ArabicNewSplits6_FineTuningAraBERT_run2_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: 0.8879
• Qwk: 0.6848
• Mse: 0.8879
• Rmse: 0.9423

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	2.2709	0.0088	2.2709	1.5070
No log	0.3636	4	1.7138	0.0400	1.7138	1.3091
No log	0.5455	6	1.7412	0.0136	1.7412	1.3195
No log	0.7273	8	1.5302	0.2050	1.5302	1.2370
No log	0.9091	10	1.4885	0.2588	1.4885	1.2200
No log	1.0909	12	1.4998	0.2405	1.4998	1.2247
No log	1.2727	14	1.6424	0.2639	1.6424	1.2815
No log	1.4545	16	1.6052	0.2996	1.6052	1.2670
No log	1.6364	18	1.4910	0.2354	1.4910	1.2211
No log	1.8182	20	1.3258	0.3025	1.3258	1.1514
No log	2.0	22	1.2220	0.3395	1.2220	1.1054
No log	2.1818	24	1.1524	0.3226	1.1524	1.0735
No log	2.3636	26	1.1310	0.2682	1.1310	1.0635
No log	2.5455	28	1.1343	0.2936	1.1343	1.0650
No log	2.7273	30	1.1651	0.4151	1.1651	1.0794
No log	2.9091	32	1.2613	0.4607	1.2613	1.1231
No log	3.0909	34	1.2590	0.5218	1.2590	1.1220
No log	3.2727	36	1.1954	0.5200	1.1954	1.0933
No log	3.4545	38	1.0391	0.5903	1.0391	1.0194
No log	3.6364	40	0.9924	0.6233	0.9924	0.9962
No log	3.8182	42	0.9426	0.6332	0.9426	0.9709
No log	4.0	44	0.8869	0.6678	0.8869	0.9418
No log	4.1818	46	0.9002	0.6559	0.9002	0.9488
No log	4.3636	48	0.9887	0.6184	0.9887	0.9943
No log	4.5455	50	1.1141	0.5550	1.1141	1.0555
No log	4.7273	52	1.1507	0.5467	1.1507	1.0727
No log	4.9091	54	1.1312	0.5334	1.1312	1.0636
No log	5.0909	56	1.0697	0.5485	1.0697	1.0343
No log	5.2727	58	1.0063	0.5771	1.0063	1.0031
No log	5.4545	60	0.9096	0.5850	0.9096	0.9537
No log	5.6364	62	0.8272	0.6061	0.8272	0.9095
No log	5.8182	64	0.7564	0.6485	0.7564	0.8697
No log	6.0	66	0.7170	0.6871	0.7170	0.8468
No log	6.1818	68	0.7125	0.7037	0.7125	0.8441
No log	6.3636	70	0.7421	0.6595	0.7421	0.8615
No log	6.5455	72	0.8479	0.6552	0.8479	0.9208
No log	6.7273	74	1.0228	0.6419	1.0228	1.0113
No log	6.9091	76	1.1326	0.6172	1.1326	1.0642
No log	7.0909	78	1.1401	0.6238	1.1401	1.0677
No log	7.2727	80	1.0679	0.6180	1.0679	1.0334
No log	7.4545	82	0.9903	0.6347	0.9903	0.9951
No log	7.6364	84	0.8867	0.6671	0.8867	0.9416
No log	7.8182	86	0.7989	0.6803	0.7989	0.8938
No log	8.0	88	0.7719	0.6959	0.7719	0.8786
No log	8.1818	90	0.7639	0.7062	0.7639	0.8740
No log	8.3636	92	0.7947	0.6985	0.7947	0.8915
No log	8.5455	94	0.8317	0.6956	0.8317	0.9120
No log	8.7273	96	0.8655	0.6778	0.8655	0.9303
No log	8.9091	98	0.8904	0.6671	0.8904	0.9436
No log	9.0909	100	0.8997	0.6671	0.8997	0.9485
No log	9.2727	102	0.8936	0.6671	0.8936	0.9453
No log	9.4545	104	0.8935	0.6769	0.8935	0.9452
No log	9.6364	106	0.8911	0.6848	0.8911	0.9440
No log	9.8182	108	0.8900	0.6848	0.8900	0.9434
No log	10.0	110	0.8879	0.6848	0.8879	0.9423

Framework versions

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