Styjun

Can a level 2 Warlock take one level in rogue, then continue advancing as a warlock?

How to pronounce 'c++' in Spanish

What was Apollo 13's "Little Jolt" after MECO?

Magical attacks and overcoming damage resistance

How can I wire a 9-position switch so that each position turns on one more LED than the one before?

How do I check if a string is entirely made of the same substring?

Co-worker works way more than he should

What does "function" actually mean in music?

Find the identical rows in a matrix

Von Neumann Extractor - Which bit is retained?

How much of a wave function must reside inside event horizon for it to be consumed by the black hole?

How to not starve gigantic beasts

What to do with someone that cheated their way through university and a PhD program?

How important is it that $TERM is correct?

Find a stone which is not the lightest one

Work requires me to come in early to start computer but wont let me clock in to get paid for it

What is this word supposed to be?

Can someone publish a story that happened to you?

How exactly does Hawking radiation decrease the mass of black holes?

Why did C use the -> operator instead of reusing the . operator?

How to be good at coming up with counter example in Topology

What makes accurate emulation of old systems a difficult task?

I preordered a game on my Xbox while on the home screen of my friend's account. Which of us owns the game?

Multiple options vs single option UI

cross validation methods in matlab

LIBSVM overfittingSVM for HOG features on MatlabPrediction of SVM with custom kernel extremely slow in MatlabRetrieve classification performance from “crossval”How to run SVC classifier after running 10-fold cross validation in sklearn?Pre-processing features before applying cross-validation without leakageScikit Learn- Decision Tree with KFold Cross ValidationEvaluating logistic regression using cross validation and ROCCross Validation metrics with PysparkHow to predict using a gcforest model when I did not keep the model in memory?

.everyoneloves__top-leaderboard:empty,.everyoneloves__mid-leaderboard:empty,.everyoneloves__bot-mid-leaderboard:empty height:90px;width:728px;box-sizing:border-box;

0

while i was working on a binary classification problem using SVM, I found two ways of crossvalidation and I don't know which works best?

first way using crossvalind and loop:

k = 10;
cvFolds = crossvalind('Kfold', data_lables, k); %# get indices of 10-fold
cp = classperf(data_lables); 

for i = 1:k %# for each fold
 testIdx = (cvFolds == i); %# get indices of test instances
 trainIdx = ~testIdx; %# get indices training instances

 %# train an SVM model over training instances
 X= features(trainIdx,:);
 Y = data_lables(trainIdx);
 svmModel = fitcsvm(X,Y,'Standardize',true,'KernelFunction','RBF','KernelScale','auto');

 %# test using test instances
 Z = features(testIdx,:);
 pred = predict(svmModel,Z);

 %# evaluate and update performance object
 cp = classperf(cp, pred, testIdx);

end

While the other way, using the cvpartition

X =Data_Features;
Y = Data_Labels;

%randomize
rand_num = randperm(100);

x_train = X(rand_num(1:80), :);
y_train = Y(rand_num(1:80), :);

x_test = X(rand_num(81: end),:);
y_test = Y(rand_num(81: end),:);

c= cvpartition(y_train, 'k', 10 );

%train SVM
svmModel = fitcsvm(X,Y,'Standardize',true,'KernelFunction','RBF','KernelScale','auto');

pre = predict(svmmodel,xtest);

matlab classification cross-validation

share|improve this question

asked Mar 22 at 16:28

gingin

3182516

add a comment | 

0

while i was working on a binary classification problem using SVM, I found two ways of crossvalidation and I don't know which works best?

first way using crossvalind and loop:

k = 10;
cvFolds = crossvalind('Kfold', data_lables, k); %# get indices of 10-fold
cp = classperf(data_lables); 

for i = 1:k %# for each fold
 testIdx = (cvFolds == i); %# get indices of test instances
 trainIdx = ~testIdx; %# get indices training instances

