|
LeastAngleRegressionOptions Class Reference |  |
Pass options to leastAngleRegression(). More...
#include <vigra/regression.hxx>
Public Member Functions | |
| LeastAngleRegressionOptions & | lars () |
| LeastAngleRegressionOptions & | lasso () |
| LeastAngleRegressionOptions () | |
| LeastAngleRegressionOptions & | leastSquaresSolutions (bool select=true) |
| LeastAngleRegressionOptions & | maxSolutionCount (unsigned int n) |
| LeastAngleRegressionOptions & | nnlasso () |
| LeastAngleRegressionOptions & | setMode (std::string mode) |
Detailed Description
Pass options to leastAngleRegression().
#include <vigra/regression.hxx>
Namespaces: vigra and vigra::linalg
Constructor & Destructor Documentation
Initialize all options with default values.
Member Function Documentation
| LeastAngleRegressionOptions& maxSolutionCount | ( | unsigned int | n | ) |
Maximum number of solutions to be computed.
If n is 0 (the default), the number of solutions is determined by the length of the solution array. Otherwise, the minimum of maxSolutionCount() and that length is taken.
Default: 0 (use length of solution array)
| LeastAngleRegressionOptions& setMode | ( | std::string | mode | ) |
Set the mode of the algorithm.
Mode must be one of "lars", "lasso", "nnlasso". The function just calls the member function of the corresponding name to set the mode.
Default: "lasso"
| LeastAngleRegressionOptions& lars | ( | ) |
Use the plain LARS algorithm.
Default: inactive
| LeastAngleRegressionOptions& lasso | ( | ) |
Use the LASSO modification of the LARS algorithm.
This allows features to be removed from the active set under certain conditions.
Default: active
| LeastAngleRegressionOptions& nnlasso | ( | ) |
Use the non-negative LASSO modification of the LARS algorithm.
This enforces all non-zero entries in the solution to be positive.
Default: inactive
| LeastAngleRegressionOptions& leastSquaresSolutions | ( | bool | select = true | ) |
Compute least squares solutions.
Use least angle regression to determine active sets, but return least squares solutions for the features in each active set, instead of constrained solutions.
Default: true
The documentation for this class was generated from the following file:
- vigra/regression.hxx