Downloads · 30 days
16
2% of all-time downloads
bert-base-chinese-ws-finetuned-ner_all
HYM/bert-base-chinese-ws-finetuned-ner_all
bert-base-chinese-ws-finetuned-ner_all is a token classification model from HYM. Use it when you need labels on individual words, such as names. It is set up for transformers. The card lists the license as gpl-3.0.
should probably proofread and complete it, then remove this comment. --
All-time downloads
830
Public
Repo size
4.9 GB
Likes
2
Public
Repo makeup
Click a slice to open those files.
.bin407 MB · 100%
At a glance
- Task
- Token Classification
- Library
- transformers
- License
- gpl-3.0
- Model type
- bert
- Access
- Public
- Created
- Aug 24, 2022
- Updated
- Aug 24, 2022
- SHA
- 81a55e91
Try a prompt
Model card
From the Hugging Face model README
<!-- 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-ws-finetuned-ner_all
This model is a fine-tuned version of ckiplab/bert-base-chinese-ws on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0330
- Precision: 0.9723
- Recall: 0.9734
- F1: 0.9728
- Accuracy: 0.9879
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: 1e-05
- train_batch_size: 18
- eval_batch_size: 18
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.0648 | 0.29 | 500 | 0.0524 | 0.9586 | 0.9572 | 0.9579 | 0.9813 |
| 0.0509 | 0.59 | 1000 | 0.0460 | 0.9615 | 0.9628 | 0.9622 | 0.9832 |
| 0.0478 | 0.88 | 1500 | 0.0429 | 0.9624 | 0.9660 | 0.9642 | 0.9840 |
| 0.0417 | 1.17 | 2000 | 0.0409 | 0.9650 | 0.9680 | 0.9665 | 0.9851 |
| 0.0402 | 1.47 | 2500 | 0.0387 | 0.9662 | 0.9693 | 0.9677 | 0.9856 |
| 0.0378 | 1.76 | 3000 | 0.0359 | 0.9699 | 0.9717 | 0.9708 | 0.9869 |
| 0.0385 | 2.05 | 3500 | 0.0353 | 0.9703 | 0.9718 | 0.9710 | 0.9871 |
| 0.0337 | 2.34 | 4000 | 0.0341 | 0.9709 | 0.9731 | 0.9720 | 0.9875 |
| 0.0348 | 2.64 | 4500 | 0.0333 | 0.9721 | 0.9733 | 0.9727 | 0.9878 |
| 0.0346 | 2.93 | 5000 | 0.0331 | 0.9722 | 0.9735 | 0.9729 | 0.9879 |
Framework versions
- Transformers 4.13.0
- Pytorch 1.8.0+cu111
- Datasets 2.4.0
- Tokenizers 0.10.3
- Task
- Token Classification
- Library
- transformers
- Type
- bert
- License
- gpl-3.0
- Created
- Aug 24, 2022
- Updated
- Aug 24, 2022