rmontanana/BayesNet
1
0
Fork 0
Code Issues 2 Pull Requests 1 Actions Packages Projects Releases 6 Wiki Activity

Optimize BoostAODE -> XBAODE #33

Merged
rmontanana merged 27 commits from WA2DE into main 2025-03-16 17:58:10 +00:00
Conversation 0 Commits 27 Files Changed 48 +1594 -66
35 changed files with 1594 additions and 66 deletions
Download Patch File Download Diff File Expand all files Collapse all files
Showing only changes of commit 71b05cc1a7 - Show all commits

4
Makefile
View File

@@ -97,7 +97,7 @@ fname = "tests/data/iris.arff"
sample: ## Build sample sample: ## Build sample
@echo ">>> Building Sample..."; @echo ">>> Building Sample...";
@if [ -d ./sample/build ]; then rm -rf ./sample/build; fi @if [ -d ./sample/build ]; then rm -rf ./sample/build; fi
@cd sample && cmake -B build -S . && cmake --build build -t bayesnet_sample @cd sample && cmake -B build -S . -D CMAKE_BUILD_TYPE=Debug && cmake --build build -t bayesnet_sample
sample/build/bayesnet_sample $(fname) sample/build/bayesnet_sample $(fname)
@echo ">>> Done"; @echo ">>> Done";
@@ -105,7 +105,7 @@ fname = "tests/data/iris.arff"
sample2: ## Build sample2 sample2: ## Build sample2
@echo ">>> Building Sample..."; @echo ">>> Building Sample...";
@if [ -d ./sample/build ]; then rm -rf ./sample/build; fi @if [ -d ./sample/build ]; then rm -rf ./sample/build; fi
@cd sample && cmake -B build -S . && cmake --build build -t bayesnet_sample_xspode @cd sample && cmake -B build -S . -D CMAKE_BUILD_TYPE=Debug && cmake --build build -t bayesnet_sample_xspode
sample/build/bayesnet_sample_xspode $(fname) sample/build/bayesnet_sample_xspode $(fname)
@echo ">>> Done"; @echo ">>> Done";

