ArabicNewSplits6_FineTuningAraBERT_run3_AugV5_k2_task3_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: 0.8531
• Qwk: 0.1545
• Mse: 0.8531
• Rmse: 0.9237

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	0.1667	2	3.2610	-0.0138	3.2610	1.8058
No log	0.3333	4	1.6647	-0.0070	1.6647	1.2902
No log	0.5	6	0.9499	0.0745	0.9499	0.9746
No log	0.6667	8	0.6053	0.1895	0.6053	0.7780
No log	0.8333	10	0.5667	0.0	0.5667	0.7528
No log	1.0	12	0.5804	0.0	0.5804	0.7618
No log	1.1667	14	0.6683	0.2688	0.6683	0.8175
No log	1.3333	16	0.7686	0.2072	0.7686	0.8767
No log	1.5	18	1.0077	0.0476	1.0077	1.0038
No log	1.6667	20	0.8404	0.1289	0.8404	0.9167
No log	1.8333	22	0.6444	0.0071	0.6444	0.8027
No log	2.0	24	0.6588	-0.0794	0.6588	0.8117
No log	2.1667	26	0.6088	0.0	0.6088	0.7803
No log	2.3333	28	0.5444	-0.0081	0.5444	0.7378
No log	2.5	30	0.5601	0.1895	0.5601	0.7484
No log	2.6667	32	0.5635	0.1813	0.5635	0.7507
No log	2.8333	34	0.6120	0.2917	0.6120	0.7823
No log	3.0	36	0.5508	0.1515	0.5508	0.7421
No log	3.1667	38	0.5475	0.1605	0.5475	0.7399
No log	3.3333	40	0.5321	0.2208	0.5321	0.7295
No log	3.5	42	0.5393	0.2471	0.5393	0.7344
No log	3.6667	44	0.5319	0.3711	0.5319	0.7293
No log	3.8333	46	0.6686	0.1795	0.6686	0.8177
No log	4.0	48	0.6801	0.1801	0.6801	0.8247
No log	4.1667	50	0.5749	0.2970	0.5749	0.7582
No log	4.3333	52	0.5966	0.2184	0.5966	0.7724
No log	4.5	54	0.6889	0.2575	0.6889	0.8300
No log	4.6667	56	0.7616	0.2670	0.7616	0.8727
No log	4.8333	58	0.7048	0.2370	0.7048	0.8395
No log	5.0	60	0.7162	0.0843	0.7162	0.8463
No log	5.1667	62	0.8158	0.1828	0.8158	0.9032
No log	5.3333	64	0.8515	0.1832	0.8515	0.9228
No log	5.5	66	0.8888	0.2442	0.8888	0.9427
No log	5.6667	68	0.8314	0.2381	0.8314	0.9118
No log	5.8333	70	0.6936	0.0909	0.6936	0.8328
No log	6.0	72	0.6400	0.2083	0.6400	0.8000
No log	6.1667	74	0.6608	0.1823	0.6608	0.8129
No log	6.3333	76	0.8031	0.2227	0.8031	0.8962
No log	6.5	78	0.7579	0.2227	0.7579	0.8706
No log	6.6667	80	0.6672	0.3016	0.6672	0.8168
No log	6.8333	82	0.7104	0.2577	0.7104	0.8429
No log	7.0	84	0.7749	0.2593	0.7749	0.8803
No log	7.1667	86	0.7127	0.2390	0.7127	0.8442
No log	7.3333	88	0.7194	0.2000	0.7194	0.8482
No log	7.5	90	0.8022	0.2593	0.8022	0.8957
No log	7.6667	92	0.9123	0.125	0.9123	0.9552
No log	7.8333	94	0.9991	0.1014	0.9991	0.9995
No log	8.0	96	0.9428	0.0958	0.9428	0.9710
No log	8.1667	98	0.9741	0.0996	0.9741	0.9869
No log	8.3333	100	0.9801	0.0996	0.9801	0.9900
No log	8.5	102	1.0366	0.1014	1.0366	1.0181
No log	8.6667	104	1.0133	0.0996	1.0133	1.0066
No log	8.8333	106	0.9291	0.1128	0.9291	0.9639
No log	9.0	108	0.9011	0.1392	0.9011	0.9493
No log	9.1667	110	0.9097	0.1093	0.9097	0.9538
No log	9.3333	112	0.9282	0.0794	0.9282	0.9634
No log	9.5	114	0.9154	0.1093	0.9154	0.9568
No log	9.6667	116	0.8835	0.1392	0.8835	0.9399
No log	9.8333	118	0.8609	0.1545	0.8609	0.9278
No log	10.0	120	0.8531	0.1545	0.8531	0.9237

Framework versions

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