Arabic_FineTuningAraBERT_AugV5_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: 0.9060
• Qwk: 0.5040
• Mse: 0.9060
• Rmse: 0.9518

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	0.1333	2	3.0375	-0.1012	3.0375	1.7429
No log	0.2667	4	2.0023	-0.0178	2.0023	1.4150
No log	0.4	6	1.5628	-0.2868	1.5628	1.2501
No log	0.5333	8	1.4236	-0.0443	1.4236	1.1932
No log	0.6667	10	1.2406	0.0809	1.2406	1.1138
No log	0.8	12	1.1435	0.1481	1.1435	1.0693
No log	0.9333	14	1.0647	0.1481	1.0647	1.0318
No log	1.0667	16	1.0504	0.1438	1.0504	1.0249
No log	1.2	18	0.9891	0.1438	0.9891	0.9946
No log	1.3333	20	0.9271	0.2920	0.9271	0.9629
No log	1.4667	22	0.8960	0.2077	0.8960	0.9466
No log	1.6	24	0.8925	0.4041	0.8925	0.9447
No log	1.7333	26	0.9790	0.3455	0.9790	0.9895
No log	1.8667	28	0.9698	0.3350	0.9698	0.9848
No log	2.0	30	0.9603	0.4102	0.9603	0.9799
No log	2.1333	32	1.0048	0.2775	1.0048	1.0024
No log	2.2667	34	0.8918	0.4262	0.8918	0.9443
No log	2.4	36	0.8874	0.3455	0.8874	0.9420
No log	2.5333	38	1.0857	0.3250	1.0857	1.0420
No log	2.6667	40	1.1136	0.3250	1.1136	1.0553
No log	2.8	42	0.8976	0.5024	0.8976	0.9474
No log	2.9333	44	0.7039	0.5704	0.7039	0.8390
No log	3.0667	46	0.7387	0.5844	0.7387	0.8595
No log	3.2	48	0.7261	0.6526	0.7261	0.8521
No log	3.3333	50	0.7850	0.5181	0.7850	0.8860
No log	3.4667	52	1.0318	0.4318	1.0318	1.0158
No log	3.6	54	1.2530	0.2954	1.2530	1.1194
No log	3.7333	56	1.2536	0.1436	1.2536	1.1196
No log	3.8667	58	1.1481	0.3490	1.1481	1.0715
No log	4.0	60	0.9497	0.5330	0.9497	0.9745
No log	4.1333	62	0.7913	0.4865	0.7913	0.8895
No log	4.2667	64	0.8270	0.5455	0.8270	0.9094
No log	4.4	66	0.8721	0.4134	0.8721	0.9339
No log	4.5333	68	0.8007	0.5455	0.8007	0.8948
No log	4.6667	70	0.7337	0.6325	0.7337	0.8566
No log	4.8	72	0.8357	0.6126	0.8357	0.9142
No log	4.9333	74	1.0382	0.4773	1.0382	1.0189
No log	5.0667	76	1.2790	0.4541	1.2790	1.1309
No log	5.2	78	1.2913	0.4541	1.2913	1.1363
No log	5.3333	80	1.0782	0.5435	1.0782	1.0384
No log	5.4667	82	0.8119	0.5455	0.8119	0.9010
No log	5.6	84	0.7329	0.5604	0.7329	0.8561
No log	5.7333	86	0.7286	0.6188	0.7286	0.8536
No log	5.8667	88	0.7255	0.6188	0.7255	0.8518
No log	6.0	90	0.7566	0.5905	0.7566	0.8698
No log	6.1333	92	0.7655	0.5905	0.7655	0.8749
No log	6.2667	94	0.8586	0.6026	0.8586	0.9266
No log	6.4	96	1.0717	0.5910	1.0717	1.0352
No log	6.5333	98	1.2574	0.5910	1.2574	1.1213
No log	6.6667	100	1.2814	0.5188	1.2814	1.1320
No log	6.8	102	1.1770	0.5910	1.1770	1.0849
No log	6.9333	104	1.0073	0.54	1.0073	1.0037
No log	7.0667	106	0.9468	0.5519	0.9468	0.9731
No log	7.2	108	0.9054	0.5519	0.9054	0.9515
No log	7.3333	110	0.8871	0.5418	0.8871	0.9419
No log	7.4667	112	0.9518	0.5519	0.9518	0.9756
No log	7.6	114	1.0376	0.5228	1.0376	1.0187
No log	7.7333	116	1.1124	0.5112	1.1124	1.0547
No log	7.8667	118	1.1793	0.5112	1.1793	1.0859
No log	8.0	120	1.1828	0.5112	1.1828	1.0876
No log	8.1333	122	1.1305	0.5643	1.1305	1.0632
No log	8.2667	124	1.0823	0.5643	1.0823	1.0404
No log	8.4	126	1.0334	0.5772	1.0334	1.0166
No log	8.5333	128	0.9949	0.5228	0.9949	0.9975
No log	8.6667	130	0.9476	0.5132	0.9476	0.9735
No log	8.8	132	0.9412	0.4726	0.9412	0.9702
No log	8.9333	134	0.9465	0.4813	0.9465	0.9729
No log	9.0667	136	0.9430	0.4726	0.9430	0.9711
No log	9.2	138	0.9301	0.4726	0.9301	0.9644
No log	9.3333	140	0.9150	0.5040	0.9150	0.9566
No log	9.4667	142	0.9062	0.5040	0.9062	0.9519
No log	9.6	144	0.9033	0.5040	0.9033	0.9504
No log	9.7333	146	0.9027	0.5040	0.9027	0.9501
No log	9.8667	148	0.9042	0.5040	0.9042	0.9509
No log	10.0	150	0.9060	0.5040	0.9060	0.9518

Framework versions

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