ArabicNewSplits6_WithDuplicationsForScore5_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: 0.6986
• Qwk: 0.7427
• Mse: 0.6986
• Rmse: 0.8359

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	2.3641	0.0579	2.3641	1.5376
No log	0.3636	4	1.4772	0.2630	1.4772	1.2154
No log	0.5455	6	1.2256	0.2737	1.2256	1.1071
No log	0.7273	8	1.2324	0.1959	1.2324	1.1101
No log	0.9091	10	1.3593	0.1926	1.3593	1.1659
No log	1.0909	12	1.4275	0.2700	1.4275	1.1948
No log	1.2727	14	1.4621	0.4035	1.4621	1.2092
No log	1.4545	16	1.6175	0.3940	1.6175	1.2718
No log	1.6364	18	1.6256	0.3769	1.6256	1.2750
No log	1.8182	20	1.4448	0.4257	1.4448	1.2020
No log	2.0	22	1.2749	0.4032	1.2749	1.1291
No log	2.1818	24	1.1558	0.3298	1.1558	1.0751
No log	2.3636	26	1.1112	0.3356	1.1112	1.0542
No log	2.5455	28	1.0616	0.4108	1.0616	1.0303
No log	2.7273	30	0.9823	0.4425	0.9823	0.9911
No log	2.9091	32	0.9610	0.5534	0.9610	0.9803
No log	3.0909	34	0.9654	0.5297	0.9654	0.9825
No log	3.2727	36	0.9966	0.5524	0.9966	0.9983
No log	3.4545	38	1.0078	0.5985	1.0078	1.0039
No log	3.6364	40	0.9842	0.6192	0.9842	0.9921
No log	3.8182	42	0.9355	0.6169	0.9355	0.9672
No log	4.0	44	0.8139	0.6378	0.8139	0.9021
No log	4.1818	46	0.7261	0.6596	0.7261	0.8521
No log	4.3636	48	0.6957	0.6952	0.6957	0.8341
No log	4.5455	50	0.6885	0.7098	0.6885	0.8297
No log	4.7273	52	0.6872	0.7003	0.6872	0.8290
No log	4.9091	54	0.7308	0.6809	0.7308	0.8549
No log	5.0909	56	0.7887	0.6864	0.7887	0.8881
No log	5.2727	58	0.8021	0.6942	0.8021	0.8956
No log	5.4545	60	0.8135	0.6855	0.8135	0.9019
No log	5.6364	62	0.7672	0.7098	0.7672	0.8759
No log	5.8182	64	0.7452	0.7031	0.7452	0.8632
No log	6.0	66	0.7024	0.6977	0.7024	0.8381
No log	6.1818	68	0.6779	0.6713	0.6779	0.8234
No log	6.3636	70	0.6832	0.7015	0.6832	0.8266
No log	6.5455	72	0.7007	0.6850	0.7007	0.8371
No log	6.7273	74	0.7242	0.6875	0.7242	0.8510
No log	6.9091	76	0.8051	0.7014	0.8051	0.8973
No log	7.0909	78	0.9199	0.6613	0.9199	0.9591
No log	7.2727	80	0.9603	0.6439	0.9603	0.9800
No log	7.4545	82	0.9467	0.6535	0.9467	0.9730
No log	7.6364	84	0.8903	0.6763	0.8903	0.9436
No log	7.8182	86	0.8205	0.6911	0.8205	0.9058
No log	8.0	88	0.7397	0.7057	0.7397	0.8601
No log	8.1818	90	0.6963	0.7211	0.6963	0.8344
No log	8.3636	92	0.6791	0.7453	0.6791	0.8241
No log	8.5455	94	0.6697	0.7393	0.6697	0.8184
No log	8.7273	96	0.6786	0.7453	0.6786	0.8238
No log	8.9091	98	0.6860	0.7346	0.6860	0.8282
No log	9.0909	100	0.6908	0.7346	0.6908	0.8312
No log	9.2727	102	0.6924	0.7346	0.6924	0.8321
No log	9.4545	104	0.6961	0.7427	0.6961	0.8343
No log	9.6364	106	0.6955	0.7427	0.6955	0.8339
No log	9.8182	108	0.6977	0.7427	0.6977	0.8353
No log	10.0	110	0.6986	0.7427	0.6986	0.8359

Framework versions

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