metadata
license: mit
base_model: dslim/bert-base-NER
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: bert-ner-finetuned-pii
results: []
bert-ner-finetuned-pii
This model is a fine-tuned version of dslim/bert-base-NER on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0019
- Precision: 0.7429
- Recall: 0.7192
- F1: 0.7309
- Accuracy: 0.9995
Model description
Model is fine tuned in a PII task in students essays
Intended uses & limitations
More information needed
Training and evaluation data
to be updated
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: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.0023 | 1.0 | 724 | 0.0033 | 0.6791 | 0.4589 | 0.5477 | 0.9992 |
| 0.0014 | 2.0 | 1448 | 0.0021 | 0.7218 | 0.6575 | 0.6882 | 0.9994 |
| 0.001 | 3.0 | 2172 | 0.0019 | 0.7429 | 0.7192 | 0.7309 | 0.9995 |
Framework versions
- Transformers 4.37.0
- Pytorch 2.1.2
- Datasets 2.1.0
- Tokenizers 0.15.1