<!-- 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. -->
albert-tiny-chinese-david-ner
This model is a fine-tuned version of ckiplab/albert-tiny-chinese-ws on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3415
- Precision: 0.6062
- Recall: 0.6690
- F1: 0.6361
- Accuracy: 0.9055
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
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.1796 | 1.4 | 500 | 0.3368 | 0.6201 | 0.6586 | 0.6388 | 0.9046 |
| 0.1374 | 2.8 | 1000 | 0.3415 | 0.6062 | 0.6690 | 0.6361 | 0.9055 |
Framework versions
- Transformers 4.29.0.dev0
- Pytorch 1.10.1+cu113
- Datasets 2.11.0
- Tokenizers 0.13.3
