#include <mlpack/core.hpp>#include <mlpack/methods/decision_tree/decision_tree_regressor.hpp>#include <mlpack/methods/decision_tree/mad_gain.hpp>#include <mlpack/methods/decision_tree/mse_gain.hpp>#include <mlpack/methods/decision_tree/random_dimension_select.hpp>#include <mlpack/methods/decision_tree/multiple_random_dimension_select.hpp>#include <mlpack/core/math/random.hpp>#include "catch.hpp"#include "serialization.hpp"#include "mock_categorical_data.hpp"#include "test_function_tools.hpp"
Include dependency graph for decision_tree_regressor_test.cpp:
Functions | |
| void | CreateMultiSplitData (arma::mat &d, arma::rowvec &r, const size_t count, arma::rowvec &values) |
| Creates dataset with 5 groups with all the points in same group have exactly same responses. | |
| TEST_CASE ("MSEGainPerfectTest", "[DecisionTreeRegressorTest]") | |
| Make sure the MSE gain is zero when the responses are perfect. | |
| TEST_CASE ("MSEGainEmptyTest", "[DecisionTreeRegressorTest]") | |
| The MSE gain of an empty vector is 0. | |
| TEST_CASE ("MSEGainHandCalculation", "[DecisionTreeRegressorTest]") | |
| Making sure that MSE gain is evaluated correctly by doing calculation by hand. | |
| TEST_CASE ("MADGainPerfectTest", "[DecisionTreeRegressorTest]") | |
| Make sure the MAD gain is zero when the responses are perfect. | |
| TEST_CASE ("MADGainNormalTest", "[DecisionTreeRegressorTest") | |
| Make sure that when mean of responses is zero, MAD_gain = mean of absolute values of the distribution. | |
| TEST_CASE ("MADGainEmptyTest", "[DecisionTreeRegressorTest]") | |
| The MAD gain of an empty vector is 0. | |
| TEST_CASE ("MADGainHandCalculation", "[DecisionTreeRegressorTest]") | |
| Making sure that MAD gain is evaluated correctly by doing calculation by hand. | |
| TEST_CASE ("AllCategoricalSplitSimpleSplitTest_", "[DecisionTreeRegressorTest]") | |
| Check that AllCategoricalSplit will split when the split is obviously better. | |
| TEST_CASE ("AllCategoricalSplitMinSamplesTest_", "[DecisionTreeRegressorTest]") | |
| Make sure that AllCategoricalSplit respects the minimum number of samples required to split. | |
| TEST_CASE ("AllCategoricalSplitNoGainTest_", "[DecisionTreeRegressorTest]") | |
| Check that no split is made when it doesn't get us anything. | |
| TEST_CASE ("BestBinaryNumericSplitSimpleSplitTest_", "[DecisionTreeRegressorTest]") | |
| Check that the BestBinaryNumericSplit will split on an obviously splittable dimension. | |
| TEST_CASE ("BestBinaryNumericSplitMinSamplesTest_", "[DecisionTreeRegressorTest]") | |
| Check that the BestBinaryNumericSplit won't split if not enough points are given. | |
| TEST_CASE ("BestBinaryNumericSplitNoGainTest_", "[DecisionTreeRegressorTest]") | |
| Check that the BestBinaryNumericSplit doesn't split a dimension that gives no gain. | |
| TEST_CASE ("RandomBinaryNumericSplitAlwaysSplit_", "[DecisionTreeRegressorTest]") | |
| Check that the RandomBinaryNumericSplit always splits when splitIfBetterGain is false. | |
| TEST_CASE ("RandomBinaryNumericSplitMinSamplesTest_", "[DecisionTreeRegressorTest]") | |
| Check that the RandomBinaryNumericSplit won't split if not enough points are given. | |
| TEST_CASE ("RandomBinaryNumericSplitNoGainTest_", "[DecisionTreeRegressorTest]") | |
| Check that the RandomBinaryNumericSplit doesn't split a dimension that gives no gain when splitIfBetterGain is true. | |
| TEST_CASE ("BasicConstructionTest_", "[DecisionTreeRegressorTest]") | |
| A basic construction of the decision tree—ensure that we can create the tree and that it split at least once. | |
| TEST_CASE ("BasicConstructionTestWithWeight_", "[DecisionTreeRegressorTest]") | |
| Construct a tree with weighted responses. | |
| TEST_CASE ("PerfectTrainingSet_", "[DecisionTreeRegressorTest]") | |
