Number of observations in X and Y disagree.

4 Mar 2022
2 Answers
Answer Accepted
Updated 4 Mar 2022
4 Views (30 days)
Ran in:

0 votes

Ran in:
clc; clear all; close all;
%Import/Upload data
load Projectdata.mat
% change to label vector
CS = categories(categorical(GR_output));
CS1 = categories(categorical(INS_output));
Z1 = []; Z2 = [];
for i = 1 : length(GR_output)
Z1(i,1) = find(GR_output(i)==CS);
end
for i = 1 : length(INS_output)
Z2(i,1) = find(INS_output(i)==CS1);
end
Yo1 = GR_output;
Yo2 = INS_output;
GR_output= Z1;
INS_output = Z2;
%transposing glucose data
GlucoseReadings_T = GlucoseReadings';
%transposing insulin data
InsulinReadings_T = InsulinReadings';
%Shuffling data to take randomly
rand('seed', 0)
ind = randperm(size(GlucoseReadings_T, 1));
GlucoseReadings_T = GlucoseReadings_T(ind, :);
GR_output = GR_output(ind);
ind = randperm(size(InsulinReadings_T, 1));
InsulinReadings_T = InsulinReadings_T(ind, :);
INS_output = INS_output(ind);
%Separating data in training, validation and testing data
GlucoseReadings_train = GlucoseReadings_T;
InsulinReadings_train = InsulinReadings_T;
%Partioning data for training 70%
train_GlucoseReadings = GlucoseReadings_train(1:84,:);
train_InsulinReadings = InsulinReadings_train(1:84,:);
%Corresponding X(input) data to Y(output) data
train_GR_output = GR_output(1:17);
train_INS_output = INS_output(1:84);
%reshaping data into 4D array
GlucoseReadingsTrain=(reshape(train_GlucoseReadings', [1749,1,1,84]));
InsulinReadingsTrain=(reshape(train_InsulinReadings',[1758,1,1,84]));
%Separating and partioning for validation data 15%
val_GlucoseReadings = GlucoseReadings_train(85:102,:);
val_InsulinReadings = InsulinReadings_train(85:102,:);
%Corresponding X(input) data to Y(output) data
val_GR_output = GR_output(85:102);
val_INS_output = INS_output(85:102);
%reshaping data into 4D array
whos
Name Size Bytes Class Attributes CS 2x1 248 cell CS1 2x1 248 cell GR_output 120x1 960 double GlucoseReadings 1749x120 1679040 double GlucoseReadingsTrain 1749x1x1x84 1175328 double GlucoseReadings_T 120x1749 1679040 double GlucoseReadings_train 120x1749 1679040 double INS_output 120x1 960 double InsulinReadings 1758x120 1687680 double InsulinReadingsTrain 1758x1x1x84 1181376 double InsulinReadings_T 120x1758 1687680 double InsulinReadings_train 120x1758 1687680 double Yo1 1x120 7296 string Yo2 1x120 7296 string Z1 120x1 960 double Z2 120x1 960 double i 1x1 8 double ind 1x120 960 double train_GR_output 17x1 136 double train_GlucoseReadings 84x1749 1175328 double train_INS_output 84x1 672 double train_InsulinReadings 84x1758 1181376 double val_GR_output 18x1 144 double val_GlucoseReadings 18x1749 251856 double val_INS_output 18x1 144 double val_InsulinReadings 18x1758 253152 double