163
bayesnet/classifiers/XSPODE.cc
View File

@@ -3,15 +3,14 @@
// SPDX-FileType: SOURCE // SPDX-FileType: SOURCE
// SPDX-License-Identifier: MIT // SPDX-License-Identifier: MIT
// *************************************************************** // ***************************************************************
#include <limits>
#include <algorithm>
#include <numeric>
#include <cmath>
#include <stdexcept>
#include <sstream>
#include "XSPODE.h" #include "XSPODE.h"
#include "bayesnet/utils/TensorUtils.h" #include "bayesnet/utils/TensorUtils.h"
#include <algorithm>
#include <cmath>
#include <limits>
#include <numeric>
#include <sstream>
#include <stdexcept>
namespace bayesnet { namespace bayesnet {
@@ -19,12 +18,9 @@ namespace bayesnet {
// Constructor // Constructor
// -------------------------------------- // --------------------------------------
XSpode::XSpode(int spIndex) XSpode::XSpode(int spIndex)
: superParent_{ spIndex }, : superParent_{ spIndex }, nFeatures_{ 0 }, statesClass_{ 0 }, alpha_{ 1.0 },
nFeatures_{ 0 }, initializer_{ 1.0 }, semaphore_{ CountingSemaphore::getInstance() },
statesClass_{ 0 }, Classifier(Network())
alpha_{ 1.0 },
initializer_{ 1.0 },
semaphore_{ CountingSemaphore::getInstance() }, Classifier(Network())
{ {
validHyperparameters = { "parent" }; validHyperparameters = { "parent" };
} }
@@ -39,10 +35,12 @@ namespace bayesnet {
Classifier::setHyperparameters(hyperparameters); Classifier::setHyperparameters(hyperparameters);
} }
void XSpode::fit(std::vector<std::vector<int>>& X, std::vector<int>& y, torch::Tensor& weights_, const Smoothing_t smoothing) void XSpode::fit(torch::Tensor & X, torch::Tensor& y, torch::Tensor& weights_, const Smoothing_t smoothing)
{ {
m = X[0].size(); m = X.size(1);
n = X.size(); n = X.size(0);
dataset = X;
buildDataset(y);
buildModel(weights_); buildModel(weights_);
trainModel(weights_, smoothing); trainModel(weights_, smoothing);
fitted = true; fitted = true;
@@ -51,7 +49,8 @@ namespace bayesnet {
// -------------------------------------- // --------------------------------------
// trainModel // trainModel
// -------------------------------------- // --------------------------------------
// Initialize storage needed for the super-parent and child features counts and probs. // Initialize storage needed for the super-parent and child features counts and
// probs.
// -------------------------------------- // --------------------------------------
void XSpode::buildModel(const torch::Tensor& weights) void XSpode::buildModel(const torch::Tensor& weights)
{ {
@@ -60,12 +59,13 @@ namespace bayesnet {
// Derive the number of states for each feature and for the class. // Derive the number of states for each feature and for the class.
// (This is just one approach; adapt to match your environment.) // (This is just one approach; adapt to match your environment.)
// Here, we assume the user also gave us the total #states per feature in e.g. statesMap. // Here, we assume the user also gave us the total #states per feature in e.g.
// We'll simply reconstruct the integer states_ array. The last entry is statesClass_. // statesMap. We'll simply reconstruct the integer states_ array. The last
// entry is statesClass_.
states_.resize(nFeatures_); states_.resize(nFeatures_);
for (int f = 0; f < nFeatures_; f++) { for (int f = 0; f < nFeatures_; f++) {
// Suppose you look up in “statesMap” by the feature name, or read directly from X. // Suppose you look up in “statesMap” by the feature name, or read directly
// We'll assume states_[f] = max value in X[f] + 1. // from X. We'll assume states_[f] = max value in X[f] + 1.
states_[f] = dataset[f].max().item<int>() + 1; states_[f] = dataset[f].max().item<int>() + 1;
} }
// For the class: states_.back() = max(y)+1 // For the class: states_.back() = max(y)+1
@@ -77,15 +77,18 @@ namespace bayesnet {
// We'll store these counts in spFeatureCounts_[spVal * statesClass_ + c]. // We'll store these counts in spFeatureCounts_[spVal * statesClass_ + c].
spFeatureCounts_.resize(states_[superParent_] * statesClass_, 0.0); spFeatureCounts_.resize(states_[superParent_] * statesClass_, 0.0);
// For each child ≠ sp, we store p(childVal| c, spVal) in a separate block of childCounts_. // For each child ≠ sp, we store p(childVal| c, spVal) in a separate block of
// childCounts_ will be sized as sum_{child≠sp} (states_[child] * statesClass_ * states_[sp]). // childCounts_. childCounts_ will be sized as sum_{child≠sp} (states_[child]
// We also need an offset for each child to index into childCounts_. // * statesClass_ * states_[sp]). We also need an offset for each child to
// index into childCounts_.
childOffsets_.resize(nFeatures_, -1); childOffsets_.resize(nFeatures_, -1);
int totalSize = 0; int totalSize = 0;
for (int f = 0; f < nFeatures_; f++) { for (int f = 0; f < nFeatures_; f++) {
if (f == superParent_) continue; // skip sp if (f == superParent_)
continue; // skip sp
childOffsets_[f] = totalSize; childOffsets_[f] = totalSize;
// block size for this child's counts: states_[f] * statesClass_ * states_[superParent_] // block size for this child's counts: states_[f] * statesClass_ *
// states_[superParent_]
totalSize += (states_[f] * statesClass_ * states_[superParent_]); totalSize += (states_[f] * statesClass_ * states_[superParent_]);
} }
childCounts_.resize(totalSize, 0.0); childCounts_.resize(totalSize, 0.0);
@@ -99,7 +102,8 @@ namespace bayesnet {
// p(x_child| c, x_sp) for all child ≠ sp // p(x_child| c, x_sp) for all child ≠ sp
// //
// -------------------------------------- // --------------------------------------
void XSpode::trainModel(const torch::Tensor& weights, const bayesnet::Smoothing_t smoothing) void XSpode::trainModel(const torch::Tensor& weights,
const bayesnet::Smoothing_t smoothing)
{ {
// Accumulate raw counts // Accumulate raw counts
