https://github.com/flurixoww/bert_toxic_filter
Model for filtering hate speech based on BERT.
https://github.com/flurixoww/bert_toxic_filter
bert bert-model hate-speech-detection hatespeech machine-learning python pytorch
Last synced: 3 months ago
JSON representation
Model for filtering hate speech based on BERT.
- Host: GitHub
- URL: https://github.com/flurixoww/bert_toxic_filter
- Owner: flurixoww
- License: mit
- Created: 2025-02-08T11:28:05.000Z (3 months ago)
- Default Branch: master
- Last Pushed: 2025-02-15T19:15:20.000Z (3 months ago)
- Last Synced: 2025-02-15T20:24:51.990Z (3 months ago)
- Topics: bert, bert-model, hate-speech-detection, hatespeech, machine-learning, python, pytorch
- Language: Jupyter Notebook
- Homepage:
- Size: 48.8 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Hate speech filter based on BERT
## Requirements
### For training your own model (main.ipynb)
| Tools | Version | Installation |
|:----------------:|:-------:|:----------------------------------------------------------:|
| **Pandas** | latest | pip install pandas |
| **Pytorch** | latest | [Latest version](https://pytorch.org/get-started/locally/) |
| **Transofrmers** | latest | pip install transformers |
| **Accelerate** | 0.26.0 | pip install accelerate |
### For using already trained model (test.ipynb)
| Tools | Version | Installation |
|:----------------:|:-------:|:------------------------:|
| **Transformers** | latest | pip install transformers |
| **Pytorch** | latest | [Latest version](https://pytorch.org/get-started/locally/) |
## Optional
* ### You can find the trained model on hugging face ([click](https://huggingface.co/flurixoww/bert_toxic_filter/tree/main))
* ### Dataset I used for training ([click](https://www.kaggle.com/datasets/julian3833/jigsaw-toxic-comment-classification-challenge)) you can also download train and test .csv from dataset directory
## Roadmap
* ### Multilingual model
* ### Create large, basic and small model
* ### Create API
* ### Distribute different types of hate speech(bulling, threat, discrimination)