Solving Regression problems with BP

#1

Hi!
This might be a rookie question, still…As explained in the video, various classification datasets achieve a decent score using BP… I was trying to apply the same algo to a regression dataset, but the MSE/ RMSE was going above 1, doubtful if except for the loss function do we need to tweak something else also in the algo to work for regression problems…

#3

hey @manmeetkaur0175 ,
to perform regression task using neural networks you need to make some changes .

  1. Change loss function to MSE/RMSE
  2. Change the output layer activation function to either relu or linear.

These two will make it a regression task , now to improve performance.
3. You might need to make changes in network , like adding BatchNormalization layers or Dropouts to improve performance.

These can help you work and achieve a decent score.

#4

Thanks! I will try to make these changes and check the score…

#5

yeah ,ofcourse.
I hope this will help you.

#6

I hope I’ve cleared your doubt. I ask you to please rate your experience here
Your feedback is very important. It helps us improve our platform and hence provide you
the learning experience you deserve.

On the off chance, you still have some questions or not find the answers satisfactory, you may reopen
the doubt.