Arabic_FineTuningAraBERT_AugV5_k2_task3_organization_fold1

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: 0.6027
• Qwk: 0.0222
• Mse: 0.6027
• Rmse: 0.7763

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	3.6943	0.0	3.6943	1.9220
No log	0.2857	4	1.8695	0.0	1.8695	1.3673
No log	0.4286	6	1.0940	0.0	1.0940	1.0459
No log	0.5714	8	0.7397	0.3654	0.7397	0.8601
No log	0.7143	10	0.6617	-0.0233	0.6617	0.8135
No log	0.8571	12	0.6396	-0.0233	0.6396	0.7997
No log	1.0	14	0.7214	-0.0577	0.7214	0.8493
No log	1.1429	16	0.8774	0.2143	0.8774	0.9367
No log	1.2857	18	0.7807	0.0984	0.7807	0.8836
No log	1.4286	20	0.6395	0.0	0.6395	0.7997
No log	1.5714	22	0.6691	0.0	0.6691	0.8180
No log	1.7143	24	0.8455	-0.0916	0.8455	0.9195
No log	1.8571	26	1.2566	0.0	1.2566	1.1210
No log	2.0	28	1.9493	0.0	1.9493	1.3962
No log	2.1429	30	1.9458	0.0	1.9458	1.3949
No log	2.2857	32	1.4166	0.0	1.4166	1.1902
No log	2.4286	34	1.0109	0.0	1.0109	1.0054
No log	2.5714	36	0.7919	0.2143	0.7919	0.8899
No log	2.7143	38	0.8311	0.1879	0.8311	0.9117
No log	2.8571	40	0.9516	0.0	0.9516	0.9755
No log	3.0	42	0.9142	0.0	0.9142	0.9562
No log	3.1429	44	0.7883	0.0763	0.7883	0.8879
No log	3.2857	46	0.8340	0.0571	0.8340	0.9132
No log	3.4286	48	0.8201	0.0571	0.8201	0.9056
No log	3.5714	50	0.7335	0.1239	0.7335	0.8565
No log	3.7143	52	0.6410	0.0	0.6410	0.8006
No log	3.8571	54	0.6221	0.0	0.6221	0.7887
No log	4.0	56	0.6607	0.0	0.6607	0.8128
No log	4.1429	58	0.8841	-0.0820	0.8841	0.9403
No log	4.2857	60	0.9114	-0.0820	0.9114	0.9547
No log	4.4286	62	0.7034	-0.0421	0.7034	0.8387
No log	4.5714	64	0.6314	0.0	0.6314	0.7946
No log	4.7143	66	0.5949	0.0	0.5949	0.7713
No log	4.8571	68	0.5761	0.0	0.5761	0.7590
No log	5.0	70	0.5803	0.0	0.5803	0.7618
No log	5.1429	72	0.6058	0.1895	0.6058	0.7783
No log	5.2857	74	0.5757	0.1895	0.5757	0.7588
No log	5.4286	76	0.5368	0.0	0.5368	0.7327
No log	5.5714	78	0.5330	0.0	0.5330	0.7301
No log	5.7143	80	0.5448	0.0	0.5448	0.7381
No log	5.8571	82	0.5658	0.0	0.5658	0.7522
No log	6.0	84	0.6706	0.0	0.6706	0.8189
No log	6.1429	86	0.8092	0.0222	0.8092	0.8996
No log	6.2857	88	0.8005	0.0222	0.8005	0.8947
No log	6.4286	90	0.6768	0.0222	0.6768	0.8227
No log	6.5714	92	0.5966	0.0	0.5966	0.7724
No log	6.7143	94	0.6005	-0.0233	0.6005	0.7749
No log	6.8571	96	0.5849	-0.0233	0.5849	0.7648
No log	7.0	98	0.5442	0.0	0.5442	0.7377
No log	7.1429	100	0.5926	0.0222	0.5926	0.7698
No log	7.2857	102	0.5723	0.0222	0.5723	0.7565
No log	7.4286	104	0.5054	0.0	0.5054	0.7109
No log	7.5714	106	0.4898	0.0	0.4898	0.6999
No log	7.7143	108	0.5001	0.0	0.5001	0.7072
No log	7.8571	110	0.5214	-0.0421	0.5214	0.7221
No log	8.0	112	0.5205	0.0	0.5205	0.7215
No log	8.1429	114	0.5150	0.0	0.5150	0.7177
No log	8.2857	116	0.5367	0.2667	0.5367	0.7326
No log	8.4286	118	0.6132	0.2414	0.6132	0.7830
No log	8.5714	120	0.6227	0.2326	0.6227	0.7891
No log	8.7143	122	0.6027	0.2524	0.6027	0.7764
No log	8.8571	124	0.5833	0.0222	0.5833	0.7637
No log	9.0	126	0.5827	0.0222	0.5827	0.7634
No log	9.1429	128	0.5799	0.0	0.5799	0.7615
No log	9.2857	130	0.5873	0.0	0.5873	0.7664
No log	9.4286	132	0.5931	0.0222	0.5931	0.7701
No log	9.5714	134	0.5958	0.0222	0.5958	0.7719
No log	9.7143	136	0.5963	0.0222	0.5963	0.7722
No log	9.8571	138	0.6013	0.0222	0.6013	0.7754
No log	10.0	140	0.6027	0.0222	0.6027	0.7763

Framework versions

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