ArabicNewSplits4_WithDuplicationsForScore5_FineTuningAraBERT_run3_AugV5_k3_task5_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: 0.5785
• Qwk: 0.7315
• Mse: 0.5785
• Rmse: 0.7606

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	2.1790	0.0082	2.1790	1.4761
No log	0.3077	4	1.4728	0.2304	1.4728	1.2136
No log	0.4615	6	1.3261	0.1568	1.3261	1.1515
No log	0.6154	8	1.3689	0.3225	1.3689	1.1700
No log	0.7692	10	1.4251	0.3789	1.4251	1.1938
No log	0.9231	12	1.6681	0.3435	1.6681	1.2915
No log	1.0769	14	1.7121	0.3138	1.7121	1.3085
No log	1.2308	16	1.4946	0.3844	1.4946	1.2225
No log	1.3846	18	1.1910	0.3336	1.1910	1.0913
No log	1.5385	20	1.0646	0.3511	1.0646	1.0318
No log	1.6923	22	1.0145	0.4172	1.0145	1.0072
No log	1.8462	24	1.0180	0.4528	1.0180	1.0090
No log	2.0	26	0.9858	0.4854	0.9858	0.9929
No log	2.1538	28	0.9269	0.4982	0.9269	0.9628
No log	2.3077	30	0.8821	0.5256	0.8821	0.9392
No log	2.4615	32	0.9886	0.5449	0.9886	0.9943
No log	2.6154	34	1.0174	0.5458	1.0174	1.0087
No log	2.7692	36	0.8418	0.6027	0.8418	0.9175
No log	2.9231	38	0.7575	0.6294	0.7575	0.8704
No log	3.0769	40	0.7354	0.6338	0.7354	0.8576
No log	3.2308	42	0.7725	0.6616	0.7725	0.8789
No log	3.3846	44	0.7842	0.6727	0.7842	0.8855
No log	3.5385	46	0.8884	0.6334	0.8884	0.9425
No log	3.6923	48	0.8606	0.6633	0.8606	0.9277
No log	3.8462	50	0.8578	0.6761	0.8578	0.9262
No log	4.0	52	0.7491	0.6799	0.7491	0.8655
No log	4.1538	54	0.6989	0.7037	0.6989	0.8360
No log	4.3077	56	0.7438	0.6898	0.7438	0.8625
No log	4.4615	58	0.6932	0.7011	0.6932	0.8326
No log	4.6154	60	0.6159	0.6783	0.6159	0.7848
No log	4.7692	62	0.6287	0.6665	0.6287	0.7929
No log	4.9231	64	0.5938	0.7156	0.5938	0.7706
No log	5.0769	66	0.6806	0.7035	0.6806	0.8250
No log	5.2308	68	0.8431	0.7256	0.8431	0.9182
No log	5.3846	70	0.7860	0.7301	0.7860	0.8866
No log	5.5385	72	0.6063	0.7220	0.6063	0.7787
No log	5.6923	74	0.5688	0.7532	0.5688	0.7542
No log	5.8462	76	0.6087	0.7027	0.6087	0.7802
No log	6.0	78	0.5901	0.6974	0.5901	0.7681
No log	6.1538	80	0.5678	0.7134	0.5678	0.7535
No log	6.3077	82	0.6307	0.7337	0.6307	0.7942
No log	6.4615	84	0.6501	0.7388	0.6501	0.8063
No log	6.6154	86	0.5932	0.7485	0.5932	0.7702
No log	6.7692	88	0.5780	0.7360	0.5780	0.7603
No log	6.9231	90	0.6080	0.7460	0.6080	0.7798
No log	7.0769	92	0.6595	0.7484	0.6595	0.8121
No log	7.2308	94	0.6795	0.7551	0.6795	0.8243
No log	7.3846	96	0.6473	0.7435	0.6473	0.8046
No log	7.5385	98	0.5970	0.7233	0.5970	0.7727
No log	7.6923	100	0.5639	0.7284	0.5639	0.7510
No log	7.8462	102	0.5576	0.7434	0.5576	0.7467
No log	8.0	104	0.5578	0.7391	0.5578	0.7469
No log	8.1538	106	0.5573	0.7393	0.5573	0.7465
No log	8.3077	108	0.5689	0.7428	0.5689	0.7542
No log	8.4615	110	0.5992	0.7255	0.5992	0.7741
No log	8.6154	112	0.6319	0.7198	0.6319	0.7949
No log	8.7692	114	0.6253	0.7301	0.6253	0.7907
No log	8.9231	116	0.6010	0.7186	0.6010	0.7752
No log	9.0769	118	0.5768	0.7382	0.5768	0.7595
No log	9.2308	120	0.5725	0.7426	0.5725	0.7566
No log	9.3846	122	0.5699	0.7428	0.5699	0.7549
No log	9.5385	124	0.5714	0.7428	0.5714	0.7559
No log	9.6923	126	0.5746	0.7428	0.5746	0.7580
No log	9.8462	128	0.5777	0.7315	0.5777	0.7601
No log	10.0	130	0.5785	0.7315	0.5785	0.7606

Framework versions

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