ArabicNewSplits6_FineTuningAraBERT_run1_AugV5_k2_task1_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: 0.6918
• Qwk: 0.6958
• Mse: 0.6918
• Rmse: 0.8318

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	5.1061	-0.0138	5.1061	2.2597
No log	0.2857	4	3.1407	0.0858	3.1407	1.7722
No log	0.4286	6	1.8978	0.1009	1.8978	1.3776
No log	0.5714	8	1.3167	0.1674	1.3167	1.1475
No log	0.7143	10	1.1906	0.2488	1.1906	1.0912
No log	0.8571	12	1.1125	0.2869	1.1125	1.0547
No log	1.0	14	1.0966	0.3074	1.0966	1.0472
No log	1.1429	16	1.0989	0.3445	1.0989	1.0483
No log	1.2857	18	1.0679	0.2803	1.0679	1.0334
No log	1.4286	20	1.0197	0.4091	1.0197	1.0098
No log	1.5714	22	1.2912	0.2804	1.2912	1.1363
No log	1.7143	24	1.9199	0.2007	1.9199	1.3856
No log	1.8571	26	2.7598	0.1298	2.7598	1.6613
No log	2.0	28	2.7077	0.1400	2.7077	1.6455
No log	2.1429	30	2.2747	0.2296	2.2747	1.5082
No log	2.2857	32	2.0854	0.2328	2.0854	1.4441
No log	2.4286	34	1.4886	0.2590	1.4886	1.2201
No log	2.5714	36	1.0672	0.4117	1.0672	1.0331
No log	2.7143	38	0.8814	0.5341	0.8814	0.9388
No log	2.8571	40	0.8292	0.5862	0.8292	0.9106
No log	3.0	42	0.7970	0.5862	0.7970	0.8928
No log	3.1429	44	0.8337	0.5434	0.8337	0.9131
No log	3.2857	46	0.8998	0.5571	0.8998	0.9486
No log	3.4286	48	0.9492	0.5087	0.9492	0.9743
No log	3.5714	50	1.1011	0.3785	1.1011	1.0494
No log	3.7143	52	1.1879	0.3857	1.1879	1.0899
No log	3.8571	54	1.1193	0.4450	1.1193	1.0580
No log	4.0	56	1.0735	0.4884	1.0735	1.0361
No log	4.1429	58	1.0039	0.5644	1.0039	1.0020
No log	4.2857	60	0.7687	0.6539	0.7687	0.8767
No log	4.4286	62	0.7227	0.6980	0.7227	0.8501
No log	4.5714	64	0.7652	0.6896	0.7652	0.8748
No log	4.7143	66	0.8730	0.6679	0.8730	0.9343
No log	4.8571	68	1.0787	0.5931	1.0787	1.0386
No log	5.0	70	1.1515	0.5476	1.1515	1.0731
No log	5.1429	72	0.9738	0.6238	0.9738	0.9868
No log	5.2857	74	0.7218	0.7032	0.7218	0.8496
No log	5.4286	76	0.5996	0.7590	0.5996	0.7744
No log	5.5714	78	0.5855	0.7611	0.5855	0.7652
No log	5.7143	80	0.5689	0.7540	0.5689	0.7542
No log	5.8571	82	0.5807	0.7248	0.5807	0.7620
No log	6.0	84	0.7101	0.6710	0.7101	0.8427
No log	6.1429	86	0.8375	0.6685	0.8375	0.9152
No log	6.2857	88	0.8033	0.6593	0.8033	0.8963
No log	6.4286	90	0.7043	0.6761	0.7043	0.8392
No log	6.5714	92	0.6226	0.7097	0.6226	0.7890
No log	6.7143	94	0.5537	0.7377	0.5537	0.7441
No log	6.8571	96	0.5292	0.7562	0.5292	0.7275
No log	7.0	98	0.5270	0.7659	0.5270	0.7259
No log	7.1429	100	0.5330	0.7570	0.5330	0.7301
No log	7.2857	102	0.5333	0.7696	0.5333	0.7303
No log	7.4286	104	0.5310	0.7697	0.5310	0.7287
No log	7.5714	106	0.5416	0.7593	0.5416	0.7360
No log	7.7143	108	0.5823	0.7199	0.5823	0.7631
No log	7.8571	110	0.6304	0.6995	0.6304	0.7940
No log	8.0	112	0.6924	0.7109	0.6924	0.8321
No log	8.1429	114	0.7001	0.7034	0.7001	0.8367
No log	8.2857	116	0.7007	0.7034	0.7007	0.8371
No log	8.4286	118	0.6754	0.6988	0.6754	0.8218
No log	8.5714	120	0.6436	0.6982	0.6436	0.8022
No log	8.7143	122	0.6456	0.6982	0.6456	0.8035
No log	8.8571	124	0.6606	0.6988	0.6606	0.8127
No log	9.0	126	0.6775	0.6915	0.6775	0.8231
No log	9.1429	128	0.6926	0.6958	0.6926	0.8322
No log	9.2857	130	0.7038	0.6958	0.7038	0.8389
No log	9.4286	132	0.7096	0.6958	0.7096	0.8424
No log	9.5714	134	0.7031	0.6958	0.7031	0.8385
No log	9.7143	136	0.6982	0.6958	0.6982	0.8356
No log	9.8571	138	0.6936	0.6958	0.6936	0.8328
No log	10.0	140	0.6918	0.6958	0.6918	0.8318

Framework versions

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