metadata

title: ReviewSummariser
emoji: 👀
colorFrom: pink
colorTo: red
sdk: gradio
sdk_version: 5.35.0
app_file: app.py
pinned: false

📝 Review Summariser GPT - Config 1

This is a Gradio-powered web application that uses a fine-tuned T5 model to summarize long product reviews into short, helpful summaries — and also predicts the overall sentiment (Positive/Negative) of the review.

🔧 Built with: 🤗 Transformers, Gradio, Hugging Face Spaces
🧠 Model: Fine-tuned from t5-small on a custom review-summary dataset
🔍 Sentiment: Powered by distilbert-base-uncased-finetuned-sst-2-english

🚀 Demo

👉 Live on Hugging Face Spaces:
🔗 Launch App

📦 Features

✂️ Summarizes long product reviews (e.g., from Amazon, Flipkart, etc.)
📊 Predicts sentiment using a pre-trained classifier
🖼️ Clean Gradio UI with examples and collapsible "About" section
⚡ Optimized for GPU inference (but also works on CPU)

🧪 Example

Input Review:

This product leaks water and smells like burnt plastic. Total waste of money.

→ Summary:

Leaks and has a bad odor. Not recommended.

→ Sentiment:

NEGATIVE (99.8%)

📁 Files in This Repo

File	Description
`app.py`	Main app file with Gradio Blocks UI
`requirements.txt`	Python dependencies for the Space
`thumbnail.png`	Preview image used on Hugging Face Space
`README.md`	This file

🧠 Model Details

Base Model: t5-small
Fine-Tuning: Trained on a 1000-sample custom dataset (review → summary)
Epochs: 5
Batch size: 4
Learning rate: 5e-5
Training Env: Google Colab + Kaggle

📄 License

This project is licensed under the MIT License.

🙋‍♂️ Author

Manish Kumar Kondoju
🌐 Hugging Face Profile
📫 [email protected]

⭐ Star this repo if you found it useful — and feel free to fork, extend, or share!

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference