Sign Up to our social questions and Answers Engine to ask questions, answer people’s questions, and connect with other people.
Login to our social questions & Answers Engine to ask questions answer people’s questions & connect with other people.
Lost your password? Please enter your email address. You will receive a link and will create a new password via email.
Please briefly explain why you feel this question should be reported.
Please briefly explain why you feel this answer should be reported.
Please briefly explain why you feel this user should be reported.
I am using the Neural Network for a classification task with (1 Hidden Layer and 10 features).
Result is not quite good. I got high error rate in training dataset itself.
What should i do now ?
Do I need to increase the number of nodes in Hidden Layers ? What will be the impact ?
Do I need to increase the number of input features to the Hidden Layer ? What will be the impact ?
Given the current configuration of your network and not knowing more about the data set, I recommend adding a second hidden layer with only a few nodes (maybe 4). That will allow for more variability in the types of decision surfaces generated (e.g., multiple distinct clusters for a single class).
Even though you are doing binary classification, I would also split the output into two nodes (one for true, one for false) and take the max value as your classification result. I usually see better convergence that way and interpreting errors is a bit more intuitive.