ArabicNewSplits_FineTuningAraBERT_AugV5_k2_task2_organization_fold0

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: 0.9873
• Qwk: 0.3542
• Mse: 0.9873
• Rmse: 0.9936

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	3.6205	0.0205	3.6205	1.9028
No log	0.5	4	2.2151	0.0975	2.2151	1.4883
No log	0.75	6	1.2713	0.1394	1.2713	1.1275
No log	1.0	8	0.9714	0.0731	0.9714	0.9856
No log	1.25	10	0.8980	0.0269	0.8980	0.9476
No log	1.5	12	1.2682	0.0816	1.2682	1.1262
No log	1.75	14	1.3561	0.0675	1.3561	1.1645
No log	2.0	16	1.0735	0.0882	1.0735	1.0361
No log	2.25	18	0.8267	0.1270	0.8267	0.9092
No log	2.5	20	0.7335	0.0826	0.7335	0.8564
No log	2.75	22	0.7324	0.1887	0.7324	0.8558
No log	3.0	24	0.7464	0.1467	0.7464	0.8639
No log	3.25	26	0.7925	0.2440	0.7925	0.8902
No log	3.5	28	0.7960	0.2934	0.7960	0.8922
No log	3.75	30	0.7873	0.2394	0.7873	0.8873
No log	4.0	32	0.7811	0.1658	0.7811	0.8838
No log	4.25	34	0.7908	0.2545	0.7908	0.8893
No log	4.5	36	0.8179	0.2930	0.8179	0.9044
No log	4.75	38	0.8891	0.3313	0.8891	0.9429
No log	5.0	40	0.8572	0.3190	0.8572	0.9259
No log	5.25	42	0.8519	0.2800	0.8519	0.9230
No log	5.5	44	0.8651	0.3114	0.8651	0.9301
No log	5.75	46	0.8632	0.3207	0.8632	0.9291
No log	6.0	48	0.8950	0.3646	0.8950	0.9460
No log	6.25	50	0.9539	0.3485	0.9539	0.9767
No log	6.5	52	0.9725	0.3542	0.9725	0.9862
No log	6.75	54	0.9180	0.3363	0.9180	0.9581
No log	7.0	56	0.8630	0.2791	0.8630	0.9290
No log	7.25	58	0.8613	0.2821	0.8613	0.9281
No log	7.5	60	0.8714	0.2479	0.8714	0.9335
No log	7.75	62	0.9099	0.3043	0.9099	0.9539
No log	8.0	64	0.9954	0.3333	0.9954	0.9977
No log	8.25	66	1.0755	0.3408	1.0755	1.0371
No log	8.5	68	1.0944	0.3408	1.0944	1.0461
No log	8.75	70	1.1255	0.3408	1.1255	1.0609
No log	9.0	72	1.1088	0.3408	1.1088	1.0530
No log	9.25	74	1.0577	0.3580	1.0577	1.0284
No log	9.5	76	1.0148	0.3667	1.0148	1.0074
No log	9.75	78	0.9975	0.3568	0.9975	0.9987
No log	10.0	80	0.9873	0.3542	0.9873	0.9936

Framework versions

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