GlucoseReadingsVal=(reshape(val_GlucoseReadings', [1749,1,1,18]));
InsulinReadingsVal=(reshape(val_InsulinReadings', [1758,1,1,18]));
%Separating and partioning for test data 15%
test_GlucoseReadings = GlucoseReadings_train(103:120,:);
test_InsulinReadings = InsulinReadings_train(103:120,:);
%Corresponding X(input) data to Y(output) data
test_GR_output = GR_output(103:120);
test_INS_output = INS_output(103:120);
%reshaping data into 4D array
GlucoseReadingsTest=(reshape(test_GlucoseReadings', [1749,1,1,18]));
InsulinReadingsTest=(reshape(test_InsulinReadings', [1758,1,1,18]));
%% NETWORK ARCHITECTURE
layers = [imageInputLayer([1758 1 1]) % Creating the image layer
convolution2dLayer([102 1],3,'Stride',1)
batchNormalizationLayer
reluLayer
maxPooling2dLayer(2,'Stride',2,'Padding',[0 0 0 1])
dropoutLayer
fullyConnectedLayer(1)
regressionLayer];
% Specify training options.
opts = trainingOptions('sgdm', ...
'MaxEpochs',1000, ...
'Shuffle','every-epoch', ...
'Plots','training-progress', ...
'Verbose',false, ...
'ValidationData',{GlucoseReadingsVal,val_GR_output,},...
'ValidationData',{InsulinReadingsVal,val_INS_output,},...
'LearnRateDropFactor',0.2,...
'LearnRateDropPeriod',5,...
'ExecutionEnvironment', 'cpu', ...
'ValidationPatience',Inf);
%% Train network
%net = trainNetwork(XTrain,Trainoutfinal,layers,opts);
yc = train_GR_output(:);
yc1 = train_INS_output(:);
net1 = trainNetwork(GlucoseReadingsTrain,yc,layers,opts);
Error using trainNetwork (line 184)
Number of observations in X and Y disagree.
net2 = trainNetwork(InsulinReadingsTrain,yc1,layers,opts);
%% Compare against testing Data
GR_outputpredicted = predict(net1, GlucoseReadingsTest)
INS_outputpredicted = predict(net1, InsulinReadingsTest)
predictionError = test_GR_output - GR_outputpredicted;
predictionError1 = test_INS_output - INS_outputpredicted;
squares = predictionError.^2;
rmse = sqrt(mean(squares))
figure
scatter(GR_outputpredicted, test_GR_output,'+')
title ('True value vs Predicted Value')
xlabel ("Predicted Value")
ylabel ("True Value")
hold on
plot([-3 3], [-7 7], 'b--')
scatter(INS_outputpredicted, test_INS_output,'+')
title ('True value vs Predicted Value')
xlabel ("Predicted Value")
ylabel ("True Value")
hold on
plot([-3 3], [-7 7], 'b--')