for (int i = 0; i < m; i++) { for (int i = 0; i < m; i++) {
@@ -120,7 +124,8 @@ namespace bayesnet {
default: default:
alpha_ = 0.0; // No smoothing alpha_ = 0.0; // No smoothing
} }
initializer_ = std::numeric_limits<double>::max() / (nFeatures_ * nFeatures_); // for numerical stability initializer_ = std::numeric_limits<double>::max() /
(nFeatures_ * nFeatures_); // for numerical stability
// Convert raw counts to probabilities // Convert raw counts to probabilities
computeProbabilities(); computeProbabilities();
} }
@@ -134,7 +139,8 @@ namespace bayesnet {
// //
void XSpode::addSample(const std::vector<int>& instance, double weight) void XSpode::addSample(const std::vector<int>& instance, double weight)
{ {
if (weight <= 0.0) return; if (weight <= 0.0)
return;
int c = instance.back(); int c = instance.back();
// (A) increment classCounts // (A) increment classCounts
@@ -146,7 +152,8 @@ namespace bayesnet {
// (C) increment child counts => p(childVal | c, x_sp) // (C) increment child counts => p(childVal | c, x_sp)
for (int f = 0; f < nFeatures_; f++) { for (int f = 0; f < nFeatures_; f++) {
if (f == superParent_) continue; if (f == superParent_)
continue;
int childVal = instance[f]; int childVal = instance[f];
int offset = childOffsets_[f]; int offset = childOffsets_[f];
// Compute index in childCounts_. // Compute index in childCounts_.
@@ -169,7 +176,8 @@ namespace bayesnet {
// -------------------------------------- // --------------------------------------
void XSpode::computeProbabilities() void XSpode::computeProbabilities()
{ {
double totalCount = std::accumulate(classCounts_.begin(), classCounts_.end(), 0.0); double totalCount =
std::accumulate(classCounts_.begin(), classCounts_.end(), 0.0);
// p(c) => classPriors_ // p(c) => classPriors_
classPriors_.resize(statesClass_, 0.0); classPriors_.resize(statesClass_, 0.0);
@@ -181,14 +189,15 @@ namespace bayesnet {
} }
} else { } else {
for (int c = 0; c < statesClass_; c++) { for (int c = 0; c < statesClass_; c++) {
classPriors_[c] = (classCounts_[c] + alpha_) classPriors_[c] =
/ (totalCount + alpha_ * statesClass_); (classCounts_[c] + alpha_) / (totalCount + alpha_ * statesClass_);
} }
} }
// p(x_sp | c) // p(x_sp | c)
spFeatureProbs_.resize(spFeatureCounts_.size()); spFeatureProbs_.resize(spFeatureCounts_.size());
// denominator for spVal * statesClass_ + c is just classCounts_[c] + alpha_ * (#states of sp) // denominator for spVal * statesClass_ + c is just classCounts_[c] + alpha_ *
// (#states of sp)
int spCard = states_[superParent_]; int spCard = states_[superParent_];
for (int spVal = 0; spVal < spCard; spVal++) { for (int spVal = 0; spVal < spCard; spVal++) {
for (int c = 0; c < statesClass_; c++) { for (int c = 0; c < statesClass_; c++) {
@@ -201,7 +210,8 @@ namespace bayesnet {
// p(x_child | c, x_sp) // p(x_child | c, x_sp)
childProbs_.resize(childCounts_.size()); childProbs_.resize(childCounts_.size());
for (int f = 0; f < nFeatures_; f++) { for (int f = 0; f < nFeatures_; f++) {
if (f == superParent_) continue; if (f == superParent_)
continue;
int offset = childOffsets_[f]; int offset = childOffsets_[f];
int childCard = states_[f]; int childCard = states_[f];
@@ -209,14 +219,14 @@ namespace bayesnet {
for (int spVal = 0; spVal < spCard; spVal++) { for (int spVal = 0; spVal < spCard; spVal++) {
for (int childVal = 0; childVal < childCard; childVal++) { for (int childVal = 0; childVal < childCard; childVal++) {
for (int c = 0; c < statesClass_; c++) { for (int c = 0; c < statesClass_; c++) {
int idx = offset + spVal * (childCard * statesClass_) int idx = offset + spVal * (childCard * statesClass_) +
+ childVal * statesClass_ childVal * statesClass_ + c;
+ c;
double num = childCounts_[idx] + alpha_; double num = childCounts_[idx] + alpha_;
// denominator = spFeatureCounts_[spVal * statesClass_ + c] + alpha_ * (#states of child) // denominator = spFeatureCounts_[spVal * statesClass_ + c] + alpha_ *
double denom = spFeatureCounts_[spVal * statesClass_ + c] // (#states of child)
+ alpha_ * childCard; double denom =
spFeatureCounts_[spVal * statesClass_ + c] + alpha_ * childCard;
childProbs_[idx] = (denom <= 0.0 ? 0.0 : num / denom); childProbs_[idx] = (denom <= 0.0 ? 0.0 : num / denom);
} }
} }
@@ -248,11 +258,13 @@ namespace bayesnet {
// Multiply by each child’s probability p(x_child | c, x_sp) // Multiply by each child’s probability p(x_child | c, x_sp)
for (int feature = 0; feature < nFeatures_; feature++) { for (int feature = 0; feature < nFeatures_; feature++) {
if (feature == superParent_) continue; // skip sp if (feature == superParent_)
continue; // skip sp
int sf = instance[feature]; int sf = instance[feature];
int offset = childOffsets_[feature]; int offset = childOffsets_[feature];
int childCard = states_[feature]; // not used directly, but for clarity int childCard = states_[feature]; // not used directly, but for clarity
// Index into childProbs_ = offset + spVal*(childCard*statesClass_) + childVal*statesClass_ + c // Index into childProbs_ = offset + spVal*(childCard*statesClass_) +
// childVal*statesClass_ + c
int base = offset + spVal * (childCard * statesClass_) + sf * statesClass_; int base = offset + spVal * (childCard * statesClass_) + sf * statesClass_;
for (int c = 0; c < statesClass_; c++) { for (int c = 0; c < statesClass_; c++) {
probs[c] *= childProbs_[base + c]; probs[c] *= childProbs_[base + c];
@@ -267,12 +279,16 @@ namespace bayesnet {
{ {
int test_size = test_data[0].size(); int test_size = test_data[0].size();
int sample_size = test_data.size(); int sample_size = test_data.size();
auto probabilities = std::vector<std::vector<double>>(test_size, std::vector<double>(statesClass_)); auto probabilities = std::vector<std::vector<double>>(
test_size, std::vector<double>(statesClass_));
