data and nn input sizes do not match..

my neural network is supposed to take an coloumn vector input [7x1] and output a value 1-3 that refers to a certain condition.

the way i have these set up is below, as well as a copy of the network code.

when i try to run this, i get an error "the numbers of input signals and networks inputs do not match"

this happens when i try to use net=configure(net, input), using my dataset.

im new to matlab so let me know if the mistake is just somewhere simple, ill have trouble finding it...

data= {temp;humid;SLP;precip;SWR;wind;dir};
%make a variable that treats all our coloumns as a matrix. i put this here
%and used it below to try and make the code work but it is making a 7 x 1
%cell instead of a 7x38000 matrix of all my input sets. 

predictionNet=network;
predictionNet.name='PredictNet';
predictionNet.numInputs= 7;
predictionNet.numLayers= 5;
%first three lines are just initialization of the predictionNetwork.
predictionNet.biasConnect=[1;1;1;1;1];
%select which layers we want biases for
predictionNet.inputConnect=[1 1 1 1 1 1 0
                            0 0 0 0 0 0 1
                            0 0 0 0 0 0 0 
                            0 0 0 0 0 0 0
                            0 0 0 0 0 0 0];
%boolean matrix of T/F connections between inputs and weights
predictionNet.outputConnect=[0 0 0 0 1];
%select which layers will connect to the final output
predictionNet.layerConnect=[0 0 0 0 0
                            1 0 0 0 0
                            0 1 0 0 0
                            0 0 1 0 0
                            0 0 0 1 0];
%outlines which layer you want to feed into a current layer out of all
%layers
predictionNet.layers{1}.size=10;
predictionNet.layers{2}.size=20;
predictionNet.layers{3}.size=20;            
predictionNet.layers{4}.size=15;              
predictionNet.layers{5}.size=1;
%set number of neurons in each layer. the last layer has to be size 1 to
%get an indication of warning/fire/safe
predictionNet.layers{1}.transferFcn='tansig';
predictionNet.layers{1}.initFcn='initnw';
predictionNet.layers{2}.transferFcn = 'logsig';
predictionNet.layers{2}.initFcn = 'initnw';
predictionNet.layers{3}.transferFcn = 'logsig';
predictionNet.layers{3}.initFcn = 'initnw';
predictionNet.layers{4}.transferFcn = 'logsig';
predictionNet.layers{4}.initFcn = 'initnw';
predictionNet.layers{5}.transferFcn = 'logsig';
predictionNet.layers{5}.initFcn = 'initnw';
%set the transfer function and the initial functions. 
%i chose tansig for the first so everything is centered around a mean, and
%then I chose logsig so everything is more probabalistic.
predictionNet.initFcn= 'initlay';
predictionNet.trainFcn='trainlm';
predictionNet.divideFcn='dividerand';
%divides training data randomly
predictionNet.plotFcns={'plotperform','plottrainstate'};
%choosing which plots i want when the data is read.
predictionNet= train(predictionNet, data, output);

view(predictionNet);