<!-- 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-wikiann-zh-ner-2
This model is a fine-tuned version of bert-base-chinese on the wikiann dataset. It achieves the following results on the evaluation set:
- Loss: 0.2036
- Precision: 0.7577
- Recall: 0.7792
- F1: 0.7683
- Accuracy: 0.9386
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: 1
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.555 | 0.16 | 400 | 0.3120 | 0.5949 | 0.7117 | 0.6481 | 0.9041 |
0.2944 | 0.32 | 800 | 0.2669 | 0.7013 | 0.7052 | 0.7032 | 0.9230 |
0.2814 | 0.48 | 1200 | 0.2354 | 0.7078 | 0.7601 | 0.7330 | 0.9317 |
0.2351 | 0.64 | 1600 | 0.2271 | 0.7295 | 0.7715 | 0.7499 | 0.9336 |
0.2101 | 0.8 | 2000 | 0.2148 | 0.7478 | 0.7764 | 0.7618 | 0.9369 |
0.23 | 0.96 | 2400 | 0.2059 | 0.7586 | 0.7752 | 0.7668 | 0.9385 |
Framework versions
- Transformers 4.27.4
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3