nabeelshan/distilbert-finetuned-agnews

DistilBERT fine-tuned on AG News for Text Classification

This is a distilbert-base-uncased model fine-tuned for text classification on the AG News dataset. It classifies news headlines into one of four categories: World, Sports, Business, and Sci/Tech.

This model was trained as part of a larger project, End-to-End NLP: From-Scratch Transformer vs. Deployed DistilBERT, which compares the performance of a Transformer built from scratch against this fine-tuned DistilBERT model.

🚀 Live Demo

You can interact with this model directly in the browser using the live Gradio demo hosted on Hugging Face Spaces:

Model Details

• Model type: distilbert-base-uncased, fine-tuned for sequence classification.
• Language: English
• Dataset: AG News
• Performance: Achieved 94.79% accuracy on the test set.
• Labels:
  • 0: World
  • 1: Sports
  • 2: Business
  • 3: Sci/Tech

How to Use

You can use this model directly with a text-classification pipeline.

from transformers import pipeline

# Load the classification pipeline
classifier = pipeline(
    "text-classification",
    model="nabeelshan/distilbert-finetuned-agnews",
    return_all_scores=True
)

# Example news headline
text = "Formula 1: Hamilton wins the British Grand Prix after a dramatic last lap."

# Get prediction
result = classifier(text)
print(result)