Sign in to answer this question.

 Accepted Answer

KSSV
KSSV on 4 Mar 2022
Edited: KSSV on 4 Mar 2022

0 votes

Your yc i.e. target for the respective input in the line:
net1 = trainNetwork(GlucoseReadingsTrain,yc,layers,opts);
is of dimension 17x1. Where as input is 1749x1x1x84. So there is a miss match and you got error. Where as the variables yc1 is of dimension 84x1 and this should work. This line should not through any error:
net2 = trainNetwork(InsulinReadingsTrain,yc1,layers,opts);
So you have to use net2. You may delete the line net1.

5 Comments
Show 3 older comments Hide 3 older comments

Nathaniel Porter
Nathaniel Porter on 4 Mar 2022
Okay i understand to change the dimensions but I am trainig two sets of data once i fix the error will it run both?
KSSV
KSSV on 4 Mar 2022
To run you need to have compatible dimensions betwwen input and output.
Nathaniel Porter
Nathaniel Porter on 4 Mar 2022
Edited: Walter Roberson on 4 Mar 2022
Ran in:
Ran in:
The validation images are of size 1758×1×1 but the input layer expects images of size 1749×1×1.
clc; clear all; close all;
%Import/Upload data
load Projectdata.mat
% change to label vector
CS = categories(categorical(GR_output));
CS1 = categories(categorical(INS_output));
Z1 = []; Z2 = [];
for i = 1 : length(GR_output)
Z1(i,1) = find(GR_output(i)==CS);
end
for i = 1 : length(INS_output)
Z2(i,1) = find(INS_output(i)==CS1);
end
Yo1 = GR_output;
Yo2 = INS_output;
GR_output= Z1;
INS_output = Z2;
%transposing glucose data
GlucoseReadings_T = GlucoseReadings';
%transposing insulin data
InsulinReadings_T = InsulinReadings';
%Shuffling data to take randomly
rand('seed', 0)
ind = randperm(size(GlucoseReadings_T, 1));
GlucoseReadings_T = GlucoseReadings_T(ind, :);
GR_output = GR_output(ind);
ind = randperm(size(InsulinReadings_T, 1));
InsulinReadings_T = InsulinReadings_T(ind, :);
INS_output = INS_output(ind);
%Separating data in training, validation and testing data
GlucoseReadings_train = GlucoseReadings_T;
InsulinReadings_train = InsulinReadings_T;
%Partioning data for training 70%
train_GlucoseReadings = GlucoseReadings_train(1:84,:);
train_InsulinReadings = InsulinReadings_train(1:84,:);
%Corresponding X(input) data to Y(output) data
train_GR_output = GR_output(1:84);
train_INS_output = INS_output(1:84);
%reshaping data into 4D array
GlucoseReadingsTrain=(reshape(train_GlucoseReadings', [1749,1,1,84]));
InsulinReadingsTrain=(reshape(train_InsulinReadings',[1758,1,1,84]));
%Separating and partioning for validation data 15%
val_GlucoseReadings = GlucoseReadings_train(85:102,:);
val_InsulinReadings = InsulinReadings_train(85:102,:);
%Corresponding X(input) data to Y(output) data
val_GR_output = GR_output(85:102);
val_INS_output = INS_output(85:102);
%reshaping data into 4D array
GlucoseReadingsVal=(reshape(val_GlucoseReadings', [1749,1,1,18]));
InsulinReadingsVal=(reshape(val_InsulinReadings', [1758,1,1,18]));
%Separating and partioning for test data 15%
test_GlucoseReadings = GlucoseReadings_train(103:120,:);
test_InsulinReadings = InsulinReadings_train(103:120,:);
%Corresponding X(input) data to Y(output) data
test_GR_output = GR_output(103:120);
test_INS_output = INS_output(103:120);
%reshaping data into 4D array
GlucoseReadingsTest=(reshape(test_GlucoseReadings', [1749,1,1,18]));
InsulinReadingsTest=(reshape(test_InsulinReadings', [1758,1,1,18]));
%% NETWORK ARCHITECTURE
layers = [imageInputLayer([1749 1 1]) % Creating the image layer
convolution2dLayer([102 1],3,'Stride',1)
batchNormalizationLayer
reluLayer
maxPooling2dLayer(2,'Stride',2,'Padding',[0 0 0 1])
dropoutLayer
fullyConnectedLayer(1)
regressionLayer];
layers1 = [imageInputLayer([1758 1 1]) % Creating the image layer
convolution2dLayer([102 1],3,'Stride',1)
batchNormalizationLayer
reluLayer
maxPooling2dLayer(2,'Stride',2,'Padding',[0 0 0 1])
dropoutLayer
fullyConnectedLayer(1)
regressionLayer];
% Specify training options.
opts = trainingOptions('sgdm', ...
'MaxEpochs',1000, ...
