bert-base-multilingual-cased-finetuned-ner-lenerBR
This model is a fine-tuned version of google-bert/bert-base-multilingual-cased on the lener_br dataset. It achieves the following results on the evaluation set:
- Loss: 0.1941
- Precision: 0.8457
- Recall: 0.8475
- F1: 0.8466
- Accuracy: 0.9642
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: 32
- eval_batch_size: 32
- 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 | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
No log | 1.0 | 245 | 0.2100 | 0.7326 | 0.7596 | 0.7459 | 0.9478 |
No log | 2.0 | 490 | 0.1885 | 0.7737 | 0.8119 | 0.7923 | 0.9548 |
0.1595 | 3.0 | 735 | 0.1491 | 0.8056 | 0.8388 | 0.8218 | 0.9616 |
0.1595 | 4.0 | 980 | 0.1787 | 0.8369 | 0.8251 | 0.8310 | 0.9612 |
0.0311 | 5.0 | 1225 | 0.1788 | 0.8303 | 0.8601 | 0.8450 | 0.9646 |
0.0311 | 6.0 | 1470 | 0.2131 | 0.7985 | 0.8463 | 0.8217 | 0.9595 |
0.0156 | 7.0 | 1715 | 0.1879 | 0.8161 | 0.8635 | 0.8392 | 0.9630 |
0.0156 | 8.0 | 1960 | 0.1975 | 0.8445 | 0.8469 | 0.8457 | 0.9636 |
0.0091 | 9.0 | 2205 | 0.1979 | 0.8460 | 0.8422 | 0.8441 | 0.9635 |
0.0091 | 10.0 | 2450 | 0.1941 | 0.8457 | 0.8475 | 0.8466 | 0.9642 |
Framework versions
- Transformers 4.41.1
- Pytorch 2.1.2
- Datasets 2.19.1
- Tokenizers 0.19.1
- Downloads last month
- 28
Inference API (serverless) is not available, repository is disabled.
Model tree for GuiTap/bert-base-multilingual-cased-finetuned-ner-lenerBR
Base model
google-bert/bert-base-multilingual-cased
Finetuned
this model
Dataset used to train GuiTap/bert-base-multilingual-cased-finetuned-ner-lenerBR
Evaluation results
- Precision on lener_brvalidation set self-reported0.846
- Recall on lener_brvalidation set self-reported0.848
- F1 on lener_brvalidation set self-reported0.847
- Accuracy on lener_brvalidation set self-reported0.964