ArabicNewSplits6_WithDuplicationsForScore5_FineTuningAraBERT_run1_AugV5_k1_task2_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: 0.5234
• Qwk: 0.5794
• Mse: 0.5234
• Rmse: 0.7235

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	0.25	2	3.6970	-0.0034	3.6970	1.9228
No log	0.5	4	2.6934	0.0556	2.6934	1.6412
No log	0.75	6	1.3705	0.1197	1.3705	1.1707
No log	1.0	8	0.9274	0.0135	0.9274	0.9630
No log	1.25	10	0.8082	0.1079	0.8082	0.8990
No log	1.5	12	0.8192	0.1063	0.8192	0.9051
No log	1.75	14	0.9281	0.0772	0.9281	0.9634
No log	2.0	16	1.0480	0.1210	1.0480	1.0237
No log	2.25	18	1.0025	0.1420	1.0025	1.0012
No log	2.5	20	0.9817	0.1525	0.9817	0.9908
No log	2.75	22	0.8630	0.2114	0.8630	0.9290
No log	3.0	24	0.6868	0.2046	0.6868	0.8288
No log	3.25	26	0.5862	0.2650	0.5862	0.7656
No log	3.5	28	0.5476	0.3588	0.5476	0.7400
No log	3.75	30	0.5354	0.4625	0.5354	0.7317
No log	4.0	32	0.5611	0.3223	0.5611	0.7490
No log	4.25	34	0.6142	0.2884	0.6142	0.7837
No log	4.5	36	0.6313	0.2558	0.6313	0.7945
No log	4.75	38	0.6507	0.3389	0.6507	0.8067
No log	5.0	40	0.6076	0.4726	0.6076	0.7795
No log	5.25	42	0.5361	0.5485	0.5361	0.7322
No log	5.5	44	0.5421	0.5792	0.5421	0.7362
No log	5.75	46	0.5666	0.5467	0.5666	0.7527
No log	6.0	48	0.5897	0.5156	0.5897	0.7679
No log	6.25	50	0.6058	0.5113	0.6058	0.7783
No log	6.5	52	0.5937	0.5507	0.5937	0.7705
No log	6.75	54	0.5385	0.5594	0.5385	0.7338
No log	7.0	56	0.5149	0.6070	0.5149	0.7176
No log	7.25	58	0.5168	0.5787	0.5168	0.7189
No log	7.5	60	0.5335	0.5882	0.5335	0.7304
No log	7.75	62	0.5545	0.5840	0.5545	0.7447
No log	8.0	64	0.5535	0.5820	0.5535	0.7440
No log	8.25	66	0.5620	0.5880	0.5620	0.7497
No log	8.5	68	0.5590	0.5880	0.5590	0.7476
No log	8.75	70	0.5539	0.5880	0.5539	0.7443
No log	9.0	72	0.5444	0.5901	0.5444	0.7378
No log	9.25	74	0.5346	0.5880	0.5346	0.7312
No log	9.5	76	0.5282	0.5922	0.5282	0.7268
No log	9.75	78	0.5233	0.5794	0.5233	0.7234
No log	10.0	80	0.5234	0.5794	0.5234	0.7235

Framework versions

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