what is the purpose of each line of this code?
3 ビュー (過去 30 日間)
古いコメントを表示
This is a code for training Neural Network:
function [net]=createff(Iin,Target)
net = newff(Iin, Target, 10, {'logsig' 'logsig'}, 'trainscg');
net.trainParam.perf = 'sse';
net.trainParam.epochs = 500;
net.trainParam.goal = 1e-5;
net.trainParam.lr=0.15;
net.trainParam.mc=0.8;
net = init(net);
net = train(net, Iin, Target);
end
Please help me to understand this code.
1 件のコメント
Greg Heath
2013 年 7 月 1 日
1. delete the net = init statement because newff is self-initializing.
2. Before using a code specifying various parameter values that overwrite default values, it is probably worth your while to try using as many defaults as possible. See the examples in
help newff
doc newff
採用された回答
Matthew Eicholtz
2013 年 6 月 29 日
The first line creates a feed-forward backpropagation network. Type this in the command window to learn more:
help newff
- '10' is the size of the hidden layer
- 'logsig' is log sigmoid transfer function
- 'trainscg' is scaled conjugate gradient backpropagation
- 'sse' is sum of squared errors
- 'epochs' refers to how many times to run through the training and validation sets
- 'goal' is the stopping criterion
- 'lr' is the learning rate
- 'mc' is the momentum constant
The last two lines initialize and train the network.
help init
help train
0 件のコメント
その他の回答 (0 件)
参考
カテゴリ
Help Center および File Exchange で Sequence and Numeric Feature Data Workflows についてさらに検索
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!