Distilbert Base Multilingual Cased Toxicity

🚀 citizenlab/distilbert-base-multilingual-cased-toxicity

A multilingual Distil - Bert model sequence classifier trained on the JIGSAW Toxic Comment Classification Challenge dataset, capable of classifying text toxicity.

🚀 Quick Start

Model Information

Widget Examples

Here are some widget examples to quickly understand the model's functionality:

• Example 1:
  • Text: "this is a lovely message"
  • Multi - class: false
• Example 2:
  • Text: "you are an idiot and you and your family should go back to your country"
  • Multi - class: false

💻 Usage Examples

Basic Usage

from transformers import pipeline

model_path = "citizenlab/distilbert-base-multilingual-cased-toxicity"

toxicity_classifier = pipeline("text-classification", model=model_path, tokenizer=model_path)
toxicity_classifier("this is a lovely message")
> [{'label': 'not_toxic', 'score': 0.9954179525375366}]

toxicity_classifier("you are an idiot and you and your family should go back to your country")
> [{'label': 'toxic', 'score': 0.9948776960372925}]

📚 Documentation

Evaluation

Accuracy

  Accuracy Score = 0.9425
F1 Score (Micro) = 0.9450549450549449
F1 Score (Macro) = 0.8491432341169309