DishaMondal2024/pcos_predictor

PCOS Prediction Web App

This repository hosts a web-based application for predicting Polycystic Ovary Syndrome (PCOS) using clinical and lifestyle data. The app is powered by Streamlit and deployed on Hugging Face Spaces.

🚀 Live Demo

Access the live app here: PCOS Predictor on Hugging Face

📖 About the Application

The PCOS Predictor allows users to input various clinical and lifestyle factors to determine the likelihood of PCOS. It utilizes a Logistic Regression model trained on a comprehensive dataset from Kaggle.

Features:

• —Interactive and user-friendly interface
• —Prediction based on clinically relevant parameters
• —Clear results indicating the likelihood of PCOS

🔢 Model Information

• —Algorithm: Logistic Regression
• —Pipeline:
• —Categorical data preprocessing: OneHotEncoder
• —Numerical data preprocessing: StandardScaler
• —Classifier: Logistic Regression with optimized hyperparameters
• —Hyperparameters:
• —Regularization (C): 0.1
• —Penalty: l2
• —Solver: liblinear

🧾 Dataset Information

The model was trained on the Polycystic Ovary Syndrome (PCOS) dataset available on Kaggle:

• —Author: Prasoon Kottarathil
• —Title: Polycystic ovary syndrome (PCOS)
• —Year: 2020
• —Publisher: Kaggle
• —Dataset Link: PCOS Dataset

📂 Repository Structure

• —app.py: Streamlit application script
• —model.pkl: Pre-trained Logistic Regression model
• —requirements.txt: Dependencies for the application
• —README.md: Documentation file (this file)

🛠️ How to Run Locally

1. 1.Clone the repository:

   git clone https://huggingface.co/spaces/DishaMondal2024/pcos_predictor
   cd pcos_predictor

1. 1.Install dependencies:

   pip install -r requirements.txt

1. 1.Run the application:

   streamlit run app.py

1. 1.Open the link provided by Streamlit (e.g., http://localhost:8501) to interact with the app.

📜 License

This project is licensed under the MIT License.

👏 Acknowledgements

Special thanks to Prasoon Kottarathil for providing the dataset that made this project possible.