When I want to use Caffe to create my own CNN, how to determine the number of convolution and pooling layer I need is suitable to extract correct features basis?

Is there any principal or documents?

Thank you~


1 Answer 1


I am new at this too, so there may be a better answer. But I would start small, have only one hidden layer get the best accuracy out of that layer as possible then add another and see if it improves. At each stage, you are changing parameters (size of filter, learning rate, weight initialization etc.) Chapter 9 of this book: http://www.deeplearningbook.org/ might be helpful.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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