int chunk_size = std::min(150, int(test_size / semaphore_.getMaxCount()) + 1); int chunk_size = std::min(150, int(test_size / semaphore_.getMaxCount()) + 1);
std::vector<std::thread> threads; std::vector<std::thread> threads;
auto worker = [&](const std::vector<std::vector<int>>& samples, int begin, int chunk, int sample_size, std::vector<std::vector<double>>& predictions) { auto worker = [&](const std::vector<std::vector<int>>& samples, int begin,
std::string threadName = "(V)PWorker-" + std::to_string(begin) + "-" + std::to_string(chunk); int chunk, int sample_size,
std::vector<std::vector<double>>& predictions) {
std::string threadName =
"(V)PWorker-" + std::to_string(begin) + "-" + std::to_string(chunk);
#if defined(__linux__) #if defined(__linux__)
pthread_setname_np(pthread_self(), threadName.c_str()); pthread_setname_np(pthread_self(), threadName.c_str());
#else #else
@@ -305,7 +321,9 @@ namespace bayesnet {
void XSpode::normalize(std::vector<double>& v) const void XSpode::normalize(std::vector<double>& v) const
{ {
double sum = 0.0; double sum = 0.0;
for (auto val : v) { sum += val; } for (auto val : v) {
sum += val;
}
if (sum <= 0.0) { if (sum <= 0.0) {
return; return;
} }
@@ -320,37 +338,47 @@ namespace bayesnet {
std::string XSpode::to_string() const std::string XSpode::to_string() const
{ {
std::ostringstream oss; std::ostringstream oss;
oss << "---- SPODE Model ----" << std::endl oss << "----- XSpode Model -----" << std::endl
<< "nFeatures_ = " << nFeatures_ << std::endl << "nFeatures_ = " << nFeatures_ << std::endl
<< "superParent_ = " << superParent_ << std::endl << "superParent_ = " << superParent_ << std::endl
<< "statesClass_ = " << statesClass_ << std::endl << "statesClass_ = " << statesClass_ << std::endl
<< std::endl; << std::endl;
oss << "States: ["; oss << "States: [";
for (int s : states_) oss << s << " "; for (int s : states_)
oss << s << " ";
oss << "]" << std::endl; oss << "]" << std::endl;
oss << "classCounts_: ["; oss << "classCounts_: [";
for (double c : classCounts_) oss << c << " "; for (double c : classCounts_)
oss << c << " ";
oss << "]" << std::endl; oss << "]" << std::endl;
oss << "classPriors_: ["; oss << "classPriors_: [";
for (double c : classPriors_) oss << c << " "; for (double c : classPriors_)
oss << c << " ";
oss << "]" << std::endl; oss << "]" << std::endl;
oss << "spFeatureCounts_: size = " << spFeatureCounts_.size() << std::endl << "["; oss << "spFeatureCounts_: size = " << spFeatureCounts_.size() << std::endl
for (double c : spFeatureCounts_) oss << c << " "; << "[";
for (double c : spFeatureCounts_)
oss << c << " ";
oss << "]" << std::endl; oss << "]" << std::endl;
oss << "spFeatureProbs_: size = " << spFeatureProbs_.size() << std::endl << "["; oss << "spFeatureProbs_: size = " << spFeatureProbs_.size() << std::endl
for (double c : spFeatureProbs_) oss << c << " "; << "[";
for (double c : spFeatureProbs_)
oss << c << " ";
oss << "]" << std::endl; oss << "]" << std::endl;
oss << "childCounts_: size = " << childCounts_.size() << std::endl << "["; oss << "childCounts_: size = " << childCounts_.size() << std::endl << "[";
for (double cc : childCounts_) oss << cc << " "; for (double cc : childCounts_)
oss << cc << " ";
oss << "]" << std::endl; oss << "]" << std::endl;
for (double cp : childProbs_) oss << cp << " "; for (double cp : childProbs_)
oss << cp << " ";
oss << "]" << std::endl; oss << "]" << std::endl;
oss << "childOffsets_: ["; oss << "childOffsets_: [";
for (int co : childOffsets_) oss << co << " "; for (int co : childOffsets_)
oss << co << " ";
oss << "]" << std::endl; oss << "]" << std::endl;
oss << "---------------------" << std::endl; oss << std::string(40,'-') << std::endl;
return oss.str(); return oss.str();
} }
int XSpode::getNumberOfNodes() const { return nFeatures_ + 1; } int XSpode::getNumberOfNodes() const { return nFeatures_ + 1; }
@@ -372,8 +400,7 @@ namespace bayesnet {
int XSpode::predict(const std::vector<int>& instance) const int XSpode::predict(const std::vector<int>& instance) const
{ {
auto p = predict_proba(instance); auto p = predict_proba(instance);
return static_cast<int>(std::distance(p.begin(), return static_cast<int>(std::distance(p.begin(), std::max_element(p.begin(), p.end())));
std::max_element(p.begin(), p.end())));
} }
std::vector<int> XSpode::predict(std::vector<std::vector<int>>& test_data) std::vector<int> XSpode::predict(std::vector<std::vector<int>>& test_data)
{ {
@@ -381,9 +408,10 @@ namespace bayesnet {
std::vector<int> predictions(probabilities.size(), 0); std::vector<int> predictions(probabilities.size(), 0);
for (size_t i = 0; i < probabilities.size(); i++) { for (size_t i = 0; i < probabilities.size(); i++) {
predictions[i] = std::distance(probabilities[i].begin(), std::max_element(probabilities[i].begin(), probabilities[i].end())); predictions[i] = std::distance(
probabilities[i].begin(),
std::max_element(probabilities[i].begin(), probabilities[i].end()));
} }
return predictions; return predictions;
} }
torch::Tensor XSpode::predict(torch::Tensor& X) torch::Tensor XSpode::predict(torch::Tensor& X)
@@ -396,9 +424,11 @@ namespace bayesnet {
{ {
auto X_ = TensorUtils::to_matrix(X); auto X_ = TensorUtils::to_matrix(X);
auto result_v = predict_proba(X_); auto result_v = predict_proba(X_);
torch::Tensor result; int n_samples = X.size(1);
torch::Tensor result =
torch::zeros({ n_samples, statesClass_ }, torch::kDouble);
for (int i = 0; i < result_v.size(); ++i) { for (int i = 0; i < result_v.size(); ++i) {
result.index_put_({ i, "..." }, torch::tensor(result_v[i], torch::kDouble)); result.index_put_({ i, "..." }, torch::tensor(result_v[i]));
} }
return result; return result;
} }
@@ -418,5 +448,4 @@ namespace bayesnet {
} }
return (double)correct / y_pred.size(); return (double)correct / y_pred.size();
} }
} } // namespace bayesnet

