preparets

(To be removed) Prepare input and target time series data for network simulation or training

preparets 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

[Xs,Xi,Ai,Ts,EWs,shift] = preparets(net,Xnf,Tnf,Tf,EW)

Description

[Xs,Xi,Ai,Ts,EWs,shift] = preparets(net,Xnf,Tnf,Tf,EW) takes these arguments:

• net — Neural network

• Xnf — Non-feedback inputs

• Tnf — Non-feedback targets

• Tf — Feedback targets

• EW — Error weights (optional)

and returns these arguments:

• Xs — Shifted inputs

• Xi — Initial input delay states

• Ai — Initial layer delay states

• Ts — Shifted targets

• EWs — Shifted error weights

• shift — The number of timesteps truncated from the front of X and T in order to properly fill Xi and Ai.

This function simplifies the normally complex and error prone task of reformatting input and target time series. It automatically shifts input and target time series as many steps as are needed to fill the initial input and layer delay states. If the network has open-loop feedback, then it copies feedback targets into the inputs as needed to define the open-loop inputs.

Each time a new network is designed, with different numbers of delays or feedback settings, preparets can reformat input and target data accordingly. Also, each time a network is transformed with openloop, closeloop, removedelay or adddelay, this function can reformat the data accordingly.

Examples

Prepare Data for Open- and Closed-Loop Networks

This example shows how to prepare data for open-loop and closed-loop networks.

Create a time-delay network with 20 hidden neurons, then train and simulate it.

[X,T] = simpleseries_dataset;
net = timedelaynet(1:2,20);
[Xs,Xi,Ai,Ts] = preparets(net,X,T);
net = train(net,Xs,Ts);
view(net)

Y = net(Xs,Xi,Ai);

Design a NARX network. The NARX network has a standard input and an open-loop feedback output to an associated feedback input.

[X,T] = simplenarx_dataset;
net = narxnet(1:2,1:2,20);
[Xs,Xi,Ai,Ts] = preparets(net,X,{},T);
net = train(net,Xs,Ts,Xi,Ai);

view(net)

y = net(Xs,Xi,Ai);

Now convert the network to closed loop, and reformat the data to simulate the network’s closed-loop response.

net = closeloop(net);
view(net)

[Xs,Xi,Ai] = preparets(net,X,{},T);
y = net(Xs,Xi,Ai);

Input Arguments

net — Input network

network

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

Xnf — Non-feedback inputs

cell array

Non-feedback input data (inputs not associated with an open loop feedback output), specified as cell array.

Tnf — Non-feedback targets

cell array

Target data for non-feedback outputs, specified as a cell array.

Tf — Feedback targets

cell array

Target data for outputs with feedback, specified as a cell array.

EW — Error weights

cell array

Error weights, specified as a cell array.

Output Arguments

Xs — Shifted inputs

cell array

Shifted inputs, returned as a cell array.

Xi — Initial input delay states

cell array

Initial input delay states, returned as cell array.

Ai — Initial layer delay states

cell array

Initial layer delay states, returned as a cell array.

Ts — Shifted targets

cell array

Shifted targets, returned as a cell array.

EWs — Shifted error weights

cell array

Shifted error weights, returned as a cell array.

shift — Timesteps

scalar

Number of timesteps truncated from the front of X and T in order to properly fill Xi and Ai, returned as a scalar.

Version History

Introduced in R2010b

R2026a: preparets will be removed

preparets will be removed in a future release.

To train a neural network with time series data, use one of these options:

• Time Series Modeler app — Interactively design and train neural networks for time series data.

• trainnet — Train neural network. Use this function when you need more flexibility in the network architecture or training algorithm.

To learn more about time series data workflows, see these examples and topics:

• Training Neural Networks with Time Series Data

• Time Series Forecasting Using Deep Learning

• Create and Train NARX Network for Time Series Forecasting

• Long Short-Term Memory Neural Networks

• Example Deep Learning Network Architectures

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