perform

(To be removed) Calculate network performance

perform will be removed in a future release. For more information, see Transition Legacy Neural Network Code to dlnetwork Workflows.

For advice on updating your code, see Version History.

Syntax

perf = perform(net,t,y,ew)

Description

perf = perform(net,t,y,ew) takes a network net, targets T, outputs Y, and optionally error weights EW, and returns network performance calculated according to the net.performFcn and net.performParam property values.

The target and output data must have the same dimensions. The error weights may be the same dimensions as the targets, in the most general case, but may also have any of its dimensions be 1. This gives the flexibility of defining error weights across any dimension desired.

Examples

Calculate Network Performance with 'perform' Function

This example shows how to calculate the performance of a feedforward network with the perform function.

Create a feedforward network using the data from the simple fit data set and calculate its performance.

[x,t] = simplefit_dataset;
net = feedforwardnet(20);
net = train(net,x,t);
y = net(x);
perf = perform(net,t,y)

perf =

   2.3654e-06

Input Arguments

net — Input network

network

Input network, specified as a network object. To create a network object, use for example, feedforwardnet or narxnet.

t — Network targets

matrix | cell array

Network targets, specified as a matrix or cell array.

y — Network outputs

matrix | cell array

Network outputs, specified as a matrix or cell array.

ew — Error weights

vector | matrix | cell array

Error weights, specified as a vector, matrix, or cell array.

Error weights can be defined by sample, output element, time step, or network output:

ew = [1.0 0.5 0.7 0.2]; % Across 4 samples
ew = [0.1; 0.5; 1.0]; % Across 3 elements
ew = {0.1 0.2 0.3 0.5 1.0}; % Across 5 timesteps
ew = {1.0; 0.5}; % Across 2 outputs

The error weights can also be defined across any combination, such as across two time-series (i.e., two samples) over four timesteps.

ew = {[0.5 0.4],[0.3 0.5],[1.0 1.0],[0.7 0.5]};

In the general case, error weights may have exactly the same dimensions as targets, in which case each target value will have an associated error weight.

The default error weight treats all errors the same.

ew = {1}

Output Arguments

perf — Network performance

scalar

Network performance, returned as a scalar.

Version History

Introduced in R2010b

perform will be removed

perform will be removed in a future release.

perform will be removed in a future release.

To learn more about using metrics and loss functions when you train neural networks, see these examples and topics:

• Deep Learning Metrics

• Custom Loss Functions

• Monitor Deep Learning Training Progress

• ROC Curve and Performance Metrics

To assess the predictive performance of Statistics and Machine Learning Toolbox™ neural network models, use one of these functions:

• loss (Statistics and Machine Learning Toolbox) — compute loss of regression neural network.

• loss (Statistics and Machine Learning Toolbox) — compute loss of classification neural network.

To assess the predictive performance of a dlnetwork object, use the testnet function.

For more information, see Transition Legacy Neural Network Code to dlnetwork Workflows.