2
bayesnet/classifiers/XSPODE.h
View File

@@ -29,7 +29,7 @@ namespace bayesnet {
int getClassNumStates() const override; int getClassNumStates() const override;
std::vector<int>& getStates(); std::vector<int>& getStates();
std::vector<std::string> graph(const std::string& title) const override { return std::vector<std::string>({ title }); } std::vector<std::string> graph(const std::string& title) const override { return std::vector<std::string>({ title }); }
void fit(std::vector<std::vector<int>>& X, std::vector<int>& y, torch::Tensor& weights_, const Smoothing_t smoothing); void fit(torch::Tensor& X, torch::Tensor& y, torch::Tensor& weights_, const Smoothing_t smoothing);
void setHyperparameters(const nlohmann::json& hyperparameters_) override; void setHyperparameters(const nlohmann::json& hyperparameters_) override;
// //

3
bayesnet/ensembles/Ensemble.h
View File

@@ -41,6 +41,7 @@ namespace bayesnet {
return output; return output;
} }
protected: protected:
void trainModel(const torch::Tensor& weights, const Smoothing_t smoothing) override;
torch::Tensor predict_average_voting(torch::Tensor& X); torch::Tensor predict_average_voting(torch::Tensor& X);
std::vector<std::vector<double>> predict_average_voting(std::vector<std::vector<int>>& X); std::vector<std::vector<double>> predict_average_voting(std::vector<std::vector<int>>& X);
torch::Tensor predict_average_proba(torch::Tensor& X); torch::Tensor predict_average_proba(torch::Tensor& X);
@@ -48,10 +49,10 @@ namespace bayesnet {
torch::Tensor compute_arg_max(torch::Tensor& X); torch::Tensor compute_arg_max(torch::Tensor& X);
std::vector<int> compute_arg_max(std::vector<std::vector<double>>& X); std::vector<int> compute_arg_max(std::vector<std::vector<double>>& X);
torch::Tensor voting(torch::Tensor& votes); torch::Tensor voting(torch::Tensor& votes);
// Attributes
unsigned n_models; unsigned n_models;
std::vector<std::unique_ptr<Classifier>> models; std::vector<std::unique_ptr<Classifier>> models;
std::vector<double> significanceModels; std::vector<double> significanceModels;
void trainModel(const torch::Tensor& weights, const Smoothing_t smoothing) override;
bool predict_voting; bool predict_voting;
}; };
} }

10
bayesnet/ensembles/XBAODE.cc
View File

@@ -36,7 +36,8 @@ namespace bayesnet {
std::vector<int> featuresSelected = featureSelection(weights_); std::vector<int> featuresSelected = featureSelection(weights_);
for (const int& feature : featuresSelected) { for (const int& feature : featuresSelected) {
std::unique_ptr<Classifier> model = std::make_unique<XSpode>(feature); std::unique_ptr<Classifier> model = std::make_unique<XSpode>(feature);
model->fit(dataset, features, className, states, weights_, smoothing); // model->fit(dataset, features, className, states, weights_, smoothing);
dynamic_cast<XSpode*>(model.get())->fit(X_train, y_train, weights_, smoothing);
add_model(std::move(model), 1.0); add_model(std::move(model), 1.0);
} }
notes.push_back("Used features in initialization: " + std::to_string(featuresSelected.size()) + " of " + std::to_string(features.size()) + " with " + select_features_algorithm); notes.push_back("Used features in initialization: " + std::to_string(featuresSelected.size()) + " of " + std::to_string(features.size()) + " with " + select_features_algorithm);
@@ -57,6 +58,7 @@ namespace bayesnet {
n_models = 0; n_models = 0;
if (selectFeatures) { if (selectFeatures) {
featuresUsed = initializeModels(smoothing); featuresUsed = initializeModels(smoothing);
std::cout << "features used: " << featuresUsed.size() << std::endl;
auto ypred = predict(X_train_); auto ypred = predict(X_train_);
auto ypred_t = torch::tensor(ypred); auto ypred_t = torch::tensor(ypred);
std::tie(weights_, alpha_t, finished) = update_weights(y_train, ypred_t, weights_); std::tie(weights_, alpha_t, finished) = update_weights(y_train, ypred_t, weights_);
@@ -103,7 +105,11 @@ namespace bayesnet {
featureSelection.erase(featureSelection.begin()); featureSelection.erase(featureSelection.begin());
std::unique_ptr<Classifier> model; std::unique_ptr<Classifier> model;
model = std::make_unique<XSpode>(feature); model = std::make_unique<XSpode>(feature);
dynamic_cast<XSpode*>(model.get())->fit(X_train_, y_train_, weights_, smoothing); // using exclusive XSpode fit method dynamic_cast<XSpode*>(model.get())->fit(X_train, y_train, weights_, smoothing); // using exclusive XSpode fit method
// DEBUG
std::cout << "Model fitted." << std::endl;
std::cout << dynamic_cast<XSpode*>(model.get())->to_string() << std::endl;
// DEBUG
std::vector<int> ypred; std::vector<int> ypred;
if (alpha_block) { if (alpha_block) {
// //

