ArabicNewSplits6_FineTuningAraBERT_run2_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: 0.6617
• Qwk: 0.7184
• Mse: 0.6617
• Rmse: 0.8135

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	5.4245	-0.0372	5.4245	2.3291
No log	0.2857	4	3.2814	0.0524	3.2814	1.8115
No log	0.4286	6	2.5494	-0.0764	2.5494	1.5967
No log	0.5714	8	1.6103	0.0918	1.6103	1.2690
No log	0.7143	10	1.5124	0.0959	1.5124	1.2298
No log	0.8571	12	1.2520	0.2613	1.2520	1.1189
No log	1.0	14	1.0725	0.3174	1.0725	1.0356
No log	1.1429	16	1.0385	0.3169	1.0385	1.0190
No log	1.2857	18	1.1912	0.3234	1.1912	1.0914
No log	1.4286	20	1.3396	0.1632	1.3396	1.1574
No log	1.5714	22	1.3607	0.1991	1.3607	1.1665
No log	1.7143	24	1.0569	0.4200	1.0569	1.0281
No log	1.8571	26	0.9703	0.4842	0.9703	0.9850
No log	2.0	28	1.0249	0.5093	1.0249	1.0124
No log	2.1429	30	1.2604	0.4347	1.2604	1.1227
No log	2.2857	32	1.8025	0.3041	1.8025	1.3426
No log	2.4286	34	2.1072	0.2432	2.1072	1.4516
No log	2.5714	36	1.6835	0.3399	1.6835	1.2975
No log	2.7143	38	1.0130	0.5597	1.0130	1.0065
No log	2.8571	40	0.7435	0.6594	0.7435	0.8623
No log	3.0	42	0.6080	0.7223	0.6080	0.7797
No log	3.1429	44	0.6114	0.7246	0.6114	0.7819
No log	3.2857	46	0.6407	0.6929	0.6407	0.8004
No log	3.4286	48	1.1929	0.5156	1.1929	1.0922
No log	3.5714	50	1.6555	0.3659	1.6555	1.2866
No log	3.7143	52	1.4935	0.4346	1.4935	1.2221
No log	3.8571	54	1.1291	0.5491	1.1291	1.0626
No log	4.0	56	0.8338	0.6362	0.8338	0.9131
No log	4.1429	58	0.6677	0.6876	0.6677	0.8171
No log	4.2857	60	0.6033	0.6997	0.6033	0.7767
No log	4.4286	62	0.6082	0.7154	0.6082	0.7799
No log	4.5714	64	0.6083	0.7170	0.6083	0.7799
No log	4.7143	66	0.6049	0.7224	0.6049	0.7777
No log	4.8571	68	0.7334	0.6654	0.7334	0.8564
No log	5.0	70	1.0638	0.5532	1.0638	1.0314
No log	5.1429	72	1.1547	0.5501	1.1547	1.0746
No log	5.2857	74	0.9750	0.5963	0.9750	0.9874
No log	5.4286	76	0.7384	0.7064	0.7384	0.8593
No log	5.5714	78	0.6557	0.7310	0.6557	0.8097
No log	5.7143	80	0.6309	0.7634	0.6309	0.7943
No log	5.8571	82	0.6455	0.7113	0.6455	0.8035
No log	6.0	84	0.6472	0.7353	0.6472	0.8045
No log	6.1429	86	0.6657	0.7298	0.6657	0.8159
No log	6.2857	88	0.6992	0.7008	0.6992	0.8362
No log	6.4286	90	0.6816	0.7106	0.6816	0.8256
No log	6.5714	92	0.6752	0.7611	0.6752	0.8217
No log	6.7143	94	0.6778	0.7598	0.6778	0.8233
No log	6.8571	96	0.6880	0.7144	0.6880	0.8295
No log	7.0	98	0.7988	0.6730	0.7988	0.8937
No log	7.1429	100	0.8650	0.6478	0.8650	0.9301
No log	7.2857	102	0.8045	0.6667	0.8045	0.8969
No log	7.4286	104	0.7116	0.6881	0.7116	0.8436
No log	7.5714	106	0.6603	0.7580	0.6603	0.8126
No log	7.7143	108	0.6973	0.7116	0.6973	0.8350
No log	7.8571	110	0.7140	0.7033	0.7140	0.8450
No log	8.0	112	0.6969	0.6997	0.6969	0.8348
No log	8.1429	114	0.6647	0.7458	0.6647	0.8153
No log	8.2857	116	0.6387	0.7439	0.6387	0.7992
No log	8.4286	118	0.6677	0.7285	0.6677	0.8171
No log	8.5714	120	0.7559	0.6730	0.7559	0.8694
No log	8.7143	122	0.8227	0.6557	0.8227	0.9070
No log	8.8571	124	0.8234	0.6679	0.8234	0.9074
No log	9.0	126	0.8002	0.6576	0.8002	0.8945
No log	9.1429	128	0.7636	0.6551	0.7636	0.8738
No log	9.2857	130	0.7244	0.6833	0.7244	0.8511
No log	9.4286	132	0.6926	0.7054	0.6926	0.8323
No log	9.5714	134	0.6771	0.7290	0.6771	0.8229
No log	9.7143	136	0.6656	0.7227	0.6656	0.8159
No log	9.8571	138	0.6621	0.7184	0.6621	0.8137
No log	10.0	140	0.6617	0.7184	0.6617	0.8135

Framework versions

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