Spam Mail Classifier

🚀 SPAM Mail Classifier

This model is fine - tuned from microsoft/Multilingual-MiniLM-L12-H384 to classify email subjects as SPAM or NOSPAM, offering an effective solution for multilingual spam detection in email subjects.

✨ Features

• Fine - tuned from microsoft/Multilingual-MiniLM-L12-H384 for text classification.
• Capable of distinguishing between 2 classes: SPAM and NOSPAM.
• Supports multilingual text, making it suitable for a wide range of users.

📦 Installation

No explicit installation steps are provided in the original README. However, to use this model, you need to have the transformers library installed. You can install it using the following command:

pip install transformers

💻 Usage Examples

Basic Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification

model_name = "Goodmotion/spam-mail-classifier"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(
    model_name
)

text = "Félicitations ! Vous avez gagné un iPhone."
inputs = tokenizer(text, return_tensors="pt")
outputs = model(**inputs)
print(outputs.logits)

Advanced Usage

import torch
from transformers import AutoTokenizer, AutoModelForSequenceClassification

model_name = "Goodmotion/spam-mail-classifier"

tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)

texts = [
'Join us for a webinar on AI innovations',
'Urgent: Verify your account immediately.',
'Meeting rescheduled to 3 PM',
'Happy Birthday!',
'Limited time offer: Act now!',
'Join us for a webinar on AI innovations',
'Claim your free prize now!',
'You have unclaimed rewards waiting!',
'Weekly newsletter from Tech World',
'Update on the project status',
'Lunch tomorrow at 12:30?',
'Get rich quick with this amazing opportunity!',
'Invoice for your recent purchase',
'Don\'t forget: Gym session at 6 AM',
'Join us for a webinar on AI innovations',
'bonjour comment allez vous ?',
'Documents suite à notre rendez-vous',
'Valentin Dupond mentioned you in a comment',
'Bolt x Supabase = 🤯',
'Modification site web de la société',
'Image de mise en avant sur les articles',
'Bring new visitors to your site',
'Le Cloud Éthique sans bullshit',
'Remix Newsletter #25: React Router v7',
'Votre essai auprès de X va bientôt prendre fin',
'Introducing a Google Docs integration, styles and more in Claude.ai',
'Carte de crédit sur le point d’expirer sur Cloudflare'
]
inputs = tokenizer(texts, padding=True, truncation=True, max_length=128, return_tensors="pt")
outputs = model(**inputs)

# Convert the logits to probabilities with softmax
logits = outputs.logits
probabilities = torch.softmax(logits, dim=1)

# Decode the classes for each text
labels = ["NOSPAM", "SPAM"]  # Mapping of indices to labels
results = [
    {"text": text, "label": labels[torch.argmax(prob).item()], "confidence": prob.max().item()}
    for text, prob in zip(texts, probabilities)
]

# Display the results
for result in results:
    print(f"Text: {result['text']}")
    print(f"Result: {result['label']} (Confidence: {result['confidence']:.2%})\n")

📚 Documentation

Model Details

📄 License

This model is licensed under the Apache-2.0 license.