How can I import data from .csv file with numeric values and texts (with column headers) into MATLAB Workspace?

I have data files (.csv of course) which contains several columns each with a column header. I want to import the data as I do in R. Using 'csvimport' I'm able to capture only the numeric values not the strings in the data. Please someone tell me the procedure of importing such data.

Answer 1

[Gareth Thomas]

If you right click on the .csv file you can use the “Import Tool” which will help you through the import process. you will be able to choose what data types are imported.

If you click the “Import” button, you can also see the generated code or script. You will see the code it generates.

Another way is to use: “readtable”. See the “readtable” page examples on reading *.csv files.

https://www.mathworks.com/help/matlab/ref/readtable.html

I hope that this helps. Gareth

Answer 2

[adam honse]

[~,~,data] = xlsread('myfile.csv')

works just fine for csvs.

Answer 3

[Augustine Ekweariri]

You can use the csvread(csv_file.csv,row_offset,col_offset) command to read data from a csv file in this format;

a = csvread('csv_file.csv',1,0);

for more information, visit;

https://www.mathworks.com/help/matlab/ref/csvread.html

Answer 4

[Stu Schwartz]

Input to csvread must be all numeric.

Mixed csv data can be input with table read and a format specification. See readtable online help and example: https://www.mathworks.com/help/matlab/import_export/import-mixed-text-and-numeric-data-from-a-text-file.html

[Walter Roberson]

Historically it was true that input to csvread had to be all numeric. However, somewhere around R2015a, the range specification to csvread became able to skip over leading text rows and leading text columns (provided that the user asked to skip them); it is not able to skip over trailing text rows or trailing text columns however.

However, this ability to skip over leading rows does not help the user read the strings as they asked. It would, though, allow for the possibility of fopen() the file, fgetl() one line, fclose(), split the input line to get the column headers, then csvread() the file telling it to skip the first row.

Answer 5

[Saeed Nematshahi]

use

a = csvread('csv_file.csv',1,0);

this only remove first column.

by changing 0 to 1 you can remove first column and first zero.

[Walter Roberson]

Someone already recommended that three years ago; https://www.mathworks.com/matlabcentral/answers/122717-how-can-i-import-data-from-csv-file-with-numeric-values-and-texts-with-column-headers-into-matlab#answer_257459

Answer 6

[ehsan ghane]

for matlab R2020b, use readmatrix fnction.

[Walter Roberson]

readmatrix() is not able to read text and numeric in the same file. The user had said "Using 'csvimport' I'm able to capture only the numeric values not the strings in the data." which I think means they wanted the text as well.

[ehsan ghane]

Absolutely.

Answer 7

[Elamin]

function HandGesture_GUI_3ButtonsRow_ExpandedPred_MotorsSelective %% GUI Setup fig = figure('Name','Hand Gesture EMG GUI','Position',[100 100 950 550]);

% Hand plot axes (يمين) axHand = axes('Parent',fig,'Position',[0.45 0.55 0.5 0.35]); axis(axHand,[0 10 0 10]); axis(axHand,'equal'); axis(axHand,'off'); axis(axHand,'manual'); title(axHand,'Hand Gesture');

% EMG plot axes (يمين تحت اليد) axEMG = axes('Parent',fig,'Position',[0.45 0.15 0.5 0.35]); xlabel(axEMG,'Time (s)'); ylabel(axEMG,'Amplitude'); title(axEMG,'EMG Signal'); grid(axEMG,'on'); axis(axEMG,'manual'); xlim(axEMG,[0 1]); ylim(axEMG,[-2 2]);

% Gestures gestures = {'Fist','Open','Point','Peace','ThumbsUp'}; nGestures = length(gestures);

% Feature text boxes and labels (يسار) txt_MAV = uicontrol('Style','edit','Position',[20 400 80 25],'Enable','inactive'); txt_RMS = uicontrol('Style','edit','Position',[20 360 80 25],'Enable','inactive'); txt_ZC = uicontrol('Style','edit','Position',[20 320 80 25],'Enable','inactive'); uicontrol('Style','text','Position',[20 430 80 20],'String','Features','FontSize',11,'FontWeight','bold'); uicontrol('Style','text','Position',[110 400 50 20],'String','MAV:','FontSize',10); uicontrol('Style','text','Position',[110 360 50 20],'String','RMS:','FontSize',10); uicontrol('Style','text','Position',[110 320 50 20],'String','ZC:','FontSize',10);

% Prediction text box (يسار أعلى الأزرار) txt_pred = uicontrol('Style','text','Position',[20 240 180 30],... 'String','Predicted Gesture: -','FontSize',11,'BackgroundColor','w');

% Motor shaft visualization (يسار تحت) axMotors = axes('Parent',fig,'Position',[0.05 0.05 0.35 0.2]); title(axMotors,'Motor Shafts (CCW Simulation)'); axis(axMotors,[0 6 0 2]); axis(axMotors,'equal'); axis(axMotors,'off'); hold(axMotors,'on');

% Create 5 circular shafts motorCenters = [1 1; 2 1; 3 1; 4 1; 5 1]; shaftRadius = 0.3; motorHandles = gobjects(1,5); for i = 1:5 motorHandles(i) = plot(axMotors, ... motorCenters(i,1)+shaftRadius*[cos(0) cos(pi/2)], ... motorCenters(i,2)+shaftRadius*[sin(0) sin(pi/2)], ... 'k','LineWidth',4); plot(axMotors,motorCenters(i,1),motorCenters(i,2),'ok','MarkerFaceColor','k'); % center end

