metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- conll2003
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: distilbert-base-uncased-finetuned-ner
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: conll2003
type: conll2003
args: conll2003
metrics:
- name: Precision
type: precision
value: 0.9300908486594284
- name: Recall
type: recall
value: 0.9391430808815304
- name: F1
type: f1
value: 0.9345950459226274
- name: Accuracy
type: accuracy
value: 0.9842407104389407
distilbert-base-uncased-finetuned-ner
This model is a fine-tuned version of distilbert-base-uncased on the conll2003 dataset. It achieves the following results on the evaluation set:
- Loss: 0.0589
- Precision: 0.9301
- Recall: 0.9391
- F1: 0.9346
- Accuracy: 0.9842
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: 16
- eval_batch_size: 16
- 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 |
|---|---|---|---|---|---|---|---|
| 0.2402 | 1.0 | 878 | 0.0692 | 0.9177 | 0.9248 | 0.9213 | 0.9815 |
| 0.0506 | 2.0 | 1756 | 0.0600 | 0.9249 | 0.9361 | 0.9305 | 0.9836 |
| 0.0304 | 3.0 | 2634 | 0.0589 | 0.9301 | 0.9391 | 0.9346 | 0.9842 |
Framework versions
- Transformers 4.12.5
- Pytorch 1.10.0+cu111
- Datasets 1.16.1
- Tokenizers 0.10.3