ArabicNewSplits6_WithDuplicationsForScore5_FineTuningAraBERT_run1_AugV5_k2_task2_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: 0.8679
• Qwk: 0.5828
• Mse: 0.8679
• Rmse: 0.9316

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	0.1429	2	3.7200	0.0073	3.7200	1.9287
No log	0.2857	4	1.9644	0.0821	1.9644	1.4016
No log	0.4286	6	1.2036	0.0787	1.2036	1.0971
No log	0.5714	8	0.8392	0.0130	0.8392	0.9161
No log	0.7143	10	0.7332	0.1494	0.7332	0.8562
No log	0.8571	12	0.7618	0.0863	0.7618	0.8728
No log	1.0	14	0.6707	0.2724	0.6707	0.8190
No log	1.1429	16	0.6400	0.2868	0.6400	0.8000
No log	1.2857	18	0.6400	0.2505	0.6400	0.8000
No log	1.4286	20	0.6786	0.3078	0.6786	0.8238
No log	1.5714	22	0.9533	0.2154	0.9533	0.9764
No log	1.7143	24	1.0807	0.2714	1.0807	1.0396
No log	1.8571	26	0.6548	0.5017	0.6548	0.8092
No log	2.0	28	0.5996	0.4421	0.5996	0.7743
No log	2.1429	30	0.6687	0.5034	0.6687	0.8178
No log	2.2857	32	1.1880	0.3555	1.1880	1.0900
No log	2.4286	34	1.3139	0.3243	1.3139	1.1463
No log	2.5714	36	1.0831	0.3836	1.0831	1.0407
No log	2.7143	38	0.9094	0.5038	0.9094	0.9536
No log	2.8571	40	0.6657	0.5468	0.6657	0.8159
No log	3.0	42	0.6559	0.5144	0.6559	0.8099
No log	3.1429	44	0.5969	0.5723	0.5969	0.7726
No log	3.2857	46	0.5732	0.5976	0.5732	0.7571
No log	3.4286	48	0.5919	0.5628	0.5919	0.7693
No log	3.5714	50	0.5679	0.5890	0.5679	0.7536
No log	3.7143	52	0.6001	0.5869	0.6001	0.7747
No log	3.8571	54	0.6023	0.6022	0.6023	0.7761
No log	4.0	56	0.6880	0.5432	0.6880	0.8295
No log	4.1429	58	0.7666	0.4918	0.7666	0.8756
No log	4.2857	60	0.7508	0.4890	0.7508	0.8665
No log	4.4286	62	0.7143	0.5779	0.7143	0.8452
No log	4.5714	64	0.7018	0.6107	0.7018	0.8377
No log	4.7143	66	0.7275	0.5918	0.7275	0.8529
No log	4.8571	68	0.7517	0.6037	0.7517	0.8670
No log	5.0	70	0.7852	0.5933	0.7852	0.8861
No log	5.1429	72	0.8059	0.6087	0.8059	0.8977
No log	5.2857	74	0.8710	0.5308	0.8710	0.9333
No log	5.4286	76	0.8897	0.5315	0.8897	0.9432
No log	5.5714	78	0.8405	0.5887	0.8405	0.9168
No log	5.7143	80	0.8582	0.5560	0.8582	0.9264
No log	5.8571	82	0.9490	0.4928	0.9490	0.9741
No log	6.0	84	0.9479	0.5275	0.9479	0.9736
No log	6.1429	86	0.9132	0.5504	0.9132	0.9556
No log	6.2857	88	0.9020	0.5848	0.9020	0.9497
No log	6.4286	90	0.8956	0.5864	0.8956	0.9464
No log	6.5714	92	0.8842	0.5864	0.8842	0.9403
No log	6.7143	94	0.8793	0.5868	0.8793	0.9377
No log	6.8571	96	0.8751	0.5861	0.8751	0.9355
No log	7.0	98	0.8813	0.5849	0.8813	0.9388
No log	7.1429	100	0.8793	0.5831	0.8793	0.9377
No log	7.2857	102	0.8696	0.5687	0.8696	0.9325
No log	7.4286	104	0.8760	0.5687	0.8760	0.9359
No log	7.5714	106	0.8757	0.5605	0.8757	0.9358
No log	7.7143	108	0.8725	0.5803	0.8725	0.9341
No log	7.8571	110	0.8745	0.5834	0.8745	0.9351
No log	8.0	112	0.8787	0.5749	0.8787	0.9374
No log	8.1429	114	0.9028	0.5341	0.9028	0.9501
No log	8.2857	116	0.9326	0.5225	0.9326	0.9657
No log	8.4286	118	0.9317	0.5225	0.9317	0.9652
No log	8.5714	120	0.9022	0.5273	0.9022	0.9499
No log	8.7143	122	0.8772	0.5735	0.8772	0.9366
No log	8.8571	124	0.8626	0.5825	0.8626	0.9288
No log	9.0	126	0.8569	0.5930	0.8569	0.9257
No log	9.1429	128	0.8592	0.5558	0.8592	0.9269
No log	9.2857	130	0.8637	0.5655	0.8637	0.9293
No log	9.4286	132	0.8659	0.5930	0.8659	0.9305
No log	9.5714	134	0.8670	0.5887	0.8670	0.9311
No log	9.7143	136	0.8688	0.5828	0.8688	0.9321
No log	9.8571	138	0.8681	0.5939	0.8681	0.9317
No log	10.0	140	0.8679	0.5828	0.8679	0.9316

Framework versions

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