ArabicNewSplits_FineTuningAraBERT_AugV5_k2_task1_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: 0.9559
• Qwk: 0.5846
• Mse: 0.9559
• Rmse: 0.9777

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	3.9371	-0.0367	3.9371	1.9842
No log	0.3636	4	2.0835	0.1248	2.0835	1.4434
No log	0.5455	6	1.7092	0.2292	1.7092	1.3074
No log	0.7273	8	1.2486	0.2784	1.2486	1.1174
No log	0.9091	10	0.9726	0.4120	0.9726	0.9862
No log	1.0909	12	1.1181	0.3683	1.1181	1.0574
No log	1.2727	14	1.0808	0.3206	1.0808	1.0396
No log	1.4545	16	0.8960	0.4363	0.8960	0.9466
No log	1.6364	18	0.8075	0.5169	0.8075	0.8986
No log	1.8182	20	0.8323	0.5213	0.8323	0.9123
No log	2.0	22	1.0774	0.3423	1.0774	1.0380
No log	2.1818	24	1.0647	0.3505	1.0647	1.0319
No log	2.3636	26	0.9041	0.3997	0.9041	0.9508
No log	2.5455	28	0.7812	0.5187	0.7812	0.8839
No log	2.7273	30	0.8243	0.4592	0.8243	0.9079
No log	2.9091	32	0.7760	0.4913	0.7760	0.8809
No log	3.0909	34	0.7663	0.5299	0.7663	0.8754
No log	3.2727	36	0.7748	0.6099	0.7748	0.8802
No log	3.4545	38	0.8489	0.5884	0.8489	0.9214
No log	3.6364	40	1.0878	0.4495	1.0878	1.0430
No log	3.8182	42	1.2034	0.4393	1.2034	1.0970
No log	4.0	44	1.0507	0.5141	1.0507	1.0250
No log	4.1818	46	0.9451	0.6008	0.9451	0.9722
No log	4.3636	48	0.8817	0.5762	0.8817	0.9390
No log	4.5455	50	0.8883	0.5745	0.8883	0.9425
No log	4.7273	52	0.9079	0.6602	0.9079	0.9528
No log	4.9091	54	1.2099	0.5106	1.2099	1.1000
No log	5.0909	56	1.3895	0.5144	1.3895	1.1788
No log	5.2727	58	1.2914	0.5421	1.2914	1.1364
No log	5.4545	60	1.0840	0.5584	1.0840	1.0412
No log	5.6364	62	0.9566	0.6115	0.9566	0.9781
No log	5.8182	64	0.9740	0.5849	0.9740	0.9869
No log	6.0	66	0.9975	0.5631	0.9975	0.9987
No log	6.1818	68	1.0176	0.5749	1.0176	1.0088
No log	6.3636	70	1.1410	0.5321	1.1410	1.0682
No log	6.5455	72	1.1833	0.5193	1.1833	1.0878
No log	6.7273	74	1.1022	0.5535	1.1022	1.0498
No log	6.9091	76	0.9794	0.5368	0.9794	0.9896
No log	7.0909	78	0.8832	0.5849	0.8832	0.9398
No log	7.2727	80	0.8625	0.6270	0.8625	0.9287
No log	7.4545	82	0.8778	0.6270	0.8778	0.9369
No log	7.6364	84	0.9472	0.5732	0.9472	0.9732
No log	7.8182	86	1.0300	0.5729	1.0300	1.0149
No log	8.0	88	1.0959	0.5452	1.0959	1.0469
No log	8.1818	90	1.1062	0.5314	1.1062	1.0517
No log	8.3636	92	1.0602	0.5506	1.0602	1.0297
No log	8.5455	94	0.9910	0.5603	0.9910	0.9955
No log	8.7273	96	0.9386	0.5972	0.9386	0.9688
No log	8.9091	98	0.9155	0.6161	0.9155	0.9568
No log	9.0909	100	0.9168	0.6161	0.9168	0.9575
No log	9.2727	102	0.9304	0.6043	0.9304	0.9646
No log	9.4545	104	0.9473	0.6043	0.9473	0.9733
No log	9.6364	106	0.9512	0.6043	0.9512	0.9753
No log	9.8182	108	0.9540	0.5846	0.9540	0.9767
No log	10.0	110	0.9559	0.5846	0.9559	0.9777

Framework versions

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