How to select histogram bins for machine learning task?

Question

I'm using a histogram values as features in the machine learning task.
How to select the best bins?

I thought at the beginning to break the range into large bins, than break significant bins again but until when?

And may be there a special methods for histogram bins selection in machine learning case?

Answer 1

This is one of the big challenges of designing any learning machine, and as far as I know not something which can be answered on a general basis. Try a few times with your real data on your specific problem, and decide on something which seems reasonable. That will probably give you a better result than anything any of us are able to give you.