You can create a standard network that usestraingdwithfeedforwardnetorcascadeforwardnet. To prepare a custom network to be trained withtraingd,

In either case, callingtrainwith the resulting network trains the network withtraingd.

Seehelp feedforwardnetandhelp cascadeforwardnetfor examples.

Thebatch steepest descent training function istraingd. The weights and biases are updated in the direction of the negative gradient of the performance function. If you want to train a network using batch steepest descent, you should set the networktrainFcntotraingd, and then call the functiontrain. There is only one training function associated with a given network.

There are seven training parameters associated withtraingd:

The learning ratelris multiplied times the negative of the gradient to determine the changes to the weights and biases. The larger the learning rate, the bigger the step. If the learning rate is made too large, the algorithm becomes unstable. If the learning rate is set too small, the algorithm takes a long time to converge. See page 12-8 of [HDB96] for a discussion of the choice of learning rate.

The training status is displayed for everyshowiterations of the algorithm. (Ifshowis set toNaN,然后训练状态不显示。)The other parameters determine when the training stops. The training stops if the number of iterations exceedsepochs, if the performance function drops belowgoal, if the magnitude of the gradient is less thanmingrad, or if the training time is longer thantimeseconds.max_fail, which is associated with the early stopping technique, is discussed inImproving Generalization.

The following code creates a training set of inputspand targetst. For batch training, all the input vectors are placed in one matrix.

p = [-1 -1 2 2; 0 5 0 5]; t = [-1 -1 1 1];

Create the feedforward network.

net = feedforwardnet(3,'traingd');

In this simple example, turn off a feature that is introduced later.

net.divideFcn = '';

At this point, you might want to modify some of the default training parameters.

net.trainParam.show = 50; net.trainParam.lr = 0.05; net.trainParam.epochs = 300; net.trainParam.goal = 1e-5;

If you want to use the default training parameters, the preceding commands are not necessary.

Now you are ready to train the network.

[net,tr] = train(net,p,t);

The training recordtrcontains information about the progress of training.

Now you can simulate the trained network to obtain its response to the inputs in the training set.

a = net(p) a = -1.0026 -0.9962 1.0010 0.9960

Try theNeural Network Designdemonstrationnnd12sd1[HDB96] for an illustration of the performance of the batch gradient descent algorithm.