---
language:
- ms
library_name: transformers
---

Safe for Work Classifier Model for Malaysian Data(WORK IN PROGRESS)


Finetuned https://huggingface.co/mesolitica/malaysian-mistral-191M-MLM-512 with Malaysian NSFW data.


![Image in a markdown cell](https://github.com/mesolitica/malaysian-llmops/raw/main/e2e.png)



### How to use

```python
from classifier import MistralForSequenceClassification
model = MistralForSequenceClassification.from_pretrained('malaysia-ai/malaysian-sfw-classifier')
```


```
                                precision    recall  f1-score   support

                       racist    0.89222   0.90653   0.89931      1808
              religion insult    0.89681   0.88123   0.88895      3570
psychiatric or mental illness    0.93524   0.86953   0.90119      5580
                       sexist    0.77826   0.80563   0.79171      1564
                   harassment    0.79693   0.78777   0.79232      2436
                  informative    0.65000   0.76098   0.70112      1640
                safe for work    0.73716   0.78421   0.75996      1849

                     accuracy                        0.84100     18447
                    macro avg    0.81237   0.82798   0.81922     18447
                 weighted avg    0.84680   0.84100   0.84304     18447

```