🧠 Sentiment Analysis with DistilBERT

This project fine-tunes a DistilBERT transformer model on the Amazon Reviews dataset to classify text into positive or negative sentiment. Built using Hugging Face Transformers and PyTorch with optional GPU acceleration.

Dataset Used: https://www.kaggle.com/datasets/bittlingmayer/amazonreviews

🚀 Features

• ✅ Binary sentiment classification (Positive / Negative)

• 📦 Fine-tuned on Amazon Reviews dataset

• 🧠 Uses Hugging Face DistilBERT and Trainer API

• ⚡ GPU support for accelerated training

• 🔍 Real-world review testing using inference pipeline

• 📊 Accuracy: ~92% on validation set

🛠️ Setup Instructions

1.  Clone  the  repository
git  clone  https://github.com/parthudawant/sentiment-analysis-using-distilbert.git
cd  sentiment-analysis-using-distilbert

2.  Install  dependencies
pip  install  -r  requirements.txt

✍️ Author

Made with ❤️ by Parth Udawant