File size: 1,546 Bytes
1f58e2d |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 |
---
tags:
- spam
- classification
- bert
- pytorch
- comment-filter
- text-classification
- content-moderation
- social-media
license: mit
language: en
datasets: custom
widget:
- text: "Click here to win a free iPhone!"
- text: "Great video, thanks for sharing!"
- text: "Follow me for daily crypto tips 💰"
- text: "This tutorial saved my life, thank you!"
- text: "🔥 Get rich quick! Limited-time offer!"
---
# 📦 Spam Detector — `vibehq/spam-detector`
A BERT-based spam classifier fine-tuned to detect **spam and promotional content** in social media-style comments. Trained on real-world-like comment data including giveaways, scams, promotions, and genuine engagement.
Perfect for content moderation on platforms like:
- YouTube
- Instagram
- Discord
- Reddit
- Facebook
- Forums or blogs
---
## 🚀 How to Use
```python
from transformers import BertTokenizer, BertForSequenceClassification
import torch
# Load model and tokenizer
model = BertForSequenceClassification.from_pretrained("vibehq/spam-detector")
tokenizer = BertTokenizer.from_pretrained("vibehq/spam-detector")
def predict_spam(comment):
inputs = tokenizer(comment, return_tensors='pt', max_length=128, padding='max_length', truncation=True)
with torch.no_grad():
outputs = model(**inputs)
prediction = torch.argmax(outputs.logits, dim=-1).item()
return "Spam" if prediction == 1 else "Non-Spam"
# Example
print(predict_spam("Subscribe to my channel for more giveaways!")) |