why mse is 0.00 for three different data sets?

Question

Sanchit 2023-7-10

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1993783-why-mse-is-0-00-for-three-different-data-sets

回答： Menika 2023-7-10

Please let me know what is wrong in matalb code given below because it is giving mean square error is for three different data sets,

clear, clc, close all

data = readtable('c:/matlab/study_data.csv');

X = data(:, 1:end-1); % Select all columns except the last one

y = data(:, end); % Select the last column

numGroundTruth = numel(y);

numTrainingSamples = round(0.8 * numGroundTruth);

trainingIndexes = randsample(numGroundTruth, numTrainingSamples);

testIndexes = setdiff((1:numGroundTruth)', trainingIndexes);

X_train = X(trainingIndexes, :);

X_test = X(testIndexes, :);

y_train = y(trainingIndexes, :);

y_test = y(testIndexes, :);

% Create a Random Forest classifier

rf_classifier = TreeBagger(100, table2array(X_train), table2array(y_train), 'OOBPrediction', 'On');

predicted = predict(rf_classifier, table2array(X_train));

YY = categorical(predicted);

ZZ = str2double(cellstr(YY));

Z = table2array(y_train);

oob_mse = immse(Z, ZZ);

disp(sprintf('Out-of-Bag Mean Square Error: %.4f', oob_mse));

Thanks for your kind help.

Sanchit

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

Menika 2023-7-10

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1993783-why-mse-is-0-00-for-three-different-data-sets#answer_1270088

Hi,

A possible problem with the above code can be that the predicted variable is being calculated using the training data X_train, rather than the test data X_test. Since the MSE is calculated using the training data, it will always be zero because the model is predicting the same data it was trained on. You can try replacing X_train with X_test when calculating the predicted variable.

Hope it helps!