'Shuffle','every-epoch', ...
'Plots','training-progress', ...
'Verbose',false, ...
'ValidationData',{GlucoseReadingsVal,val_GR_output,},...
'ValidationData',{InsulinReadingsVal,val_INS_output,},...
'LearnRateDropFactor',0.2,...
'LearnRateDropPeriod',5,...
'ExecutionEnvironment', 'cpu', ...
'ValidationPatience',Inf);
%% Train network
%net = trainNetwork(XTrain,Trainoutfinal,layers,opts);
yc = train_GR_output(:);
yc1 = train_INS_output(:);
whos
Name Size Bytes Class Attributes CS 2x1 248 cell CS1 2x1 248 cell GR_output 120x1 960 double GlucoseReadings 1749x120 1679040 double GlucoseReadingsTest 1749x1x1x18 251856 double GlucoseReadingsTrain 1749x1x1x84 1175328 double GlucoseReadingsVal 1749x1x1x18 251856 double GlucoseReadings_T 120x1749 1679040 double GlucoseReadings_train 120x1749 1679040 double INS_output 120x1 960 double InsulinReadings 1758x120 1687680 double InsulinReadingsTest 1758x1x1x18 253152 double InsulinReadingsTrain 1758x1x1x84 1181376 double InsulinReadingsVal 1758x1x1x18 253152 double InsulinReadings_T 120x1758 1687680 double InsulinReadings_train 120x1758 1687680 double Yo1 1x120 7296 string Yo2 1x120 7296 string Z1 120x1 960 double Z2 120x1 960 double i 1x1 8 double ind 1x120 960 double layers 8x1 4018 nnet.cnn.layer.Layer layers1 8x1 4018 nnet.cnn.layer.Layer opts 1x1 253757 nnet.cnn.TrainingOptionsSGDM test_GR_output 18x1 144 double test_GlucoseReadings 18x1749 251856 double test_INS_output 18x1 144 double test_InsulinReadings 18x1758 253152 double train_GR_output 84x1 672 double train_GlucoseReadings 84x1749 1175328 double train_INS_output 84x1 672 double train_InsulinReadings 84x1758 1181376 double val_GR_output 18x1 144 double val_GlucoseReadings 18x1749 251856 double val_INS_output 18x1 144 double val_InsulinReadings 18x1758 253152 double yc 84x1 672 double yc1 84x1 672 double
net1 = trainNetwork(GlucoseReadingsTrain,yc,layers,opts);
Error using trainNetwork (line 184)
The validation images are of size 1758×1×1 but the input layer expects images of size 1749×1×1.
net2 = trainNetwork(InsulinReadingsTrain,yc1,layers1,opts);
%% Compare against testing Data
GR_outputpredicted = predict(net1, GlucoseReadingsTest)
INS_outputpredicted = predict(net2, InsulinReadingsTest)
predictionError = test_GR_output - GR_outputpredicted;
predictionError1 = test_INS_output - INS_outputpredicted;
squares = predictionError.^2;
rmse = sqrt(mean(squares))
figure
scatter(GR_outputpredicted, test_GR_output,'+')
title ('True value vs Predicted Value')
xlabel ("Predicted Value")
ylabel ("True Value")
hold on
plot([-3 3], [-7 7], 'b--')
scatter(INS_outputpredicted, test_INS_output,'+')
title ('True value vs Predicted Value')
xlabel ("Predicted Value")
ylabel ("True Value")
hold on
plot([-3 3], [-7 7], 'b--')
KSSV
KSSV on 4 Mar 2022
Now you messed up with the input: GlucoseReadingsTrain
It should be 1758x1x1x84, but it is 1749x1x1x84.
Nathaniel Porter
Nathaniel Porter on 4 Mar 2022
GlucoseReadingsTrain=(reshape(train_GlucoseReadings', [1758,1,1,84]));
Changed it but got this:
Error using reshape
Number of elements must not change. Use [] as one of the size inputs to automatically calculate the
appropriate size for that dimension.

Sign in to comment.

More Answers (1)

Walter Roberson
Walter Roberson on 4 Mar 2022
Edited: Walter Roberson on 4 Mar 2022

0 votes

net2 = trainNetwork(train_GlucoseReadings,yc1,layers1,opts);
Your code uses a series of variables that are not well distinguished from each other in meaning, so it is easy to get wrong which ones you are talking about.

1 Comment
Show -1 older comments Hide -1 older comments

Nathaniel Porter
Nathaniel Porter on 4 Mar 2022
No well this is suppose to be my net two which are my insulin readings
net2 = trainNetwork(InsulinReadingsTrain,yc1,layers1,opts);

Sign in to comment.

Categories

Find more on Visualization and Interpretability in Help Center and File Exchange

Products

Release

R2021b

Tags

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!