SupportVectorRegressor

🍷 Wine Quality Prediction using SVR

This project builds a Support Vector Regression (SVR) model to predict the quality of red wine based on selected physicochemical features. The model uses a radial basis function (RBF) kernel with feature scaling via StandardScaler.

---

📌 Project Overview

Wine quality is influenced by various physicochemical properties. This project aims to predict the wine quality (on a scale of 0–10) using machine learning with selected features from the dataset. The model is trained using:

• Support Vector Regression (SVR)
• StandardScaler (for feature scaling)
• A pipeline to simplify preprocessing and modeling

---

📊 Dataset

• Source: UCI Machine Learning Repository
• Filename: winequality-red.csv
• Total Entries: 1599
• Target Variable: quality
• Selected Features:
  • fixed acidity
  • volatile acidity
  • citric acid
  • alcohol
  • pH

---

🧠 ML Pipeline

model = make_pipeline(StandardScaler(), SVR(kernel='rbf'))

• Train-test split: 80-20
• Model persistence: Using pickle (svr_model.pkl)

---

🚀 Live Deployment

🔗 Live App: Wine Quality Predictor

---

🖼️ Screenshots

📌 Home Page

📌 Prediction Result

---

🗂️ Project Structure

.
├── winequality-red.csv
├── svr_model.pkl
├── train_model.py
├── app.py
├── templates/
│   ├── index.html
│   └── result.html
├── static/
│   └── style.css
├── screenshots/
│   ├── home.png
│   └── result.png
└── README.md

📚 Requirements

• pandas
• scikit-learn
• flask
• pickle (standard library)