Documentation Index
Fetch the complete documentation index at: https://mintlify.com/jonatan-leal/ia-proyecto-sustituto/llms.txt
Use this file to discover all available pages before exploring further.
Diabetes Prediction ML
A comprehensive machine learning system for predicting diabetes using clinical and lifestyle data
Overview
Diabetes Prediction ML is a complete machine learning solution that predicts diabetes risk based on patient health metrics. Built with RandomForest classification and deployed as a FastAPI REST API, this system demonstrates the full ML lifecycle from exploratory analysis to production deployment. The project is structured in three progressive phases, each building on the previous one:Phase 1: Notebook
Exploratory analysis and model development in Jupyter
Phase 2: CLI
Command-line tools for training and prediction
Phase 3: API
REST API deployment with FastAPI and Docker
Key Features
RandomForest Classifier
Robust ensemble learning method for accurate diabetes prediction
SMOTEENN Balancing
Handles imbalanced datasets with combined over/under-sampling
FastAPI REST API
High-performance async API with automatic OpenAPI documentation
Docker Deployment
Containerized application for consistent deployment environments
Feature Normalization
StandardScaler preprocessing for optimal model performance
Batch & Single Predictions
Support for both individual and batch prediction workflows
Quick Links
Get Started
Install and run your first prediction in minutes
API Reference
Explore the complete REST API documentation
Model Architecture
Understand how the prediction model works
Dataset Information
Learn about the training data from Kaggle
Use Cases
This diabetes prediction system is designed for:- Healthcare Providers: Screen patients for diabetes risk during routine checkups
- Research Teams: Analyze factors contributing to diabetes in populations
- ML Engineers: Reference implementation for healthcare ML workflows
- Students: Learn end-to-end ML system development and deployment
Model Input
The prediction model accepts eight patient features:- Gender: Patient’s biological sex (Female, Male, Other)
- Age: Patient’s age in years
- Hypertension: History of high blood pressure (0/1)
- Heart Disease: History of heart disease (0/1)
- Smoking History: Smoking status (never, current, former, etc.)
- BMI: Body Mass Index (kg/m²)
- HbA1c Level: Average blood glucose over 3 months (%)
- Blood Glucose Level: Current blood glucose measurement (mg/dL)
What’s Next?
Explore the Dataset
Learn about the diabetes prediction dataset from Kaggle and its features
Run the Quickstart
Get your first prediction running in under 5 minutes
Understand the Model
Deep dive into the model architecture and preprocessing pipeline
Deploy to Production
Follow the deployment guide to containerize and deploy the API