Find optimal parameters for regression
Greetings.
I've recently been toying with this incredible tool, however, I've a question, parhaps a very newbie one.
I've sucessfully created a linear regression model by using the x-validation operator, modeling, apply model, etc. Now I've a valid model, is there a process or operator that allows me to, given a threshold for the attributes, to find the values of them for a maximum output?
I've recently been toying with this incredible tool, however, I've a question, parhaps a very newbie one.
I've sucessfully created a linear regression model by using the x-validation operator, modeling, apply model, etc. Now I've a valid model, is there a process or operator that allows me to, given a threshold for the attributes, to find the values of them for a maximum output?
Find more posts tagged with
Sort by:
1 - 2 of
21
Hi Edin, thanks for your reply!
Thanks again!
I actually do not intend to optimize the datasent (I've already tried the Optimize selection operators, for example), but what I need to do is, for a given linear model already determined with rapidmaner, to find which combination of values of the attributes will produce the estimated value to be the maximum.
Edin Klapic wrote: Edin Klapic wrote: Edin Klapic wrote:
Which values do you want to optimize? What do mean with ?
Thanks again!
Since RapidMiner offers various ways to optimize your results (e.g. Optimize Parameters, Drop Uncertain Predictions, ...) it is necessary to know which part of the result you want to optimize.
Which values do you want to optimize? What do mean with ?
Best,
Edin