5
sample/sample.cc
View File

@@ -6,7 +6,7 @@
#include <ArffFiles.hpp> #include <ArffFiles.hpp>
#include <CPPFImdlp.h> #include <CPPFImdlp.h>
#include <bayesnet/ensembles/BoostAODE.h> #include <bayesnet/ensembles/XBAODE.h>
std::vector<mdlp::labels_t> discretizeDataset(std::vector<mdlp::samples_t>& X, mdlp::labels_t& y) std::vector<mdlp::labels_t> discretizeDataset(std::vector<mdlp::samples_t>& X, mdlp::labels_t& y)
{ {
@@ -57,7 +57,7 @@ int main(int argc, char* argv[])
std::vector<std::string> features; std::vector<std::string> features;
std::string className; std::string className;
map<std::string, std::vector<int>> states; map<std::string, std::vector<int>> states;
auto clf = bayesnet::BoostAODE(false); // false for not using voting in predict auto clf = bayesnet::XBAODE(); // false for not using voting in predict
std::cout << "Library version: " << clf.getVersion() << std::endl; std::cout << "Library version: " << clf.getVersion() << std::endl;
tie(X, y, features, className, states) = loadDataset(file_name, true); tie(X, y, features, className, states) = loadDataset(file_name, true);
torch::Tensor weights = torch::full({ X.size(1) }, 15, torch::kDouble); torch::Tensor weights = torch::full({ X.size(1) }, 15, torch::kDouble);
@@ -73,7 +73,6 @@ int main(int argc, char* argv[])
oss << "y dimensions: " << y.sizes(); oss << "y dimensions: " << y.sizes();
throw std::runtime_error(oss.str()); throw std::runtime_error(oss.str());
} }
//Classifier& fit(torch::Tensor& dataset, const std::vector<std::string>& features, const std::string& className, std::map<std::string, std::vector<int>>& states, const torch::Tensor& weights, const Smoothing_t smoothing) override;
clf.fit(dataset, features, className, states, weights, bayesnet::Smoothing_t::LAPLACE); clf.fit(dataset, features, className, states, weights, bayesnet::Smoothing_t::LAPLACE);
auto score = clf.score(X, y); auto score = clf.score(X, y);
std::cout << "File: " << file_name << " Model: BoostAODE score: " << score << std::endl; std::cout << "File: " << file_name << " Model: BoostAODE score: " << score << std::endl;

2
tests/CMakeLists.txt
View File

@@ -10,7 +10,7 @@ if(ENABLE_TESTING)
) )
file(GLOB_RECURSE BayesNet_SOURCES "${BayesNet_SOURCE_DIR}/bayesnet/*.cc") file(GLOB_RECURSE BayesNet_SOURCES "${BayesNet_SOURCE_DIR}/bayesnet/*.cc")
add_executable(TestBayesNet TestBayesNetwork.cc TestBayesNode.cc TestBayesClassifier.cc add_executable(TestBayesNet TestBayesNetwork.cc TestBayesNode.cc TestBayesClassifier.cc
TestBayesModels.cc TestBayesMetrics.cc TestFeatureSelection.cc TestBoostAODE.cc TestA2DE.cc TestWA2DE.cc TestBayesModels.cc TestBayesMetrics.cc TestFeatureSelection.cc TestBoostAODE.cc TestXBAODE.cc TestA2DE.cc TestWA2DE.cc
TestUtils.cc TestBayesEnsemble.cc TestModulesVersions.cc TestBoostA2DE.cc TestMST.cc ${BayesNet_SOURCES}) TestUtils.cc TestBayesEnsemble.cc TestModulesVersions.cc TestBoostA2DE.cc TestMST.cc ${BayesNet_SOURCES})
target_link_libraries(TestBayesNet PUBLIC "${TORCH_LIBRARIES}" fimdlp PRIVATE Catch2::Catch2WithMain) target_link_libraries(TestBayesNet PUBLIC "${TORCH_LIBRARIES}" fimdlp PRIVATE Catch2::Catch2WithMain)
add_test(NAME BayesNetworkTest COMMAND TestBayesNet) add_test(NAME BayesNetworkTest COMMAND TestBayesNet)

