|
--- |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: bert-base-chinese-david-ner |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# bert-base-chinese-david-ner |
|
|
|
This model is a fine-tuned version of [bert-base-chinese](https://huggingface.co/bert-base-chinese) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0677 |
|
- Precision: 0.8954 |
|
- Recall: 0.8935 |
|
- F1: 0.8945 |
|
- Accuracy: 0.9830 |
|
|
|
## 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 |
|
- lr_scheduler_warmup_ratio: 0.1 |
|
- num_epochs: 3 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| 1.1093 | 0.14 | 50 | 0.5193 | 0.16 | 0.1304 | 0.1437 | 0.8199 | |
|
| 0.3453 | 0.28 | 100 | 0.1877 | 0.5811 | 0.6696 | 0.6222 | 0.9390 | |
|
| 0.2117 | 0.42 | 150 | 0.1344 | 0.6907 | 0.7087 | 0.6996 | 0.9526 | |
|
| 0.193 | 0.56 | 200 | 0.1159 | 0.7228 | 0.7370 | 0.7298 | 0.9593 | |
|
| 0.1625 | 0.7 | 250 | 0.1191 | 0.7367 | 0.7543 | 0.7454 | 0.9603 | |
|
| 0.1302 | 0.84 | 300 | 0.1448 | 0.7332 | 0.7587 | 0.7457 | 0.9550 | |
|
| 0.1396 | 0.98 | 350 | 0.0899 | 0.8226 | 0.8370 | 0.8297 | 0.9720 | |
|
| 0.0966 | 1.12 | 400 | 0.0918 | 0.8240 | 0.8348 | 0.8294 | 0.9732 | |
|
| 0.1077 | 1.26 | 450 | 0.0824 | 0.7944 | 0.8565 | 0.8243 | 0.9742 | |
|
| 0.0895 | 1.4 | 500 | 0.0793 | 0.8121 | 0.8457 | 0.8285 | 0.9761 | |
|
| 0.0968 | 1.54 | 550 | 0.0797 | 0.8409 | 0.85 | 0.8454 | 0.9773 | |
|
| 0.1172 | 1.68 | 600 | 0.0694 | 0.8422 | 0.8587 | 0.8504 | 0.9792 | |
|
| 0.0974 | 1.82 | 650 | 0.0710 | 0.8354 | 0.8609 | 0.8480 | 0.9780 | |
|
| 0.0941 | 1.96 | 700 | 0.0650 | 0.8543 | 0.8543 | 0.8543 | 0.9804 | |
|
| 0.0755 | 2.09 | 750 | 0.0673 | 0.8789 | 0.8674 | 0.8731 | 0.9816 | |
|
| 0.0559 | 2.23 | 800 | 0.0744 | 0.8544 | 0.8674 | 0.8608 | 0.9792 | |
|
| 0.0689 | 2.37 | 850 | 0.0707 | 0.8596 | 0.8652 | 0.8624 | 0.9799 | |
|
| 0.0525 | 2.51 | 900 | 0.0677 | 0.8954 | 0.8935 | 0.8945 | 0.9830 | |
|
| 0.0631 | 2.65 | 950 | 0.0646 | 0.8886 | 0.8848 | 0.8867 | 0.9830 | |
|
| 0.0699 | 2.79 | 1000 | 0.0630 | 0.8932 | 0.8913 | 0.8923 | 0.9840 | |
|
| 0.053 | 2.93 | 1050 | 0.0636 | 0.8950 | 0.8891 | 0.8920 | 0.9837 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.29.0.dev0 |
|
- Pytorch 1.10.1+cu113 |
|
- Datasets 2.11.0 |
|
- Tokenizers 0.13.3 |
|
|
