Arabic_FineTuningAraBERT_noAug_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: 0.7619
• Qwk: 0.7454
• Mse: 0.7619
• Rmse: 0.8729

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	0.1818	2	2.8279	0.0171	2.8279	1.6816
No log	0.3636	4	2.2488	0.0400	2.2488	1.4996
No log	0.5455	6	1.6228	0.0	1.6228	1.2739
No log	0.7273	8	1.2957	0.0402	1.2957	1.1383
No log	0.9091	10	1.1708	0.0847	1.1708	1.0820
No log	1.0909	12	1.1213	0.1315	1.1213	1.0589
No log	1.2727	14	1.0661	0.1337	1.0661	1.0325
No log	1.4545	16	1.0087	0.2366	1.0087	1.0043
No log	1.6364	18	0.9387	0.25	0.9387	0.9688
No log	1.8182	20	0.8709	0.25	0.8709	0.9332
No log	2.0	22	0.8250	0.2839	0.8250	0.9083
No log	2.1818	24	0.8200	0.2839	0.8200	0.9056
No log	2.3636	26	0.7726	0.3602	0.7726	0.8790
No log	2.5455	28	0.7253	0.4444	0.7253	0.8517
No log	2.7273	30	0.8181	0.4101	0.8181	0.9045
No log	2.9091	32	0.9323	0.6460	0.9323	0.9656
No log	3.0909	34	0.8816	0.6528	0.8816	0.9389
No log	3.2727	36	0.7752	0.6690	0.7752	0.8805
No log	3.4545	38	0.7313	0.6860	0.7313	0.8552
No log	3.6364	40	0.7014	0.6860	0.7014	0.8375
No log	3.8182	42	0.7972	0.6690	0.7972	0.8928
No log	4.0	44	0.9486	0.5556	0.9486	0.9739
No log	4.1818	46	1.0658	0.5062	1.0658	1.0324
No log	4.3636	48	1.0538	0.5062	1.0538	1.0266
No log	4.5455	50	0.9729	0.6690	0.9729	0.9863
No log	4.7273	52	0.8401	0.6690	0.8401	0.9166
No log	4.9091	54	0.7317	0.6860	0.7317	0.8554
No log	5.0909	56	0.6792	0.7176	0.6792	0.8241
No log	5.2727	58	0.6828	0.7176	0.6828	0.8263
No log	5.4545	60	0.7411	0.6690	0.7411	0.8609
No log	5.6364	62	0.8518	0.6368	0.8518	0.9229
No log	5.8182	64	0.8708	0.6514	0.8708	0.9332
No log	6.0	66	0.8761	0.6667	0.8761	0.9360
No log	6.1818	68	0.8494	0.6528	0.8494	0.9216
No log	6.3636	70	0.8465	0.6544	0.8465	0.9200
No log	6.5455	72	0.8179	0.6696	0.8179	0.9044
No log	6.7273	74	0.8030	0.6696	0.8030	0.8961
No log	6.9091	76	0.7552	0.7149	0.7552	0.8690
No log	7.0909	78	0.7056	0.7149	0.7056	0.8400
No log	7.2727	80	0.6861	0.7149	0.6861	0.8283
No log	7.4545	82	0.6870	0.7149	0.6870	0.8289
No log	7.6364	84	0.7442	0.7149	0.7442	0.8627
No log	7.8182	86	0.8680	0.6065	0.8680	0.9317
No log	8.0	88	0.9583	0.6222	0.9583	0.9789
No log	8.1818	90	0.9770	0.6222	0.9770	0.9884
No log	8.3636	92	0.9388	0.6667	0.9388	0.9689
No log	8.5455	94	0.8745	0.6825	0.8745	0.9352
No log	8.7273	96	0.8389	0.6825	0.8389	0.9159
No log	8.9091	98	0.8180	0.7454	0.8180	0.9044
No log	9.0909	100	0.7749	0.7454	0.7749	0.8803
No log	9.2727	102	0.7351	0.7743	0.7351	0.8574
No log	9.4545	104	0.7308	0.7005	0.7308	0.8549
No log	9.6364	106	0.7466	0.7454	0.7466	0.8641
No log	9.8182	108	0.7567	0.7454	0.7567	0.8699
No log	10.0	110	0.7619	0.7454	0.7619	0.8729

Framework versions

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