Neural Network Same Constant Result Value Problem

Adam

2015 3 30

0 个回答

1 次查看（30 天）

0 个投票

I am trying to predict using 2 years data with NNet for electricity prices. However when I run the below code my forecast Prices for all year is same result. I did not figure it out where the problem is. Do you have any idea. (For example I get 213.68 for whole hours in a year)

Thank you very much.

 % code
% % 1
load('data')
% % 2
[num, text] = xlsread('Holidays.xls');
holidays = text(2:end,1);
% % 3
term='short';
[X, dates, labels] = genPredictors(data, term, holidays);
% % 4
ind = data.ElecPrice==0;
data.ElecPrice(ind) = interp1(find(~ind), data.ElecPrice(~ind), find(ind));
trainInd = data.NumDate < datenum('2014-01-01');
trainX = X(trainInd,:);
trainY = data.ElecPrice(trainInd);
testInd = data.NumDate >= datenum('2014-01-01');
testX = X(testInd,:);
testY = data.ElecPrice(testInd);
testDates = dates(testInd);
save testSet testDates testX testY
clear X data trainInd testInd term holidays dates ans num text
% % 5
reTrain = false;
if reTrain || ~exist('Models\NNModel.mat', 'file')
    net = newfit(trainX', trainY', 20);
    net.performFcn = 'mae';
    net = train(net, trainX', trainY');
    save Models\NNModel.mat net
else
    load Models\NNModel.mat
end
% % 6
load Data\testSet
forecastPrice = sim(net, testX')';
  end