pos_thai
This model is a fine-tuned version of Geotrend/bert-base-th-cased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0935
- Precision: 0.9525
- Recall: 0.9540
- F1: 0.9533
- Accuracy: 0.9693
Model description
This model is train on thai pos_tag datasets to help with pos tagging in Thai language.
Example
from transformers import AutoModelForTokenClassification, AutoTokenizer, TokenClassificationPipeline
model = AutoModelForTokenClassification.from_pretrained("lunarlist/pos_thai")
tokenizer = AutoTokenizer.from_pretrained("lunarlist/pos_thai")
pipeline = TokenClassificationPipeline(model=model, tokenizer=tokenizer, grouped_entities=True)
outputs = pipeline("ภาษาไทย ง่าย นิดเดียว")
print(outputs)
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.1124 | 1.0 | 7344 | 0.1048 | 0.9505 | 0.9478 | 0.9492 | 0.9670 |
| 0.0866 | 2.0 | 14688 | 0.0935 | 0.9525 | 0.9540 | 0.9533 | 0.9693 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1