ArabicNewSplits6_FineTuningAraBERT_run2_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: 0.7037
• Qwk: 0.3061
• Mse: 0.7037
• Rmse: 0.8389

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	3.5919	-0.0078	3.5919	1.8952
No log	0.5714	4	1.9609	-0.0390	1.9609	1.4003
No log	0.8571	6	0.9395	0.0118	0.9395	0.9693
No log	1.1429	8	0.7189	0.1828	0.7189	0.8479
No log	1.4286	10	0.5866	0.0569	0.5866	0.7659
No log	1.7143	12	0.6798	0.0769	0.6798	0.8245
No log	2.0	14	0.5462	0.0448	0.5462	0.7390
No log	2.2857	16	0.5481	0.0448	0.5481	0.7403
No log	2.5714	18	0.5493	0.1515	0.5493	0.7411
No log	2.8571	20	0.5689	0.1278	0.5689	0.7543
No log	3.1429	22	0.6048	0.1884	0.6048	0.7777
No log	3.4286	24	0.6327	0.0843	0.6327	0.7954
No log	3.7143	26	0.6779	0.0843	0.6779	0.8233
No log	4.0	28	0.8276	0.2079	0.8276	0.9097
No log	4.2857	30	0.9390	0.1464	0.9390	0.9690
No log	4.5714	32	0.7907	0.1230	0.7907	0.8892
No log	4.8571	34	0.9166	0.1493	0.9166	0.9574
No log	5.1429	36	0.9284	0.1803	0.9284	0.9635
No log	5.4286	38	0.7695	0.1527	0.7695	0.8772
No log	5.7143	40	0.9695	0.0795	0.9695	0.9846
No log	6.0	42	1.2415	0.0813	1.2415	1.1142
No log	6.2857	44	1.2311	0.1095	1.2311	1.1095
No log	6.5714	46	0.9033	0.1388	0.9033	0.9504
No log	6.8571	48	0.7084	0.2577	0.7084	0.8417
No log	7.1429	50	0.6782	0.2653	0.6782	0.8235
No log	7.4286	52	0.6675	0.2593	0.6675	0.8170
No log	7.7143	54	0.6641	0.1739	0.6641	0.8149
No log	8.0	56	0.7504	0.2917	0.7504	0.8662
No log	8.2857	58	0.8544	0.1336	0.8544	0.9243
No log	8.5714	60	0.8665	0.1351	0.8665	0.9309
No log	8.8571	62	0.7900	0.2762	0.7900	0.8888
No log	9.1429	64	0.7408	0.3131	0.7408	0.8607
No log	9.4286	66	0.7091	0.3061	0.7091	0.8421
No log	9.7143	68	0.7039	0.2563	0.7039	0.8390
No log	10.0	70	0.7037	0.3061	0.7037	0.8389

Framework versions

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