|
|
--- |
|
|
license: apache-2.0 |
|
|
datasets: |
|
|
- UngLong/ViSIR |
|
|
language: |
|
|
- vi |
|
|
metrics: |
|
|
- f1 |
|
|
- recall |
|
|
- precision |
|
|
base_model: |
|
|
- uitnlp/CafeBERT |
|
|
pipeline_tag: text-classification |
|
|
tags: |
|
|
- socialmedia |
|
|
- toxiccomment |
|
|
- classification_toxic_comment |
|
|
- transformer |
|
|
library_name: transformers |
|
|
--- |
|
|
# CafeBERT Fine-tuned for Toxic Classification |
|
|
This model is a fine-tuned version of CafeBERT on a toxic comment classification task. |
|
|
|
|
|
## Usage |
|
|
```python |
|
|
from transformers import AutoTokenizer, AutoModelForSequenceClassification |
|
|
|
|
|
tokenizer = AutoTokenizer.from_pretrained("UngLong/cafebert-classification-ft") |
|
|
model = AutoModelForSequenceClassification.from_pretrained("UngLong/cafebert-classification-ft") |
|
|
|