 %# train an SVM model over training instances
 X= features(trainIdx,:);
 Y = data_lables(trainIdx);
 svmModel = fitcsvm(X,Y,'Standardize',true,'KernelFunction','RBF','KernelScale','auto');

 %# test using test instances
 Z = features(testIdx,:);
 pred = predict(svmModel,Z);

 %# evaluate and update performance object
 cp = classperf(cp, pred, testIdx);

end

While the other way, using the cvpartition

X =Data_Features;
Y = Data_Labels;

%randomize
rand_num = randperm(100);

x_train = X(rand_num(1:80), :);
y_train = Y(rand_num(1:80), :);

x_test = X(rand_num(81: end),:);
y_test = Y(rand_num(81: end),:);

c= cvpartition(y_train, 'k', 10 );

%train SVM
svmModel = fitcsvm(X,Y,'Standardize',true,'KernelFunction','RBF','KernelScale','auto');

pre = predict(svmmodel,xtest);

matlab classification cross-validation

share|improve this question

asked Mar 22 at 16:28

gingin

3182516

add a comment | 

while i was working on a binary classification problem using SVM, I found two ways of crossvalidation and I don't know which works best?

first way using crossvalind and loop:

k = 10;
cvFolds = crossvalind('Kfold', data_lables, k); %# get indices of 10-fold
cp = classperf(data_lables); 

for i = 1:k %# for each fold
 testIdx = (cvFolds == i); %# get indices of test instances
 trainIdx = ~testIdx; %# get indices training instances

 %# train an SVM model over training instances
 X= features(trainIdx,:);
 Y = data_lables(trainIdx);
 svmModel = fitcsvm(X,Y,'Standardize',true,'KernelFunction','RBF','KernelScale','auto');

 %# test using test instances
 Z = features(testIdx,:);
 pred = predict(svmModel,Z);

 %# evaluate and update performance object
 cp = classperf(cp, pred, testIdx);

end

While the other way, using the cvpartition

X =Data_Features;
Y = Data_Labels;

%randomize
rand_num = randperm(100);

x_train = X(rand_num(1:80), :);
y_train = Y(rand_num(1:80), :);

x_test = X(rand_num(81: end),:);
y_test = Y(rand_num(81: end),:);

c= cvpartition(y_train, 'k', 10 );

%train SVM
svmModel = fitcsvm(X,Y,'Standardize',true,'KernelFunction','RBF','KernelScale','auto');

pre = predict(svmmodel,xtest);

matlab classification cross-validation

share|improve this question

asked Mar 22 at 16:28

gingin

3182516

k = 10;
cvFolds = crossvalind('Kfold', data_lables, k); %# get indices of 10-fold
cp = classperf(data_lables); 

for i = 1:k %# for each fold
 testIdx = (cvFolds == i); %# get indices of test instances
 trainIdx = ~testIdx; %# get indices training instances

 %# train an SVM model over training instances
 X= features(trainIdx,:);
 Y = data_lables(trainIdx);
 svmModel = fitcsvm(X,Y,'Standardize',true,'KernelFunction','RBF','KernelScale','auto');

 %# test using test instances
 Z = features(testIdx,:);
 pred = predict(svmModel,Z);

 %# evaluate and update performance object
 cp = classperf(cp, pred, testIdx);

end

X =Data_Features;
Y = Data_Labels;

%randomize
rand_num = randperm(100);

x_train = X(rand_num(1:80), :);
y_train = Y(rand_num(1:80), :);

x_test = X(rand_num(81: end),:);
y_test = Y(rand_num(81: end),:);

c= cvpartition(y_train, 'k', 10 );

%train SVM
svmModel = fitcsvm(X,Y,'Standardize',true,'KernelFunction','RBF','KernelScale','auto');

pre = predict(svmmodel,xtest);

share|improve this question

asked Mar 22 at 16:28

gingin

3182516

share|improve this question

asked Mar 22 at 16:28

gingin

3182516

asked Mar 22 at 16:28

gingin

3182516

asked Mar 22 at 16:28

gingin

3182516