Use the sidebar to the left to navigate or use Search to find the information you need
about Minitab Model
Ops. If the Help navigation is not visible, select the
Help menu button 
to open the sidebar.
Model Ops Basics
Minitab® Statistical Software creates sophisticated predictive analytics models that you use for applications like prediction and forecasting. To predict and monitor with Minitab Model Ops, the first step is to import and deploy your model, including the baseline data. Then, you can upload prediction and stability data through standard HTTP requests. Next, you can monitor the drift and stability of the deployment using graphs and statistics to evaluate the performance of the champion and challenger models.
- Model Ops Environment
- Includes information about the Model Ops user interface, supported models, API keys, and the system audit log.
- Import and Deploy Models
- Create a new deployment from the Deployments page or deploy an existing model from the Model Repository. Learn how to add and delete models, change deployment settings and more.
- Data Requirements and Integration
- Each model requires a connection to the API endpoint. Learn how to integrate between platforms to easily retrieve prediction and stability data.
- Monitor Deployed Models
- Learn how to create drift and stability reports and assess the performance of the champion and challenger models.
Model Ops Tutorials
- Deploy a Minitab model
- Deploy and monitor a CART® Classification model that was created in Minitab® Statistical Software.
- Add and monitor challenger models
- Add challenger models to the deployment that was created in the Deploy and monitor a Minitab model example.
- Compare a Python model with a Minitab model
- Compare a MLPRegressor model, created in sci-kit to a Random Forests® Regression model that was created in Minitab Statistical Software.
Note
You can also download a pdf of the Getting Started with Minitab Model Ops® guide from Documentation.
