metadata
base_model: gyr66/RoBERTa-ext-large-chinese-finetuned-ner
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: Ernie-3.0-large-crf-chinese-finetuned-ner
results: []
datasets:
- gyr66/privacy_detection
language:
- zh
library_name: transformers
pipeline_tag: token-classification
RoBERTa-ext-large-crf-chinese-finetuned-ner
This model is a fine-tuned version of gyr66/RoBERTa-ext-large-chinese-finetuned-ner on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5907
- Precision: 0.7278
- Recall: 0.75
- F1: 0.7387
- Accuracy: 0.9629
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: 4
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.0061 | 1.0 | 503 | 0.6739 | 0.6747 | 0.7457 | 0.7084 | 0.9608 |
| 0.0078 | 2.0 | 1006 | 0.6343 | 0.7083 | 0.7518 | 0.7294 | 0.9622 |
| 0.0072 | 3.0 | 1509 | 0.6237 | 0.6867 | 0.7621 | 0.7224 | 0.9607 |
| 0.0052 | 4.0 | 2012 | 0.5929 | 0.7136 | 0.7616 | 0.7368 | 0.9635 |
| 0.0031 | 5.0 | 2515 | 0.5907 | 0.7278 | 0.75 | 0.7387 | 0.9629 |
| 0.0014 | 6.0 | 3018 | 0.6080 | 0.7172 | 0.7558 | 0.7360 | 0.9636 |
| 0.001 | 7.0 | 3521 | 0.6179 | 0.7198 | 0.7586 | 0.7387 | 0.9637 |
| 0.0005 | 8.0 | 4024 | 0.6208 | 0.7211 | 0.7518 | 0.7361 | 0.9632 |
| 0.0004 | 9.0 | 4527 | 0.6169 | 0.7271 | 0.7487 | 0.7378 | 0.9636 |
| 0.0002 | 10.0 | 5030 | 0.6202 | 0.7266 | 0.7495 | 0.7379 | 0.9636 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0