Difference between revisions of "Adaptive Modeling Mode"
From SUMOwiki
Jump to navigationJump to search| (One intermediate revision by one other user not shown) | |||
| Line 1: | Line 1: | ||
It is possible to switch off sample selection and run the toolbox for a fixed set of samples, only optimizing the model parameters. This is what we call running in Adaptive Modeling Mode. | It is possible to switch off sample selection and run the toolbox for a fixed set of samples, only optimizing the model parameters. This is what we call running in Adaptive Modeling Mode. | ||
| − | You can switch off adaptive sample selection if you simply do not specify a < | + | You can switch off adaptive sample selection if you simply do not specify a <SequentialDesign> tag in your configuration file. Just remove it from the <Plan> section. When you then run the toolbox, all the available data will be used and only adaptive modeling will be done. There are two possibilities: |
#Your simulator is a dataset: the whole dataset will be read in one go and all of the data therein will be used for modeling | #Your simulator is a dataset: the whole dataset will be read in one go and all of the data therein will be used for modeling | ||
Latest revision as of 16:35, 27 February 2014
It is possible to switch off sample selection and run the toolbox for a fixed set of samples, only optimizing the model parameters. This is what we call running in Adaptive Modeling Mode.
You can switch off adaptive sample selection if you simply do not specify a <SequentialDesign> tag in your configuration file. Just remove it from the <Plan> section. When you then run the toolbox, all the available data will be used and only adaptive modeling will be done. There are two possibilities:
- Your simulator is a dataset: the whole dataset will be read in one go and all of the data therein will be used for modeling
- Your simulator is an executable or Matlab script: only the initial design will be generated and used for adaptive modeling, no further samples will be selected.
This is useful if you just want to see what the best model is you can get for a fixed dataset in a certain amount of time.
