znafian/multilabel-text-classifier

Multilabel Text Classification Model

This model is fine-tuned from DistilBERT for multilabel text classification.

Model Details

• Base Model: distilbert-base-uncased
• Parameters: ~774M (under 1B limit)
• Task: Multilabel Classification
• Number of Labels: 14
• Max Sequence Length: 157

Training Details

• Partial layer freezing applied
• Trained for 3 epochs
• Learning rate: 2e-5
• Batch size: 16

Usage

import torch
from transformers import DistilBertTokenizer

# Load model and tokenizer
tokenizer = DistilBertTokenizer.from_pretrained('multilabel-text-classifier')
model = torch.load('multilabel-text-classifier/pytorch_model.bin')

# Example prediction
text = "Your input text here"
inputs = tokenizer(text, return_tensors="pt", max_length=157, truncation=True, padding=True)
with torch.no_grad():
    outputs = model(**inputs)
    predictions = torch.sigmoid(outputs).numpy()

Labels

['amusement', 'anger', 'annoyance', 'caring', 'confusion', 'disappointment', 'disgust', 'embarrassment', 'excitement', 'fear', 'gratitude', 'joy', 'love', 'sadness']