Home Loan Approval

Welcome to the Home Loan Approval project! This machine learning-based web application predicts whether an individual is eligible for a home loan. It combines the power of Streamlit for the front end, FastAPI for the API, and various Python libraries for data processing and machine learning.

• Home_Loan_Approval Streamlit

• Home_Loan_Approval API

• Home_Loan_Approval Kaggle

• Home_Loan_Approval notion

Tech Stack

This project leverages a robust tech stack to deliver accurate and efficient loan predictions:

• Frontend: Streamlit
• API: FastAPI, Pydantic
• API hosting: Uvicorn (local), Render
• Machine Learning: Scikit-Learn
• Data Manipulation: Pandas, NumPy
• Data Visualization: Matplotlib, Seaborn
• Model Serialization: Pickle

Getting Started

Prerequisites

Before diving into the project, ensure you have Python installed. You can download the latest version from python.org.

Installation

Clone the repository to your local machine:

git clone https://github.com/kinba09/Home_Loan_Approval.git

Install the required dependencies:

pip install -r requirements.txt

Usage

Start the FastAPI server:

uvicorn mlapi:app --reload

Host Streamlit:

streamlit run app.py

To host it on render:

• Pushed only the API-required files to the repo first and hosted them on Render
• The start command in render is uvicorn mlapi:app --host 0.0.0.0 --port $PORT

API
  |---- CustomData.py
  |---- encoding.py
  |---- mlapi.py
  |---- requirements.txt

• Then pushed the other files and hosted them on Streamlit

Streamlit
  |---- app.py
  |---- encoding.py

Note

• The version of Scikit Learn on Render is only up to 1.0.2. Therefore, if your pickled model is from a later version of Scikit-Learn, you may need to use the older pickled model () instead of the file. If you are planning to use the latest of Scikit Learn then use