metadata
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: KoELECTRA-small-v3-modu-ner
results: []
language:
- ko
pipeline_tag: token-classification
examples: null
widget:
- text: 서울역으로 안내해줘
example_title: Sentence_1
- text: 에어컨 온도를 3도 올려줘
example_title: Sentence_2
KoELECTRA-small-v3-modu-ner
This model is a fine-tuned version of monologg/koelectra-small-v3-discriminator on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1354
- Precision: 0.8084
- Recall: 0.8311
- F1: 0.8196
- Accuracy: 0.9599
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 7575
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 3788 | 0.2991 | 0.6481 | 0.6373 | 0.6426 | 0.9229 |
| No log | 2.0 | 7576 | 0.1904 | 0.7479 | 0.7418 | 0.7448 | 0.9438 |
| No log | 3.0 | 11364 | 0.1620 | 0.7577 | 0.7940 | 0.7754 | 0.9502 |
| No log | 4.0 | 15152 | 0.1505 | 0.7890 | 0.7982 | 0.7936 | 0.9544 |
| No log | 5.0 | 18940 | 0.1417 | 0.7905 | 0.8163 | 0.8032 | 0.9563 |
| No log | 6.0 | 22728 | 0.1392 | 0.7914 | 0.8250 | 0.8079 | 0.9572 |
| No log | 7.0 | 26516 | 0.1363 | 0.8060 | 0.8231 | 0.8144 | 0.9589 |
| No log | 8.0 | 30304 | 0.1367 | 0.8035 | 0.8294 | 0.8162 | 0.9592 |
| No log | 9.0 | 34092 | 0.1349 | 0.8085 | 0.8296 | 0.8189 | 0.9597 |
| 0.2299 | 10.0 | 37880 | 0.1354 | 0.8084 | 0.8311 | 0.8196 | 0.9599 |
Framework versions
- Transformers 4.27.3
- Pytorch 1.13.1+cu116
- Datasets 2.10.1
- Tokenizers 0.13.2