Views
No views yet
1from transformers import AutoTokenizer, AutoModelForSequenceClassification
2import torch
3
4# Load model and tokenizer
5model = AutoModelForSequenceClassification.from_pretrained("indic-toxicity-detector")
6tokenizer = AutoTokenizer.from_pretrained("indic-toxicity-detector")
7
8# Predict
9def predict_toxicity(text):
10 inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=128)
11 outputs = model(**inputs)
12 probabilities = torch.softmax(outputs.logits, dim=-1)
13 predicted_class = torch.argmax(probabilities, dim=-1).item()
14 confidence = probabilities[0][predicted_class].item()
15
16 label = model.config.id2label[predicted_class]
17 return {"label": label, "confidence": confidence}
18
19# Example
20result = predict_toxicity("You are amazing!")
21print(result) # {'label': 'non-toxic', 'confidence': 0.95}1@misc{indic-toxicity-detector,
2 author = {Your Name},
3 title = {IndicBERT Multilingual Toxicity Detector},
4 year = {2025},
5 publisher = {Hugging Face},
6 url = {https://huggingface.co/indic-toxicity-detector}
7}