ArabicNewSplits6_FineTuningAraBERT_run3_AugV5_k1_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: 0.7955
• Qwk: 0.0380
• Mse: 0.7955
• Rmse: 0.8919

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	0.2857	2	3.3847	0.0013	3.3847	1.8398
No log	0.5714	4	1.7075	-0.0070	1.7075	1.3067
No log	0.8571	6	0.8806	0.0431	0.8806	0.9384
No log	1.1429	8	0.6111	0.1628	0.6111	0.7817
No log	1.4286	10	0.5695	0.0569	0.5695	0.7546
No log	1.7143	12	0.6063	0.0725	0.6063	0.7787
No log	2.0	14	1.2292	0.0075	1.2292	1.1087
No log	2.2857	16	1.0425	-0.0196	1.0425	1.0210
No log	2.5714	18	0.6164	0.0303	0.6164	0.7851
No log	2.8571	20	0.6193	0.0	0.6193	0.7869
No log	3.1429	22	0.7512	0.1407	0.7512	0.8667
No log	3.4286	24	0.7528	0.0720	0.7528	0.8677
No log	3.7143	26	0.6546	0.0	0.6546	0.8091
No log	4.0	28	0.5797	0.0	0.5797	0.7614
No log	4.2857	30	0.6165	0.0222	0.6165	0.7852
No log	4.5714	32	0.6072	-0.0435	0.6072	0.7792
No log	4.8571	34	0.6236	0.0	0.6236	0.7897
No log	5.1429	36	0.6909	0.0	0.6909	0.8312
No log	5.4286	38	0.6907	0.0	0.6907	0.8311
No log	5.7143	40	0.6322	0.0222	0.6322	0.7951
No log	6.0	42	0.6989	0.0692	0.6989	0.8360
No log	6.2857	44	0.7577	-0.0169	0.7577	0.8705
No log	6.5714	46	0.6897	0.0933	0.6897	0.8305
No log	6.8571	48	0.6895	-0.0286	0.6895	0.8304
No log	7.1429	50	0.7050	-0.0207	0.7050	0.8396
No log	7.4286	52	0.7190	0.0210	0.7190	0.8479
No log	7.7143	54	0.7689	0.1138	0.7689	0.8769
No log	8.0	56	0.8325	-0.0204	0.8325	0.9124
No log	8.2857	58	0.8521	-0.0204	0.8521	0.9231
No log	8.5714	60	0.8204	0.0316	0.8204	0.9058
No log	8.8571	62	0.7803	0.0380	0.7803	0.8833
No log	9.1429	64	0.7784	0.0256	0.7784	0.8823
No log	9.4286	66	0.7859	0.0256	0.7859	0.8865
No log	9.7143	68	0.7940	0.0380	0.7940	0.8911
No log	10.0	70	0.7955	0.0380	0.7955	0.8919

Framework versions

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