Decision Tree module. More...
#include <fstream>#include <stack>#include <algorithm>#include <cstring>#include "random_number_generator.h"
Include dependency graph for decision_tree.h:
This graph shows which files directly or indirectly include this file:
Go to the source code of this file.
Classes | |
| class | koho::Node |
| Node of a binary tree. More... | |
| class | koho::Tree |
| Binary tree structure build up of nodes. More... | |
| class | koho::GiniCriterion |
| Gini Index impurity criterion. More... | |
| class | koho::BestSplitter |
| Splitter to find the best split for a node. More... | |
| class | koho::DepthFirstTreeBuilder |
| Build a binary decision tree in depth-first order. More... | |
| class | koho::DecisionTreeClassifier |
| A decision tree classifier. More... | |
Namespaces | |
| koho | |
Typedefs | |
| typedef double | koho::Features_t |
| typedef long | koho::Classes_t |
| typedef double | koho::ClassWeights_t |
| typedef double | koho::Histogram_t |
| typedef unsigned long | koho::SamplesIdx_t |
| typedef unsigned long | koho::FeaturesIdx_t |
| typedef unsigned long | koho::ClassesIdx_t |
| typedef unsigned long | koho::OutputsIdx_t |
| typedef unsigned long | koho::NodesIdx_t |
| typedef unsigned long | koho::TreeDepthIdx_t |
Variables | |
| const double | koho::PRECISION_EQUAL = 1e-7 |
Detailed Description
Decision Tree module.
- Classification
- Numerical (dense) data
- Missing values (Not Missing At Random (NMAR)
- Class balancing
- Multi-Class
- Multi-Output (single model)
- Build order: depth first
- Impurity criteria: gini
- Split a. features: best over k (incl. all) random features
- Split b. thresholds: 1 random or all thresholds
- Stop criteria: max depth, (pure, no improvement)
- Important Features
Optimized Implementation: stack, samples LUT with in-place partitioning, incremental histogram updates
C++ implementation.