% Define which motors (fingers) move for each gesture % order = [Thumb Index Middle Ring Pinky] gestureMotors = { [1 2 3 4 5], % Fist → all [1 2 3 4 5], % Open → all [2], % Point → only index [2 3], % Peace → index + middle [1] % ThumbsUp → only thumb };

% Simulated EMG data (2 electrodes) fs = 200; t = 0:1/fs:1; reps = 3; nElectrodes = 2; rawData = zeros(nGestures*reps,length(t),nElectrodes); labels = zeros(nGestures*reps,1); for g=1:nGestures for r=1:reps idx = (g-1)*reps + r; for e=1:nElectrodes amp = 0.2 + 0.15*e + 0.1*g; freq = 5 + g + 0.5*e + 0.2*r; rawData(idx,:,e) = amp*sin(2*pi*freq*t) + 0.05*randn(1,length(t)); end labels(idx) = g; end end

% Train neural network features = []; targets = []; for i=1:size(rawData,1) sig = squeeze(rawData(i,:,:)); sigF = preprocessEMG(sig,fs); featVec = extractFeatures(sigF); features = [features; featVec]; tVec = zeros(1,nGestures); tVec(labels(i)) = 1; targets = [targets; tVec]; end trainedNN = feedforwardnet(20); trainedNN = train(trainedNN,features',targets');

% Create gesture buttons in rows of 3 buttonWidth = 80; buttonHeight = 35; hSpacing = 10; vSpacing = 10; % spacing buttonsPerRow = 3; startX = 20; startY = 200;

for i = 1:nGestures row = floor((i-1)/buttonsPerRow); col = mod((i-1),buttonsPerRow); xPos = startX + col*(buttonWidth+hSpacing); yPos = startY - row*(buttonHeight+vSpacing); uicontrol('Style','pushbutton', ... 'String',gestures{i}, ... 'Position',[xPos yPos buttonWidth buttonHeight], ... 'FontSize',10, ... 'Callback',@(src,evt) executeGesture(i)); end

%% --- Functions --- function executeGesture(idx) % Hand drawing fingers = getFingersFromGesture(idx); plotHand(fingers,axHand);

    % EMG signal (2 electrodes)
    sig = squeeze(rawData(idx,:,:));
    cla(axEMG); hold(axEMG,'on');
    colors = ['r','b'];
    labelsElectrodes = {'Electrode 1','Electrode 2'};
    for e=1:nElectrodes
        plot(axEMG,t,sig(:,e),colors(e),'LineWidth',1.2);
    end
    legend(axEMG,labelsElectrodes,'Location','northeast');
    xlabel(axEMG,'Time (s)'); ylabel(axEMG,'Amplitude'); title(axEMG,'EMG Signal'); grid(axEMG,'on');

    % Preprocessing
    sigF = preprocessEMG(sig,fs);

    % Feature extraction
    featVec = extractFeatures(sigF);

    % Display features
    set(txt_MAV,'String',num2str(featVec(1),'%.3f'));
    set(txt_RMS,'String',num2str(featVec(2),'%.3f'));
    set(txt_ZC,'String',num2str(featVec(3),'%.0f'));

    % Prediction (currently direct match)
    set(txt_pred,'String',['Predicted Gesture: ',gestures{idx}]);

    % --- Simulated Motor Shafts CCW Rotation ---
    nSteps = 15; % animation steps
    thetaStep = -pi/12; % CCW
    activeMotors = gestureMotors{idx}; % select motors for this gesture

    for step = 1:nSteps
        for m = activeMotors
            x = motorCenters(m,1) + shaftRadius * [cos(step*thetaStep) cos(step*thetaStep + pi/2)];
            y = motorCenters(m,2) + shaftRadius * [sin(step*thetaStep) sin(step*thetaStep + pi/2)];
            set(motorHandles(m),'XData',x,'YData',y);
        end
        drawnow;
        pause(0.05);
    end
end

function fingers = getFingersFromGesture(idx) switch idx case 1, fingers = [0 0 0 0 0]; % Fist case 2, fingers = [1 1 1 1 1]; % Open case 3, fingers = [0 1 0 0 0]; % Point case 4, fingers = [0 1 1 0 0]; % Peace case 5, fingers = [1 0 0 0 0]; % ThumbsUp otherwise, fingers = [0 0 0 0 0]; end end

function preprocessed = preprocessEMG(signal,fs) [b,a] = butter(4,[20 90]/(fs/2)); preprocessed = filtfilt(b,a,double(signal)); preprocessed = (preprocessed - mean(preprocessed))/std(preprocessed); end

function feat = extractFeatures(signal) nCh = size(signal,2); feat = []; for ch=1:nCh x = signal(:,ch); MAV = mean(abs(x)); RMSval = rms(x); ZC = sum(diff(x>0)~=0); feat = [feat MAV RMSval ZC]; end end

function plotHand(fingers,ax) cla(ax); hold(ax,'on'); baseX = [2 3 4 5 6]; baseY = 2*ones(1,5); fingerLength = 6; for i = 1:5 angle = fingers(i)*pi/2; x = [baseX(i) baseX(i) - fingerLength*sin(angle)]; y = [baseY(i) baseY(i) + fingerLength*cos(angle)]; plot(ax,x,y,'k','LineWidth',6); end axis(ax,[0 10 0 10]); axis(ax,'equal'); axis(ax,'off'); end

end