For the latest version of JMP Help, visit

Publication date: 11/10/2021

Neural Networks

Fit Nonlinear Models Using Nodes and Layers

Image shown here Most features in this platform are available only in JMP Pro and noted with this icon.

The Neural platform implements a fully connected multi-layer perceptron with one or two layers. Use neural networks to predict one or more response variables using a flexible function of the input variables. Neural networks can be very good predictors when it is not necessary to describe the functional form of the response surface, or to describe the relationship between the inputs and the response.

Figure 3.1 Example of a Neural Network 

Example of a Neural Network


Overview of the Neural Platform

Launch the Neural Platform

The Neural Launch Window
The Model Launch Control Panel
Validation Methods for Neural
Hidden Layer Structure
Fitting Options

Model Reports

Training and Validation Measures of Fit
Confusion Statistics

Neural Platform Options

Neural Model Options

Example of a Neural Network

Want more information? Have questions? Get answers in the JMP User Community (