Toxic Comment Classifier

Model Details

• Base model: distilbert-base-uncased
• Task: Multi-label text classification (6 labels)
• Labels: toxic, severe_toxic, obscene, threat, insult, identity_hate
• Training data: Jigsaw Toxic Comment Classification (159,571 comments)
• Framework: HuggingFace Transformers + Trainer API
• Training time: ~28 minutes on T4 GPU

Performance

F1 is lower due to severe class imbalance — threat label has only 478 positive examples out of 159k. ROC-AUC of 0.990 reflects true model quality.

Latency Benchmark (GPU, single request)

Live API

• Swagger UI: https://chandkr123-toxic-classifier-api.hf.space/docs
• Predict endpoint: https://chandkr123-toxic-classifier-api.hf.space/predict

Example Predictions

Limitations

• Trained on Wikipedia comments, may not generalise to all domains
• Severe class imbalance: threat label has only 478 training examples
• Predictions above 0.5 threshold are flagged (adjustable)
• Not suitable for production content moderation without human review