Confusionmatrix for linear regression

Hi,

I'm using the code below to train a logistic regression classifier. I'd like to plot a confusion matrix but even though i waited 30 minutes, it doesn't show the matrix. I just run the code and it goes on debugging but never shows the result. The predicted and response matrix consits of 5000x1 matrix that has either 0 or 1. I don't think plotting a confusion matrix for this data would take that long. Can anybody help me about the problem ?

strokedata=importstroke("healthcare-dataset-stroke-data")
inputTable=strokedata;
predictorNames={'gender','age','hypertension','heart_disease','work_type','Residence_type','avg_glucose_level','bmi','smoking_status'}
predictors = inputTable(:, predictorNames)
response=inputTable.stroke
isCategoricalPredictor=[true,false,false,false,true,true,true,false,false,true]

successClass = double(1);
failureClass = double(0);

numSuccess = sum(response == successClass);
numFailure = sum(response == failureClass);
if numSuccess > numFailure
    missingClass = successClass
else
    missingClass = failureClass
end
successFailureAndMissingClasses = [successClass; failureClass; missingClass];

isMissing = isnan(response)
zeroOneResponse = double(ismember(response, successClass))
zeroOneResponse(isMissing) = NaN
% Prepare input arguments to fitglm.
concatenatedPredictorsAndResponse = [predictors, table(zeroOneResponse)]

GeneralizedLinearModel = fitglm(...
    concatenatedPredictorsAndResponse, ...
    'Distribution', 'binomial','link','logit')

yPredicted=predict(GeneralizedLinearModel,inputTable) > 0.47

plotconfusion(response,yPredicted)

[the cyclist]

Can you upload the data, so that we can run your code?

[Gorkem Akgul]

Of course.

You can download the dataset from the following link.

Stroke Dataset

[the cyclist]

Is importstroke() a function you wrote? I don't see it on Kaggle.

[Gorkem Akgul]

@the cyclist Yes, i just used the import data option and save it as importstroke()

function healthcaredatasetstrokedata = importfile(filename, dataLines)
%IMPORTFILE Import data from a text file
%  HEALTHCAREDATASETSTROKEDATA = IMPORTFILE(FILENAME) reads data from
%  text file FILENAME for the default selection.  Returns the data as a
%  table.
%
%  HEALTHCAREDATASETSTROKEDATA = IMPORTFILE(FILE, DATALINES) reads data
%  for the specified row interval(s) of text file FILENAME. Specify
%  DATALINES as a positive scalar integer or a N-by-2 array of positive
%  scalar integers for dis-contiguous row intervals.
%
%  Example:
%  healthcaredatasetstrokedata = importfile("F:\Documents\Neural Networks Matlab\Course 1\COVID-19-master\healthcare-dataset-stroke-data.csv", [2, Inf]);
%
%  See also READTABLE.
%
% Auto-generated by MATLAB on 17-Apr-2021 22:47:18

%% Input handling

% If dataLines is not specified, define defaults
if nargin < 2
    dataLines = [2, Inf];
end

%% Setup the Import Options and import the data
opts = delimitedTextImportOptions("NumVariables", 12);

% Specify range and delimiter
opts.DataLines = dataLines;
opts.Delimiter = ",";

% Specify column names and types
opts.VariableNames = ["id", "gender", "age", "hypertension", "heart_disease", "ever_married", "work_type", "Residence_type", "avg_glucose_level", "bmi", "smoking_status", "stroke"];
opts.VariableTypes = ["double", "categorical", "double", "double", "double", "categorical", "categorical", "categorical", "double", "double", "categorical", "double"];

% Specify file level properties
opts.ExtraColumnsRule = "ignore";
opts.EmptyLineRule = "read";

% Specify variable properties
opts = setvaropts(opts, ["gender", "ever_married", "work_type", "Residence_type", "smoking_status"], "EmptyFieldRule", "auto");

% Import the data
healthcaredatasetstrokedata = readtable(filename, opts);

end

[the cyclist]

I don't have the Deep Learning Toolbox. But

confusionchart(logical(response),yPredicted) % requires Stats & Machine Learning Toolbox

returned the chart in under a second. So, I think your instinct is correct. I'm not sure what's going on in your code.

[Gorkem Akgul]

Thank you I solved the problem with the function you say. I think there's a problem with the plotconfusion() matrix. I asked one of my friends to try it and he said he had the same problem as well. Or perhaps, it is only suitable for deep learning models.