YAML Metadata
Error:
"datasets[0]" must be a string
YAML Metadata
Error:
"datasets[1]" must be a string
YAML Metadata
Error:
"license" does not match any of the allowed types
YAML Metadata
Error:
"language[0]" must be a string
YAML Metadata
Error:
"language[1]" must be a string
YAML Metadata
Error:
"tags[0]" must be a string
YAML Metadata
Error:
"tags[1]" must be a string
YAML Metadata
Error:
"tags[2]" must be a string
Toxic language detection
Model description
A toxic language detection model trained on tweets. The base model is Roberta-large. For more information, including the training data, limitations and bias, please refer to the paper and Github repo for more details.
How to use
Note that LABEL_1 means toxic and LABEL_0 means non-toxic in the output.
from transformers import pipeline
classifier = pipeline("text-classification",model='Xuhui/ToxDect-roberta-large', return_all_scores=True)
prediction = classifier("You are f**king stupid!", )
print(prediction)
"""
Output:
[[{'label': 'LABEL_0', 'score': 0.002632011892274022}, {'label': 'LABEL_1', 'score': 0.9973680377006531}]]
"""
Training procedure
The random seed for this model is 22. For other details, please refer to the Github repo for more details.
BibTeX entry and citation info
@inproceedings{zhou-etal-2020-debiasing,
title = {Challenges in Automated Debiasing for Toxic Language Detection},
author = {Zhou, Xuhui and Sap, Maarten and Swayamdipta, Swabha and Choi, Yejin and Smith, Noah A.},
booktitle = {EACL},
abbr = {EACL},
html = {https://www.aclweb.org/anthology/2021.eacl-main.274.pdf},
code = {https://github.com/XuhuiZhou/Toxic_Debias},
year = {2021},
bibtex_show = {true},
selected = {true}
}
- Downloads last month
- 360
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.