234
tests/TestBoostXBAODE.cc
View File

@@ -1,234 +0,0 @@
// ***************************************************************
// SPDX-FileCopyrightText: Copyright 2024 Ricardo Montañana Gómez
// SPDX-FileType: SOURCE
// SPDX-License-Identifier: MIT
// ***************************************************************
#include <type_traits>
#include <catch2/catch_test_macros.hpp>
#include <catch2/catch_approx.hpp>
#include <catch2/generators/catch_generators.hpp>
#include <catch2/matchers/catch_matchers.hpp>
#include "bayesnet/ensembles/XBAODE.h"
#include "TestUtils.h"
TEST_CASE("Feature_select CFS", "[XBAODE]")
{
auto raw = RawDatasets("glass", true);
auto clf = bayesnet::XBAODE();
clf.setHyperparameters({ {"select_features", "CFS"} });
clf.fit(raw.Xv, raw.yv, raw.features, raw.className, raw.states, raw.smoothing);
REQUIRE(clf.getNumberOfNodes() == 90);
REQUIRE(clf.getNumberOfEdges() == 153);
REQUIRE(clf.getNotes().size() == 2);
REQUIRE(clf.getNotes()[0] == "Used features in initialization: 6 of 9 with CFS");
REQUIRE(clf.getNotes()[1] == "Number of models: 9");
}
TEST_CASE("Feature_select IWSS", "[XBAODE]")
{
auto raw = RawDatasets("glass", true);
auto clf = bayesnet::XBAODE();
clf.setHyperparameters({ {"select_features", "IWSS"}, {"threshold", 0.5 } });
clf.fit(raw.Xv, raw.yv, raw.features, raw.className, raw.states, raw.smoothing);
REQUIRE(clf.getNumberOfNodes() == 90);
REQUIRE(clf.getNumberOfEdges() == 153);
REQUIRE(clf.getNotes().size() == 2);
REQUIRE(clf.getNotes()[0] == "Used features in initialization: 4 of 9 with IWSS");
REQUIRE(clf.getNotes()[1] == "Number of models: 9");
}
TEST_CASE("Feature_select FCBF", "[XBAODE]")
{
auto raw = RawDatasets("glass", true);
auto clf = bayesnet::XBAODE();
clf.setHyperparameters({ {"select_features", "FCBF"}, {"threshold", 1e-7 } });
clf.fit(raw.Xv, raw.yv, raw.features, raw.className, raw.states, raw.smoothing);
REQUIRE(clf.getNumberOfNodes() == 90);
REQUIRE(clf.getNumberOfEdges() == 153);
REQUIRE(clf.getNotes().size() == 2);
REQUIRE(clf.getNotes()[0] == "Used features in initialization: 4 of 9 with FCBF");
REQUIRE(clf.getNotes()[1] == "Number of models: 9");
}
TEST_CASE("Test used features in train note and score", "[XBAODE]")
{
auto raw = RawDatasets("diabetes", true);
auto clf = bayesnet::XBAODE(true);
clf.setHyperparameters({
{"order", "asc"},
{"convergence", true},
{"select_features","CFS"},
});
clf.fit(raw.Xv, raw.yv, raw.features, raw.className, raw.states, raw.smoothing);
REQUIRE(clf.getNumberOfNodes() == 72);
REQUIRE(clf.getNumberOfEdges() == 120);
REQUIRE(clf.getNotes().size() == 2);
REQUIRE(clf.getNotes()[0] == "Used features in initialization: 6 of 8 with CFS");
REQUIRE(clf.getNotes()[1] == "Number of models: 8");
auto score = clf.score(raw.Xv, raw.yv);
auto scoret = clf.score(raw.Xt, raw.yt);
REQUIRE(score == Catch::Approx(0.809895813).epsilon(raw.epsilon));
REQUIRE(scoret == Catch::Approx(0.809895813).epsilon(raw.epsilon));
}
TEST_CASE("Voting vs proba", "[XBAODE]")
{
auto raw = RawDatasets("iris", true);
auto clf = bayesnet::XBAODE(false);
clf.fit(raw.Xv, raw.yv, raw.features, raw.className, raw.states, raw.smoothing);
auto score_proba = clf.score(raw.Xv, raw.yv);
auto pred_proba = clf.predict_proba(raw.Xv);
clf.setHyperparameters({
{"predict_voting",true},
});
auto score_voting = clf.score(raw.Xv, raw.yv);
auto pred_voting = clf.predict_proba(raw.Xv);
REQUIRE(score_proba == Catch::Approx(0.97333).epsilon(raw.epsilon));
REQUIRE(score_voting == Catch::Approx(0.98).epsilon(raw.epsilon));
REQUIRE(pred_voting[83][2] == Catch::Approx(1.0).epsilon(raw.epsilon));
REQUIRE(pred_proba[83][2] == Catch::Approx(0.86121525).epsilon(raw.epsilon));
REQUIRE(clf.dump_cpt() == "");
REQUIRE(clf.topological_order() == std::vector<std::string>());
}
TEST_CASE("Order asc, desc & random", "[XBAODE]")
{
auto raw = RawDatasets("glass", true);
std::map<std::string, double> scores{
{"asc", 0.83645f }, { "desc", 0.84579f }, { "rand", 0.84112 }
};
for (const std::string& order : { "asc", "desc", "rand" }) {
auto clf = bayesnet::XBAODE();
clf.setHyperparameters({
{"order", order},
{"bisection", false},
{"maxTolerance", 1},
{"convergence", false},
});
clf.fit(raw.Xv, raw.yv, raw.features, raw.className, raw.states, raw.smoothing);
auto score = clf.score(raw.Xv, raw.yv);
auto scoret = clf.score(raw.Xt, raw.yt);
INFO("XBAODE order: " << order);
REQUIRE(score == Catch::Approx(scores[order]).epsilon(raw.epsilon));
REQUIRE(scoret == Catch::Approx(scores[order]).epsilon(raw.epsilon));
}
}
TEST_CASE("Oddities", "[XBAODE]")
{
auto clf = bayesnet::XBAODE();
auto raw = RawDatasets("iris", true);
auto bad_hyper = nlohmann::json{
{ { "order", "duck" } },
{ { "select_features", "duck" } },
{ { "maxTolerance", 0 } },
{ { "maxTolerance", 7 } },
};
for (const auto& hyper : bad_hyper.items()) {
INFO("XBAODE hyper: " << hyper.value().dump());
REQUIRE_THROWS_AS(clf.setHyperparameters(hyper.value()), std::invalid_argument);
}
REQUIRE_THROWS_AS(clf.setHyperparameters({ {"maxTolerance", 0 } }), std::invalid_argument);
auto bad_hyper_fit = nlohmann::json{
{ { "select_features","IWSS" }, { "threshold", -0.01 } },
{ { "select_features","IWSS" }, { "threshold", 0.51 } },
{ { "select_features","FCBF" }, { "threshold", 1e-8 } },
{ { "select_features","FCBF" }, { "threshold", 1.01 } },
};
for (const auto& hyper : bad_hyper_fit.items()) {
INFO("XBAODE hyper: " << hyper.value().dump());
clf.setHyperparameters(hyper.value());
REQUIRE_THROWS_AS(clf.fit(raw.Xv, raw.yv, raw.features, raw.className, raw.states, raw.smoothing), std::invalid_argument);
}
auto bad_hyper_fit2 = nlohmann::json{
{ { "alpha_block", true }, { "block_update", true } },
{ { "bisection", false }, { "block_update", true } },
};
for (const auto& hyper : bad_hyper_fit2.items()) {
INFO("XBAODE hyper: " << hyper.value().dump());
REQUIRE_THROWS_AS(clf.setHyperparameters(hyper.value()), std::invalid_argument);
}
}
TEST_CASE("Bisection Best", "[XBAODE]")
{
auto clf = bayesnet::XBAODE();
auto raw = RawDatasets("kdd_JapaneseVowels", true, 1200, true, false);
clf.setHyperparameters({
{"bisection", true},
{"maxTolerance", 3},
{"convergence", true},
{"convergence_best", false},
});
clf.fit(raw.X_train, raw.y_train, raw.features, raw.className, raw.states, raw.smoothing);
REQUIRE(clf.getNumberOfNodes() == 210);
REQUIRE(clf.getNumberOfEdges() == 378);
REQUIRE(clf.getNotes().size() == 1);
REQUIRE(clf.getNotes().at(0) == "Number of models: 14");
auto score = clf.score(raw.X_test, raw.y_test);
auto scoret = clf.score(raw.X_test, raw.y_test);
REQUIRE(score == Catch::Approx(0.991666675f).epsilon(raw.epsilon));
REQUIRE(scoret == Catch::Approx(0.991666675f).epsilon(raw.epsilon));
}
TEST_CASE("Bisection Best vs Last", "[XBAODE]")
{
auto raw = RawDatasets("kdd_JapaneseVowels", true, 1500, true, false);
auto clf = bayesnet::XBAODE(true);
auto hyperparameters = nlohmann::json{
{"bisection", true},
{"maxTolerance", 3},
{"convergence", true},
{"convergence_best", true},
};
clf.setHyperparameters(hyperparameters);
clf.fit(raw.X_train, raw.y_train, raw.features, raw.className, raw.states, raw.smoothing);
auto score_best = clf.score(raw.X_test, raw.y_test);
REQUIRE(score_best == Catch::Approx(0.980000019f).epsilon(raw.epsilon));
// Now we will set the hyperparameter to use the last accuracy
hyperparameters["convergence_best"] = false;
clf.setHyperparameters(hyperparameters);
clf.fit(raw.X_train, raw.y_train, raw.features, raw.className, raw.states, raw.smoothing);
auto score_last = clf.score(raw.X_test, raw.y_test);
REQUIRE(score_last == Catch::Approx(0.976666689f).epsilon(raw.epsilon));
}
TEST_CASE("Block Update", "[XBAODE]")
{
auto clf = bayesnet::XBAODE();
auto raw = RawDatasets("mfeat-factors", true, 500);
clf.setHyperparameters({
{"bisection", true},
{"block_update", true},
{"maxTolerance", 3},
{"convergence", true},
});
clf.fit(raw.X_train, raw.y_train, raw.features, raw.className, raw.states, raw.smoothing);
REQUIRE(clf.getNumberOfNodes() == 868);
REQUIRE(clf.getNumberOfEdges() == 1724);
REQUIRE(clf.getNotes().size() == 3);
REQUIRE(clf.getNotes()[0] == "Convergence threshold reached & 15 models eliminated");
REQUIRE(clf.getNotes()[1] == "Used features in train: 19 of 216");
REQUIRE(clf.getNotes()[2] == "Number of models: 4");
auto score = clf.score(raw.X_test, raw.y_test);
auto scoret = clf.score(raw.X_test, raw.y_test);
REQUIRE(score == Catch::Approx(0.99f).epsilon(raw.epsilon));
REQUIRE(scoret == Catch::Approx(0.99f).epsilon(raw.epsilon));
//
// std::cout << "Number of nodes " << clf.getNumberOfNodes() << std::endl;
// std::cout << "Number of edges " << clf.getNumberOfEdges() << std::endl;
// std::cout << "Notes size " << clf.getNotes().size() << std::endl;
// for (auto note : clf.getNotes()) {
// std::cout << note << std::endl;
// }
// std::cout << "Score " << score << std::endl;
}
TEST_CASE("Alphablock", "[XBAODE]")
{
auto clf_alpha = bayesnet::XBAODE();
auto clf_no_alpha = bayesnet::XBAODE();
auto raw = RawDatasets("diabetes", true);
clf_alpha.setHyperparameters({
{"alpha_block", true},
});
clf_alpha.fit(raw.X_train, raw.y_train, raw.features, raw.className, raw.states, raw.smoothing);
clf_no_alpha.fit(raw.X_train, raw.y_train, raw.features, raw.className, raw.states, raw.smoothing);
auto score_alpha = clf_alpha.score(raw.X_test, raw.y_test);
auto score_no_alpha = clf_no_alpha.score(raw.X_test, raw.y_test);
REQUIRE(score_alpha == Catch::Approx(0.720779f).epsilon(raw.epsilon));
REQUIRE(score_no_alpha == Catch::Approx(0.733766f).epsilon(raw.epsilon));
}

