Arabic_FineTuningAraBERT_AugV0_k1_task5_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: 0.9025
• Qwk: 0.4904
• Mse: 0.9025
• Rmse: 0.9500

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	0.1538	2	4.8571	-0.0019	4.8571	2.2039
No log	0.3077	4	3.2406	0.0390	3.2406	1.8002
No log	0.4615	6	2.1366	0.1295	2.1366	1.4617
No log	0.6154	8	1.3807	0.1437	1.3807	1.1750
No log	0.7692	10	1.2109	0.0	1.2109	1.1004
No log	0.9231	12	1.1472	0.0734	1.1472	1.0711
No log	1.0769	14	1.0463	0.2657	1.0463	1.0229
No log	1.2308	16	1.0573	0.375	1.0573	1.0282
No log	1.3846	18	1.0933	0.4269	1.0933	1.0456
No log	1.5385	20	1.0469	0.3353	1.0469	1.0232
No log	1.6923	22	0.9741	0.2708	0.9741	0.9870
No log	1.8462	24	0.9136	0.2708	0.9136	0.9558
No log	2.0	26	0.8856	0.2279	0.8856	0.9411
No log	2.1538	28	0.8410	0.3980	0.8410	0.9171
No log	2.3077	30	0.8176	0.3980	0.8176	0.9042
No log	2.4615	32	0.7829	0.4334	0.7829	0.8848
No log	2.6154	34	0.7447	0.5399	0.7447	0.8630
No log	2.7692	36	0.7231	0.4357	0.7231	0.8503
No log	2.9231	38	0.7205	0.4357	0.7205	0.8488
No log	3.0769	40	0.7075	0.4516	0.7075	0.8412
No log	3.2308	42	0.7143	0.4059	0.7143	0.8452
No log	3.3846	44	0.6533	0.6078	0.6533	0.8083
No log	3.5385	46	0.7730	0.6247	0.7730	0.8792
No log	3.6923	48	0.8588	0.6073	0.8588	0.9267
No log	3.8462	50	0.8608	0.5929	0.8608	0.9278
No log	4.0	52	0.7368	0.6094	0.7368	0.8583
No log	4.1538	54	0.6310	0.5902	0.6310	0.7944
No log	4.3077	56	0.6050	0.6078	0.6050	0.7778
No log	4.4615	58	0.6029	0.6264	0.6029	0.7764
No log	4.6154	60	0.6909	0.5929	0.6909	0.8312
No log	4.7692	62	0.7828	0.625	0.7828	0.8848
No log	4.9231	64	0.7070	0.5396	0.7070	0.8408
No log	5.0769	66	0.6682	0.6800	0.6682	0.8174
No log	5.2308	68	0.6888	0.5929	0.6888	0.8299
No log	5.3846	70	0.7814	0.5929	0.7814	0.8840
No log	5.5385	72	0.8431	0.5929	0.8431	0.9182
No log	5.6923	74	0.8419	0.5972	0.8419	0.9175
No log	5.8462	76	0.7389	0.5052	0.7389	0.8596
No log	6.0	78	0.6461	0.5870	0.6461	0.8038
No log	6.1538	80	0.6413	0.6325	0.6413	0.8008
No log	6.3077	82	0.6631	0.6264	0.6631	0.8143
No log	6.4615	84	0.8114	0.5249	0.8114	0.9008
No log	6.6154	86	1.0179	0.5249	1.0179	1.0089
No log	6.7692	88	1.1803	0.5625	1.1803	1.0864
No log	6.9231	90	1.1846	0.5625	1.1846	1.0884
No log	7.0769	92	1.0459	0.4904	1.0459	1.0227
No log	7.2308	94	0.8653	0.5157	0.8653	0.9302
No log	7.3846	96	0.7346	0.6114	0.7346	0.8571
No log	7.5385	98	0.6802	0.6264	0.6802	0.8247
No log	7.6923	100	0.6912	0.6264	0.6912	0.8314
No log	7.8462	102	0.7606	0.5443	0.7606	0.8721
No log	8.0	104	0.9137	0.5157	0.9137	0.9559
No log	8.1538	106	1.0525	0.5249	1.0525	1.0259
No log	8.3077	108	1.1142	0.5625	1.1142	1.0556
No log	8.4615	110	1.1128	0.5625	1.1128	1.0549
No log	8.6154	112	1.0546	0.4904	1.0546	1.0270
No log	8.7692	114	0.9727	0.4904	0.9727	0.9863
No log	8.9231	116	0.9005	0.4904	0.9005	0.9490
No log	9.0769	118	0.8743	0.5249	0.8743	0.9350
No log	9.2308	120	0.8671	0.5249	0.8671	0.9312
No log	9.3846	122	0.8736	0.5249	0.8736	0.9347
No log	9.5385	124	0.8883	0.4904	0.8883	0.9425
No log	9.6923	126	0.8948	0.4904	0.8948	0.9460
No log	9.8462	128	0.9006	0.4904	0.9006	0.9490
No log	10.0	130	0.9025	0.4904	0.9025	0.9500

Framework versions

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