This merge request adds a new application VectorRegression that performs regression on input vector data according to a regression model.
This application is almost the same as VectorClassifier. The differences are:
The label types are different in classification and regression (unsigned int vs float), as well as the ogr field type (OFTInteger versus OFTFloat)
we need to call SetRegressionMode(1) on the model created by the MachineLearningModelFactory
Instead of creating the VectorRegression by copy-pasting almost all the VectorClassifier application, I tried to factorize the code in VectorClassifier using templates.
I created a new application VectorPrediction<bool RegressionMode> that performs regression when RegressionMode =1 and classification when RegressionMode = 0. The code of this application is the code of the old VectorClassifier, whith some modification to use the right types depending on RegressionMode.
VectorClassifier is defined VectorPrediction<0> and VectorRegression as VectorPrediction<1>.
The DoInit method is defined differently for VectorClassifier and VectorRegression using template specialization, to generate different documentations for the two applications
Copyright
The copyright owner is CNES and has signed the ORFEO ToolBox Contributor License Agreement.
Check before merging:
All discussions are resolved
At least 2 votes from core developers, no vote.
The feature branch is (reasonably) up-to-date with the base branch
Dashboard is green
Copyright owner has signed the ORFEO ToolBox Contributor License Agreement
Optionally, run git diff develop... -U0 --no-color | clang-format-diff.py -p1 -i on latest changes and commit
