|
ppforest2 v0.1.0
Projection Pursuit Decision Trees and Random Forests
|
Minimal mode-correct TrainingSpec::Ptr for structural tests.
More...
#include "models/TrainingSpec.hpp"Go to the source code of this file.
Namespaces | |
| namespace | ppforest2 |
| Binarization strategies for multiclass-to-binary reduction. | |
| namespace | ppforest2::test |
Functions | |
| TrainingSpec::Ptr | ppforest2::test::classification_spec () |
| Default-valued classification spec. | |
| TrainingSpec::Ptr | ppforest2::test::regression_spec () |
| Minimal well-formed regression spec. | |
Minimal mode-correct TrainingSpec::Ptr for structural tests.
Used by tests that build a Tree or Forest directly from a hand-built TreeNode graph and just need to satisfy the subclass/mode invariants on ClassificationTree, RegressionTree, etc. Strategy choices are irrelevant — the spec is never trained against.
Tests that actually train should construct their TrainingSpec inline via TrainingSpec::builder(...); the kind of model being trained is part of what the test is documenting.