xmatthias
commented
8 months ago The objective column is the value that's being optimized by hyperopt (in a multi-dimensional space, usually).
How that's calculated (and what factors will be included in it's calculation) will depend on the actual loss function used. In general, hyperopt will simply try to minimize that function - which means that lower values are "better" hyperopt results (not necessarily more profitable - depending on the loss function used).
It has no meaning / comparison value outside of the same loss function - as every loss-function has other rules - and other ranges the objective value can take.
Hello, I have a question about the output of hyperopts. What does the "Objective" column mean? I couldn't find anything about it in the documentation. Sometimes it contains positive numbers, sometimes negative. Thanks in advance! VerDoe