	0.6093	0.5515	0.6083
No log	8.8	132	0.6055	0.5501	0.6046
No log	8.9333	134	0.5879	0.5569	0.5870
No log	9.0667	136	0.5763	0.5636	0.5754
No log	9.2	138	0.5630	0.5646	0.5622
No log	9.3333	140	0.5585	0.5852	0.5577
No log	9.4667	142	0.5527	0.5852	0.5519
No log	9.6	144	0.5526	0.5852	0.5518
No log	9.7333	146	0.5553	0.5745	0.5545
No log	9.8667	148	0.5592	0.5728	0.5584
No log	10.0	150	0.5605	0.5728	0.5597

Framework versions

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