https://www.youtube.com/watch?v=i-iC6KQ0HJg
Below are the detailed steps to run the application based on the app.py file:
```
# Install Python:
# Ensure Python 3.7 or later is installed. Check the version using:
python --version
# Create a virtual environment (optional but recommended):
python -m venv venv
source venv/bin/activate # For Windows: venv\Scripts\activate
# Install dependencies:
# Create a `requirements.txt` file with the following content:
# fastapi
# uvicorn
# gradio
# pandas
# numpy
# matplotlib
# plotly
# joblib
# Then, run:
pip install -r requirements.txt
```
```
# Verify project files:
# Ensure the project directory contains the following files and folders:
# - app.py: Main application file.
# - archive/2022/heart_2022_no_nans.csv: Dataset file.
# - static/heart_attack_model.pkl: Pretrained random forest model file.
# - static/app.css: Custom CSS file.
# - static/app.js: Custom JavaScript file.
# - prompt.md: Prompt file for generating health summaries using LLMs.
# Validate file paths:
# Ensure the above file paths are correct within the project directory.
# For example, heart_attack_model.pkl should be located in the static folder.
```
```
# Start the application:
# From the project root directory, run:
python app.py
# This starts the Uvicorn server, listening on 0.0.0.0:4242. Output will look like:
# INFO: Uvicorn running on http://0.0.0.0:4242 (Press CTRL+C to quit)
# Keep the terminal window open to continue using the app.
# Access the application:
# Open a browser and navigate to http://localhost:4242/ to view the Gradio-based heart disease prediction and analysis interface.
```
```
# Refresh data:
# Click the "Refresh Data" button to load the latest data and generate relevant charts.
# View data distribution:
# Go to the "💻 Distribution" tab to see charts for state distribution, gender distribution, health status distribution, and heart attack distribution.
# View feature importance:
# Go to the "📊 Feature Importances" tab to view the ranking of feature importance in the model.
# Track data:
# Go to the "🔎 Tracking Data" tab to paginate through the dataset records.
# Use the "◀️" and "▶️" buttons to navigate pages,
# or enter a specific page number and click "🔄 Display" to refresh the table.
# View risk distribution:
# Go to the "🏃♂️ Risk Distribution" tab and click the "🔄 Refresh" button to update the risk distribution chart.
# Perform risk assessment:
# Go to the "🤖 Risk Factoring" tab to:
# - Load data: Click "Load Random Data", "Load Healthy Data", or "Load Unhealthy Data" to auto-fill random healthy or unhealthy data.
# - Input health factors: Manually enter health-related factors such as state, gender, health status, smoking status, etc.
# - Calculate risk: After input, click the "Calculate Risk" button to view the risk probability and generate a health summary.
```
```
# Public access:
# The host parameter in app.py is set to "0.0.0.0", allowing public access.
# - Local access: http://localhost:4242/
# - Public access: http://<your-ip>:4242/
```
```
# Static file management:
# Ensure the static files (CSS, JavaScript, and model files) in the static folder are correct and their paths align with app.py.
# Data updates:
# To update the dataset, replace the heart_2022_no_nans.csv file,
# then click the "Refresh Data" button to reload the data.
# Model updates:
# To update the model, replace the heart_attack_model.pkl file (a random forest model).
# Ensure the new model is compatible with the data preprocessing steps.
```