metadata
license: mit
tags:
- generated_from_trainer
datasets:
- conll2003
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: bert-base-NER-finetuned-ner
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: conll2003
type: conll2003
args: conll2003
metrics:
- name: Precision
type: precision
value: 0.9327342290239345
- name: Recall
type: recall
value: 0.9405167773192177
- name: F1
type: f1
value: 0.9366093366093367
- name: Accuracy
type: accuracy
value: 0.9850621063165951
bert-base-NER-finetuned-ner
This model is a fine-tuned version of dslim/bert-base-NER on the conll2003 dataset. It achieves the following results on the evaluation set:
- Loss: 0.0723
- Precision: 0.9327
- Recall: 0.9405
- F1: 0.9366
- Accuracy: 0.9851
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: 64
- eval_batch_size: 64
- 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 |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 220 | 0.0754 | 0.9225 | 0.9296 | 0.9260 | 0.9831 |
| No log | 2.0 | 440 | 0.0688 | 0.9319 | 0.9407 | 0.9363 | 0.9849 |
| 0.0717 | 3.0 | 660 | 0.0723 | 0.9327 | 0.9405 | 0.9366 | 0.9851 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.12.0
- Datasets 2.7.1
- Tokenizers 0.11.0