metadata
license: apache-2.0
base_model: bert-base-cased
tags:
- generated_from_trainer
metrics:
- precision
- recall
- accuracy
model-index:
- name: bert-finetuned-ner
results: []
datasets:
- conll2002
bert-finetuned-ner
This model is a fine-tuned version of bert-base-cased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1793
- Precision: 0.7556
- Recall: 0.8015
- Overall F1: 0.7779
- Accuracy: 0.9669
Model description
This is a model trained on the conll2002 dataset that can be used for Named Entity Recognition. This model uses bert-base-cased as the underlying encoder.
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 | Overall F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.0221 | 1.0 | 1041 | 0.1840 | 0.7401 | 0.7976 | 0.7678 | 0.9662 |
| 0.0362 | 2.0 | 2082 | 0.1571 | 0.7490 | 0.8028 | 0.7750 | 0.9662 |
| 0.0187 | 3.0 | 3123 | 0.1793 | 0.7556 | 0.8015 | 0.7779 | 0.9669 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1