0

Here I have a multi-task network to detect if there is a talking user and where he is in a clip of audio, which has two outputs. One of them is a binary classification problem indicating if there is a user. Another output is a regression problem used to estimate the direction of the user. The input is a feature vector extracted from this audio clip.

If the classification output is 1, which means there is a user, then the output of regression has a meaningful value. So I want to set the target label data. The target label of classification can be easily set to 0 or 1. But I do not know how to set the label for regression for no-user cases. Is it valid to set it as a negative value (to differ it from the positive meaningful direction value)?

I know in a classification problem, we can set another new class to represent the no-exist case. How to do this in a regression model?

Qiang
  • 1
  • 2
  • 1
    Why is this a problem, please give us more context! But, if the problem is that this output is further used as input in other models, then one solution is here: https://stats.stackexchange.com/questions/372257/how-do-you-deal-with-nested-variables-in-a-regression-model/372258#372258 – kjetil b halvorsen Jul 08 '21 at 15:06
  • Hello, I have updated my problem description. I want to use a network to predict if there is a talking user and where he is in a clip of audio. – Qiang Jul 09 '21 at 07:12
  • That's fine, but unless you tell us more details nobody will be able to say more than generalities! – kjetil b halvorsen Jul 13 '21 at 02:02

0 Answers0