ArabicNewSplits6_FineTuningAraBERT_run3_AugV5_k2_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: 0.9081
• Qwk: 0.6293
• Mse: 0.9081
• Rmse: 0.9529

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	2.3189	0.0143	2.3189	1.5228
No log	0.3636	4	1.6061	0.1453	1.6061	1.2673
No log	0.5455	6	1.4858	0.1096	1.4858	1.2190
No log	0.7273	8	1.5517	0.1209	1.5517	1.2457
No log	0.9091	10	1.4942	0.2036	1.4942	1.2224
No log	1.0909	12	1.5345	0.2898	1.5345	1.2388
No log	1.2727	14	1.5094	0.2954	1.5094	1.2286
No log	1.4545	16	1.3849	0.1475	1.3849	1.1768
No log	1.6364	18	1.3435	0.1826	1.3435	1.1591
No log	1.8182	20	1.2991	0.2186	1.2991	1.1398
No log	2.0	22	1.2653	0.1508	1.2653	1.1249
No log	2.1818	24	1.3091	0.1502	1.3091	1.1441
No log	2.3636	26	1.3817	0.3818	1.3817	1.1755
No log	2.5455	28	1.3800	0.4268	1.3800	1.1748
No log	2.7273	30	1.2511	0.4215	1.2511	1.1185
No log	2.9091	32	1.1082	0.3357	1.1082	1.0527
No log	3.0909	34	1.0879	0.3439	1.0879	1.0430
No log	3.2727	36	1.1033	0.4070	1.1033	1.0504
No log	3.4545	38	1.0586	0.3708	1.0586	1.0289
No log	3.6364	40	1.0327	0.3457	1.0327	1.0162
No log	3.8182	42	1.1099	0.5156	1.1099	1.0535
No log	4.0	44	1.1714	0.5159	1.1714	1.0823
No log	4.1818	46	1.1697	0.5226	1.1697	1.0815
No log	4.3636	48	1.0905	0.5444	1.0905	1.0443
No log	4.5455	50	0.9878	0.5594	0.9878	0.9939
No log	4.7273	52	0.9020	0.5518	0.9020	0.9497
No log	4.9091	54	0.8852	0.5067	0.8852	0.9409
No log	5.0909	56	0.8631	0.5916	0.8631	0.9290
No log	5.2727	58	0.9656	0.5982	0.9656	0.9827
No log	5.4545	60	1.1558	0.4951	1.1558	1.0751
No log	5.6364	62	1.2013	0.4897	1.2013	1.0960
No log	5.8182	64	1.1060	0.5182	1.1060	1.0517
No log	6.0	66	0.9641	0.5885	0.9641	0.9819
No log	6.1818	68	0.8159	0.6761	0.8159	0.9033
No log	6.3636	70	0.7358	0.6802	0.7358	0.8578
No log	6.5455	72	0.7102	0.6759	0.7102	0.8428
No log	6.7273	74	0.7270	0.6802	0.7270	0.8526
No log	6.9091	76	0.7979	0.6730	0.7979	0.8933
No log	7.0909	78	0.9243	0.5899	0.9243	0.9614
No log	7.2727	80	1.0149	0.5546	1.0149	1.0074
No log	7.4545	82	1.0652	0.5580	1.0652	1.0321
No log	7.6364	84	1.0646	0.5806	1.0646	1.0318
No log	7.8182	86	1.0260	0.5890	1.0260	1.0129
No log	8.0	88	0.9834	0.5951	0.9834	0.9916
No log	8.1818	90	0.9557	0.6049	0.9557	0.9776
No log	8.3636	92	0.9616	0.6049	0.9616	0.9806
No log	8.5455	94	0.9833	0.5971	0.9833	0.9916
No log	8.7273	96	1.0104	0.5945	1.0104	1.0052
No log	8.9091	98	1.0084	0.6046	1.0084	1.0042
No log	9.0909	100	0.9901	0.6046	0.9901	0.9950
No log	9.2727	102	0.9648	0.5901	0.9648	0.9822
No log	9.4545	104	0.9375	0.6130	0.9375	0.9683
No log	9.6364	106	0.9168	0.6276	0.9168	0.9575
No log	9.8182	108	0.9110	0.6293	0.9110	0.9544
No log	10.0	110	0.9081	0.6293	0.9081	0.9529

Framework versions

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