SafeOrNot Safety Classification Model
This model detects whether a given text is safe or unsafe based on context-aware classification with a focus on Indian and Hinglish language usage.
Use Case
- Community safety moderation
- Location/place safety scoring (e.g. Maps-style apps)
- Harassment and abuse detection in Indian contexts
Model Details
- Architecture: Transformer-based (BERT family)
- Task: Binary text classification (Safe / Unsafe)
- Trained on: Custom curated Indian safety dataset
How to Use
from transformers import AutoTokenizer, AutoModelForSequenceClassification
model_id = "aryaman1222/safeornot-safety-model"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForSequenceClassification.from_pretrained(model_id)
- Downloads last month
- 6