metadata
library_name: transformers
base_model: huawei-noah/TinyBERT_General_4L_312D
tags:
- generated_from_trainer
datasets:
- conll2003
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: TinyBERT-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: 0.8465303458777463
- name: Recall
type: recall
value: 0.870679046873252
- name: F1
type: f1
value: 0.8584348977003253
- name: Accuracy
type: accuracy
value: 0.9670516466233497
TinyBERT-finetuned-NER
This model is a fine-tuned version of huawei-noah/TinyBERT_General_4L_312D on the conll2003 dataset. It achieves the following results on the evaluation set:
- Loss: 0.1232
- Precision: 0.8465
- Recall: 0.8707
- F1: 0.8584
- Accuracy: 0.9671
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: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.5173 | 1.0 | 878 | 0.2116 | 0.7429 | 0.7756 | 0.7589 | 0.9493 |
| 0.196 | 2.0 | 1756 | 0.1528 | 0.8262 | 0.8383 | 0.8323 | 0.9620 |
| 0.1444 | 3.0 | 2634 | 0.1355 | 0.8447 | 0.8606 | 0.8526 | 0.9652 |
| 0.116 | 4.0 | 3512 | 0.1255 | 0.8452 | 0.8660 | 0.8555 | 0.9663 |
| 0.1116 | 5.0 | 4390 | 0.1232 | 0.8465 | 0.8707 | 0.8584 | 0.9671 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.5.0+cu121
- Datasets 3.1.0
- Tokenizers 0.19.1