How to import csv deep learning dataset with labels to matlab?

Question

rami dishlo 2022-3-5

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1664229-how-to-import-csv-deep-learning-dataset-with-labels-to-matlab

评论： rami dishlo 2022-3-15

采纳的回答： yanqi liu

Hello everyone,

I am trying to train a fully connected deep learning model.

I have my data set in a csv file so that each row represents a different signal.

The first 56 coloums represents the signal and the 2 last coloums represents the labels for the signals (there are two labels).

How can I import the cvs file in a way that i will be able to train a deep learning network with it?

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

yanqi liu 2022-3-7

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1664229-how-to-import-csv-deep-learning-dataset-with-labels-to-matlab#answer_911524

yes，sir，may be read csv and reshape the data(:,1:56) into 4-D as train_input，data(:, 57:58) make to label vector as train_output

if possible，may be upload your csv to analysis

3 个评论
显示 1更早的评论隐藏 1更早的评论

yanqi liu 2022-3-10

在 MATLAB Online 中打开

yes，sir，now we can use

data = load('Train Data.csv');
% make X and Y 
X = data(:, 10 : 65);
Y = data(:, 66 : 67);
[~, Y] = max(Y');
X = X';
Y = Y';
% make cnn
num_class = length(unique(Y));
% make data shuffle
rand('seed', 0)
ind = randperm(size(X, 2));
X = X(:,ind);
Y = Y(ind);
Y = categorical(Y);
% Split Data
rate = 0.8;
ind_split = round(length(Y)*rate);
train_X = X(:,1:ind_split);
train_Y = Y(1:ind_split);
val_X = X(:,ind_split+1:end);
val_Y = Y(ind_split+1:end);
% Data Batch
XTrain=(reshape(train_X, [size(X,1),1,1,size(train_X,2)]));
XVal=(reshape(val_X', [size(X,1),1,1,size(val_X,2)]));
% CNN
layers = [imageInputLayer([size(X,1) 1 1])  
    convolution2dLayer([30 1],3,'Stride',1)
    dropoutLayer
    batchNormalizationLayer
    reluLayer
    maxPooling2dLayer(2,'Stride',2,'Padding',[0 0 0 1])
    dropoutLayer
    fullyConnectedLayer(num_class)
    softmaxLayer
    classificationLayer];
% Specify training options.
opts = trainingOptions('adam', ...
    'MaxEpochs',200, ...
    'Shuffle','every-epoch', ...
    'Plots','training-progress', ...
    'Verbose',false, ...
    'ValidationData',{XVal,val_Y},...
    'ExecutionEnvironment', 'cpu', ...
    'ValidationPatience',Inf);
% Train
yc = categorical(train_Y);
net1 = trainNetwork(XTrain,yc,layers,opts);
% Test
miniBatchSize = 27;
YPred = classify(net1,XVal, ...
    'MiniBatchSize',miniBatchSize,...
    'ExecutionEnvironment', 'cpu');
acc = mean(YPred(:) == val_Y(:))
figure
t = confusionchart(val_Y(:),YPred(:));