<!-- 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. -->
chinese-macbert-base-finetuned-ner
This model is a fine-tuned version of hfl/chinese-macbert-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2420
- F1: 0.9224
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: 5e-05
- train_batch_size: 57
- eval_batch_size: 57
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| 3.6141 | 1.0 | 1 | 2.6454 | 0.0 |
| 2.7076 | 2.0 | 2 | 2.0034 | 0.0 |
| 2.0979 | 3.0 | 3 | 1.6276 | 0.0 |
| 1.7264 | 4.0 | 4 | 1.3419 | 0.3522 |
| 1.4691 | 5.0 | 5 | 1.1239 | 0.4091 |
| 1.2504 | 6.0 | 6 | 0.9532 | 0.5514 |
| 1.0798 | 7.0 | 7 | 0.8129 | 0.5895 |
| 0.9279 | 8.0 | 8 | 0.6987 | 0.625 |
| 0.8179 | 9.0 | 9 | 0.6081 | 0.6392 |
| 0.7202 | 10.0 | 10 | 0.5346 | 0.6667 |
| 0.6377 | 11.0 | 11 | 0.4731 | 0.7451 |
| 0.5751 | 12.0 | 12 | 0.4226 | 0.7925 |
| 0.5202 | 13.0 | 13 | 0.3804 | 0.7685 |
| 0.4733 | 14.0 | 14 | 0.3447 | 0.7928 |
| 0.44 | 15.0 | 15 | 0.3145 | 0.8509 |
| 0.4047 | 16.0 | 16 | 0.2899 | 0.8918 |
| 0.3773 | 17.0 | 17 | 0.2707 | 0.8966 |
| 0.353 | 18.0 | 18 | 0.2563 | 0.9052 |
| 0.3413 | 19.0 | 19 | 0.2468 | 0.9224 |
| 0.3314 | 20.0 | 20 | 0.2420 | 0.9224 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.12.0+cu102
- Datasets 1.18.4
- Tokenizers 0.12.1
