Welcome to the Model Integration Software ToolKit (MISTK)

This is the documentation for the Model Integration Software ToolKit (MISTK) SDK. MISTK is an ecosystem for integrating, validating, and evaluating machine learning models and algorithms. This guide serves to instruct a developer on adapting a machine learning model for use within the ecosystem by implementing the model interface, running the implemented model as a web service, testing and validating the model implementation, and packaging the model as a Docker image for distribution.

Contents:

• Getting Started
  • Prerequisites
  • MISTK Python Library
  • Installation
  • Assumptions
• Guide
  • Model Interface
  • Data Transformation Interface
  • Evaluation Interface
• Testing
  • Running the Model
  • Running in Eclipse and PyDev
  • Testing through the RESTful API
  • Test and Validation Harness
  • Validation Metrics
• Deployment
  • Containerization
• MISTK API
  • mistk package

Indices and Tables

• Index

• Module Index

• Search Page