car13mesquita/bert-finetuned-sem_eval-rest14-english

bert-finetuned-sem_eval-english

This model is a fine-tuned version of bert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0808
• F1: 0.4747
• Roc Auc: 0.8705
• Accuracy: 0.6863
• Hamming Loss: 0.0246

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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy	Hamming Loss
No log	1.0	381	0.1614	0.0450	0.6326	0.2188	0.0506
0.2203	2.0	762	0.1223	0.1837	0.7240	0.3962	0.0394
0.1234	3.0	1143	0.1005	0.3528	0.8231	0.6038	0.0293
0.0852	4.0	1524	0.0912	0.3786	0.8317	0.6212	0.0276
0.0852	5.0	1905	0.0871	0.3947	0.8481	0.6538	0.0267
0.0623	6.0	2286	0.0859	0.4013	0.8493	0.645	0.0269
0.0485	7.0	2667	0.0828	0.4231	0.8573	0.6575	0.0255
0.0382	8.0	3048	0.0812	0.4368	0.8598	0.6725	0.0251
0.0382	9.0	3429	0.0803	0.4538	0.8697	0.6837	0.0244
0.0321	10.0	3810	0.0817	0.4513	0.8653	0.675	0.0249
0.0267	11.0	4191	0.0819	0.4732	0.8655	0.6737	0.0254
0.023	12.0	4572	0.0815	0.4648	0.8675	0.6825	0.0252
0.023	13.0	4953	0.0808	0.4733	0.8699	0.68	0.0249
0.0211	14.0	5334	0.0808	0.4747	0.8705	0.6863	0.0246
0.0192	15.0	5715	0.0807	0.4727	0.8717	0.6837	0.025

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1