metadata
library_name: transformers
license: apache-2.0
base_model: bert-base-cased
tags:
- generated_from_trainer
datasets:
- conll2003
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: bert-finetuned-ner
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: conll2003
type: conll2003
config: conll2003
split: validation
args: conll2003
metrics:
- name: Precision
type: precision
value: 0.9325508348487354
- name: Recall
type: recall
value: 0.9493436553349041
- name: F1
type: f1
value: 0.9408723209073472
- name: Accuracy
type: accuracy
value: 0.9857538117383882
bert-finetuned-ner
This model is a fine-tuned version of bert-base-cased on the conll2003 dataset. It achieves the following results on the evaluation set:
- Loss: 0.0618
- Precision: 0.9326
- Recall: 0.9493
- F1: 0.9409
- Accuracy: 0.9858
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: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.0756 | 1.0 | 1756 | 0.0697 | 0.8940 | 0.9308 | 0.9120 | 0.9808 |
| 0.0347 | 2.0 | 3512 | 0.0663 | 0.9277 | 0.9441 | 0.9359 | 0.9841 |
| 0.0206 | 3.0 | 5268 | 0.0618 | 0.9326 | 0.9493 | 0.9409 | 0.9858 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.5.0+cu121
- Datasets 3.1.0
- Tokenizers 0.19.1