| Construct the decision tree on numeric data only and see that we can fit it exactly and achieve perfect performance on the training set. | |
| TEST_CASE ("PerfectTrainingSetWithWeight_", "[DecisionTreeRegressorTest]") | |
| Construct the decision tree with weighted responses. | |
| TEST_CASE ("MultiSplitTest1", "[DecisionTreeRegressorTest]") | |
| Test that the tree is able to perfectly fit all the obvious splits present in the data. More... | |
| TEST_CASE ("MultiSplitTest2", "[DecisionTreeRegressorTest]") | |
| Test that the tree is able to perfectly fit all the obvious splits present in the data. More... | |
| TEST_CASE ("MultiSplitTest3", "[DecisionTreeRegressorTest]") | |
| Test that the tree is able to perfectly fit all the obvious splits present in the data. More... | |
| TEST_CASE ("NumericalBuildTest", "[DecisionTreeRegressorTest]") | |
| Test that the tree builds correctly on unweighted numerical dataset. | |
| TEST_CASE ("NumericalBuildTestWithWeights", "[DecisionTreeRegressorTest]") | |
| Test that the tree builds correctly on weighted numerical dataset. | |
| TEST_CASE ("CategoricalBuildTest_", "[DecisionTreeRegressorTest]") | |
| Test that we can build a decision tree on a simple categorical dataset. | |
| TEST_CASE ("CategoricalBuildTestWithWeight_", "[DecisionTreeRegressorTest]") | |
| Test that we can build a decision tree with weights on a simple categorical dataset. | |
| TEST_CASE ("CategoricalWeightedBuildTest_", "[DecisionTreeRegressorTest]") | |
| Test that we can build a decision tree on a simple categorical dataset using weights, with low-weight noise added. | |
| TEST_CASE ("CategoricalMADGainWeightedBuildTest", "[DecisionTreeRegressorTest]") | |
| Test that we can build a decision tree using MAD gain on a simple categorical dataset using weights, with low-weight noise added. | |
| TEST_CASE ("SimpleGeneralizationTest_", "[DecisionTreeRegressorTest]") | |
| Test that the decision tree generalizes reasonably. | |
| TEST_CASE ("SimpleGeneralizationFMatTest_", "[DecisionTreeRegressorTest]") | |
| Test that the decision tree generalizes reasonably when built on float data. | |
| TEST_CASE ("WeightedDecisionTreeTest_", "[DecisionTreeRegressorTest]") | |
| Test that we can build a decision tree using weighted data (where the low-weighted data is random noise), and that the tree still builds correctly enough to get good results. | |
| TEST_CASE ("WeightedDecisionTreeMADGainTest", "[DecisionTreeRegressorTest]") | |
| Test that we can build a decision tree using weighted data (where the low-weighted data is random noise) with MAD gain, and that the tree still builds correctly enough to get good results. | |
Detailed Description
Tests for the DecisionTreeRegressor class and related classes.
mlpack is free software; you may redistribute it and/or modify it under the terms of the 3-clause BSD license. You should have received a copy of the 3-clause BSD license along with mlpack. If not, see http://www.opensource.org/licenses/BSD-3-Clause for more information.
Function Documentation
◆ TEST_CASE() [1/3]
| TEST_CASE | ( | "MultiSplitTest1" | , |
| "" | [DecisionTreeRegressorTest] | ||
| ) |
Test that the tree is able to perfectly fit all the obvious splits present in the data.
| | 2 | xxxxxx | | 1 | xxxxxx xxxxxx | | 0 |xxxxxx xxxxxx |___________________________________
◆ TEST_CASE() [2/3]
| TEST_CASE | ( | "MultiSplitTest2" | , |
| "" | [DecisionTreeRegressorTest] | ||
| ) |
Test that the tree is able to perfectly fit all the obvious splits present in the data.
Same test as above, but with less data.
◆ TEST_CASE() [3/3]
| TEST_CASE | ( | "MultiSplitTest3" | , |
| "" | [DecisionTreeRegressorTest] | ||
| ) |
Test that the tree is able to perfectly fit all the obvious splits present in the data.
| 20 | xxxxxx | | 15 | xxxxxx | | 10 | xxxxxx | | 5 | xxxxxx | | 0 |xxxxxx |________________________________________
