ArabicNewSplits6_FineTuningAraBERT_run3_AugV5_k2_task2_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: 1.0019
• Qwk: 0.4064
• Mse: 1.0019
• Rmse: 1.0009

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	4.2135	-0.0259	4.2135	2.0527
No log	0.3077	4	2.6414	0.0712	2.6414	1.6252
No log	0.4615	6	1.2669	0.0884	1.2669	1.1256
No log	0.6154	8	0.9761	0.0135	0.9761	0.9880
No log	0.7692	10	0.8787	-0.0284	0.8787	0.9374
No log	0.9231	12	0.7310	0.1706	0.7310	0.8550
No log	1.0769	14	0.7029	0.1569	0.7029	0.8384
No log	1.2308	16	0.6952	0.1638	0.6952	0.8338
No log	1.3846	18	0.6957	0.1529	0.6957	0.8341
No log	1.5385	20	0.6848	0.2248	0.6848	0.8276
No log	1.6923	22	0.6767	0.1907	0.6767	0.8226
No log	1.8462	24	0.6901	0.2126	0.6901	0.8307
No log	2.0	26	0.6706	0.2416	0.6706	0.8189
No log	2.1538	28	0.7627	0.2493	0.7627	0.8733
No log	2.3077	30	0.7147	0.2090	0.7147	0.8454
No log	2.4615	32	0.7783	0.2532	0.7783	0.8822
No log	2.6154	34	0.8754	0.2102	0.8754	0.9356
No log	2.7692	36	1.0469	0.1444	1.0469	1.0232
No log	2.9231	38	0.9168	0.1683	0.9168	0.9575
No log	3.0769	40	0.6342	0.3420	0.6342	0.7963
No log	3.2308	42	0.6029	0.4427	0.6029	0.7765
No log	3.3846	44	0.6343	0.3766	0.6343	0.7964
No log	3.5385	46	0.5821	0.3919	0.5821	0.7630
No log	3.6923	48	0.6571	0.4018	0.6571	0.8106
No log	3.8462	50	0.7716	0.3445	0.7716	0.8784
No log	4.0	52	0.7364	0.3654	0.7364	0.8582
No log	4.1538	54	0.6476	0.4164	0.6476	0.8047
No log	4.3077	56	0.7711	0.4250	0.7711	0.8781
No log	4.4615	58	0.9730	0.3850	0.9730	0.9864
No log	4.6154	60	0.9366	0.3553	0.9366	0.9678
No log	4.7692	62	0.8107	0.4480	0.8107	0.9004
No log	4.9231	64	0.9182	0.4070	0.9182	0.9582
No log	5.0769	66	0.9764	0.3774	0.9764	0.9881
No log	5.2308	68	0.9137	0.3947	0.9137	0.9559
No log	5.3846	70	0.8957	0.4224	0.8957	0.9464
No log	5.5385	72	0.9613	0.3774	0.9613	0.9805
No log	5.6923	74	0.9711	0.3774	0.9711	0.9854
No log	5.8462	76	0.9074	0.4240	0.9074	0.9526
No log	6.0	78	0.8764	0.4202	0.8764	0.9361
No log	6.1538	80	0.8561	0.3780	0.8561	0.9253
No log	6.3077	82	0.8555	0.3630	0.8555	0.9249
No log	6.4615	84	0.8664	0.3987	0.8664	0.9308
No log	6.6154	86	0.8705	0.3987	0.8705	0.9330
No log	6.7692	88	0.8774	0.4066	0.8774	0.9367
No log	6.9231	90	0.8755	0.3863	0.8755	0.9357
No log	7.0769	92	0.8758	0.4223	0.8758	0.9358
No log	7.2308	94	0.8860	0.3964	0.8860	0.9413
No log	7.3846	96	0.9080	0.3944	0.9080	0.9529
No log	7.5385	98	0.9194	0.3868	0.9194	0.9588
No log	7.6923	100	0.9083	0.3944	0.9083	0.9530
No log	7.8462	102	0.8968	0.3824	0.8968	0.9470
No log	8.0	104	0.9164	0.3621	0.9164	0.9573
No log	8.1538	106	0.9399	0.3954	0.9399	0.9695
No log	8.3077	108	0.9472	0.4094	0.9472	0.9733
No log	8.4615	110	0.9472	0.3929	0.9472	0.9733
No log	8.6154	112	0.9583	0.3963	0.9583	0.9789
No log	8.7692	114	0.9634	0.3884	0.9634	0.9816
No log	8.9231	116	0.9778	0.3926	0.9778	0.9889
No log	9.0769	118	0.9951	0.4009	0.9951	0.9975
No log	9.2308	120	1.0029	0.4077	1.0029	1.0015
No log	9.3846	122	1.0049	0.3978	1.0049	1.0024
No log	9.5385	124	1.0045	0.3929	1.0045	1.0023
No log	9.6923	126	1.0031	0.3950	1.0031	1.0016
No log	9.8462	128	1.0025	0.4064	1.0025	1.0012
No log	10.0	130	1.0019	0.4064	1.0019	1.0009

Framework versions

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