0
$\begingroup$

I have a multi vairable binary target variable like this with multiple TRUE values.

    Y1  Y2  Y3  Y4
1   0   0   1   1
2   1   1   1   1
3   1   0   1   0
4   1   0   0   1
5   0   1   1   0
6   1   1   1   0
7   0   0   0   1
8   1   1   1   0
9   1   1   1   0
10  0   1   1   0
11  0   0   0   1
12  1   1   1   1

How should I prepare my output neural network such a way that I can predict multiple true variable?

I have tried using softmax activation function in the output layer for a multi variable single TRUE classification problems whose actual value is like this

Actual output value= [0, 0, 0, 1]
Predicted output value= [0.1,   0.2,    0.1,    0.6]

Then I take argmax of the predicted array which gives output 3 which result in [0, 0, 0, 1]. But what method should I follow for a classification problem with multiple true value output.

$\endgroup$
1
$\begingroup$

You could simply replace the softmax in the output layer with sigmoid, and train it with the binary cross entropy loss.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.