243
tests/TestXBAODE.cc Normal file
View File

@@ -0,0 +1,243 @@
// ***************************************************************
// SPDX-FileCopyrightText: Copyright 2024 Ricardo Montañana Gómez
// SPDX-FileType: SOURCE
// SPDX-License-Identifier: MIT
// ***************************************************************
#include <type_traits>
#include <catch2/catch_test_macros.hpp>
#include <catch2/catch_approx.hpp>
#include <catch2/generators/catch_generators.hpp>
#include <catch2/matchers/catch_matchers.hpp>
#include "bayesnet/ensembles/XBAODE.h"
#include "TestUtils.h"
TEST_CASE("Normal test", "[XBAODE]")
{
auto raw = RawDatasets("iris", true);
auto clf = bayesnet::XBAODE();
clf.fit(raw.Xv, raw.yv, raw.features, raw.className, raw.states, raw.smoothing);
REQUIRE(clf.getNumberOfNodes() == 20);
REQUIRE(clf.getNumberOfEdges() == 112);
REQUIRE(clf.getNotes().size() == 1);
}
//TEST_CASE("Feature_select CFS", "[XBAODE]")
//{
// auto raw = RawDatasets("glass", true);
// auto clf = bayesnet::XBAODE();
// clf.setHyperparameters({ {"select_features", "CFS"} });
// clf.fit(raw.Xv, raw.yv, raw.features, raw.className, raw.states, raw.smoothing);
// REQUIRE(clf.getNumberOfNodes() == 97);
// REQUIRE(clf.getNumberOfEdges() == 153);
// REQUIRE(clf.getNotes().size() == 2);
// REQUIRE(clf.getNotes()[0] == "Used features in initialization: 6 of 9 with CFS");
// REQUIRE(clf.getNotes()[1] == "Number of models: 9");
//}
// TEST_CASE("Feature_select IWSS", "[XBAODE]")
// {
// auto raw = RawDatasets("glass", true);
// auto clf = bayesnet::XBAODE();
// clf.setHyperparameters({ {"select_features", "IWSS"}, {"threshold", 0.5 } });
// clf.fit(raw.Xv, raw.yv, raw.features, raw.className, raw.states, raw.smoothing);
// REQUIRE(clf.getNumberOfNodes() == 90);
// REQUIRE(clf.getNumberOfEdges() == 153);
// REQUIRE(clf.getNotes().size() == 2);
// REQUIRE(clf.getNotes()[0] == "Used features in initialization: 4 of 9 with IWSS");
// REQUIRE(clf.getNotes()[1] == "Number of models: 9");
// }
// TEST_CASE("Feature_select FCBF", "[XBAODE]")
// {
// auto raw = RawDatasets("glass", true);
// auto clf = bayesnet::XBAODE();
// clf.setHyperparameters({ {"select_features", "FCBF"}, {"threshold", 1e-7 } });
// clf.fit(raw.Xv, raw.yv, raw.features, raw.className, raw.states, raw.smoothing);
// REQUIRE(clf.getNumberOfNodes() == 90);
// REQUIRE(clf.getNumberOfEdges() == 153);
// REQUIRE(clf.getNotes().size() == 2);
// REQUIRE(clf.getNotes()[0] == "Used features in initialization: 4 of 9 with FCBF");
// REQUIRE(clf.getNotes()[1] == "Number of models: 9");
// }
// TEST_CASE("Test used features in train note and score", "[XBAODE]")
// {
// auto raw = RawDatasets("diabetes", true);
// auto clf = bayesnet::XBAODE(true);
// clf.setHyperparameters({
// {"order", "asc"},
// {"convergence", true},
// {"select_features","CFS"},
// });
// clf.fit(raw.Xv, raw.yv, raw.features, raw.className, raw.states, raw.smoothing);
// REQUIRE(clf.getNumberOfNodes() == 72);
// REQUIRE(clf.getNumberOfEdges() == 120);
// REQUIRE(clf.getNotes().size() == 2);
// REQUIRE(clf.getNotes()[0] == "Used features in initialization: 6 of 8 with CFS");
// REQUIRE(clf.getNotes()[1] == "Number of models: 8");
// auto score = clf.score(raw.Xv, raw.yv);
// auto scoret = clf.score(raw.Xt, raw.yt);
// REQUIRE(score == Catch::Approx(0.809895813).epsilon(raw.epsilon));
// REQUIRE(scoret == Catch::Approx(0.809895813).epsilon(raw.epsilon));
// }
// TEST_CASE("Voting vs proba", "[XBAODE]")
// {
// auto raw = RawDatasets("iris", true);
// auto clf = bayesnet::XBAODE(false);
// clf.fit(raw.Xv, raw.yv, raw.features, raw.className, raw.states, raw.smoothing);
// auto score_proba = clf.score(raw.Xv, raw.yv);
// auto pred_proba = clf.predict_proba(raw.Xv);
// clf.setHyperparameters({
// {"predict_voting",true},
// });
// auto score_voting = clf.score(raw.Xv, raw.yv);
// auto pred_voting = clf.predict_proba(raw.Xv);
// REQUIRE(score_proba == Catch::Approx(0.97333).epsilon(raw.epsilon));
// REQUIRE(score_voting == Catch::Approx(0.98).epsilon(raw.epsilon));
// REQUIRE(pred_voting[83][2] == Catch::Approx(1.0).epsilon(raw.epsilon));
// REQUIRE(pred_proba[83][2] == Catch::Approx(0.86121525).epsilon(raw.epsilon));
// REQUIRE(clf.dump_cpt() == "");
// REQUIRE(clf.topological_order() == std::vector<std::string>());
// }
// TEST_CASE("Order asc, desc & random", "[XBAODE]")
// {
// auto raw = RawDatasets("glass", true);
// std::map<std::string, double> scores{
// {"asc", 0.83645f }, { "desc", 0.84579f }, { "rand", 0.84112 }
// };
// for (const std::string& order : { "asc", "desc", "rand" }) {
// auto clf = bayesnet::XBAODE();
// clf.setHyperparameters({
// {"order", order},
// {"bisection", false},
// {"maxTolerance", 1},
// {"convergence", false},
// });
// clf.fit(raw.Xv, raw.yv, raw.features, raw.className, raw.states, raw.smoothing);
// auto score = clf.score(raw.Xv, raw.yv);
// auto scoret = clf.score(raw.Xt, raw.yt);
// INFO("XBAODE order: " << order);
// REQUIRE(score == Catch::Approx(scores[order]).epsilon(raw.epsilon));
// REQUIRE(scoret == Catch::Approx(scores[order]).epsilon(raw.epsilon));
// }
// }
// TEST_CASE("Oddities", "[XBAODE]")
// {
// auto clf = bayesnet::XBAODE();
// auto raw = RawDatasets("iris", true);
// auto bad_hyper = nlohmann::json{
// { { "order", "duck" } },
// { { "select_features", "duck" } },
// { { "maxTolerance", 0 } },
// { { "maxTolerance", 7 } },
// };
// for (const auto& hyper : bad_hyper.items()) {
// INFO("XBAODE hyper: " << hyper.value().dump());
// REQUIRE_THROWS_AS(clf.setHyperparameters(hyper.value()), std::invalid_argument);
// }
// REQUIRE_THROWS_AS(clf.setHyperparameters({ {"maxTolerance", 0 } }), std::invalid_argument);
// auto bad_hyper_fit = nlohmann::json{
// { { "select_features","IWSS" }, { "threshold", -0.01 } },
// { { "select_features","IWSS" }, { "threshold", 0.51 } },
// { { "select_features","FCBF" }, { "threshold", 1e-8 } },
// { { "select_features","FCBF" }, { "threshold", 1.01 } },
// };
// for (const auto& hyper : bad_hyper_fit.items()) {
// INFO("XBAODE hyper: " << hyper.value().dump());
// clf.setHyperparameters(hyper.value());
// REQUIRE_THROWS_AS(clf.fit(raw.Xv, raw.yv, raw.features, raw.className, raw.states, raw.smoothing), std::invalid_argument);
// }
// auto bad_hyper_fit2 = nlohmann::json{
// { { "alpha_block", true }, { "block_update", true } },
// { { "bisection", false }, { "block_update", true } },
// };
// for (const auto& hyper : bad_hyper_fit2.items()) {
// INFO("XBAODE hyper: " << hyper.value().dump());
// REQUIRE_THROWS_AS(clf.setHyperparameters(hyper.value()), std::invalid_argument);
// }
// }
// TEST_CASE("Bisection Best", "[XBAODE]")
// {
// auto clf = bayesnet::XBAODE();
// auto raw = RawDatasets("kdd_JapaneseVowels", true, 1200, true, false);
// clf.setHyperparameters({
// {"bisection", true},
// {"maxTolerance", 3},
// {"convergence", true},
// {"convergence_best", false},
// });
// clf.fit(raw.X_train, raw.y_train, raw.features, raw.className, raw.states, raw.smoothing);
// REQUIRE(clf.getNumberOfNodes() == 210);
// REQUIRE(clf.getNumberOfEdges() == 378);
// REQUIRE(clf.getNotes().size() == 1);
// REQUIRE(clf.getNotes().at(0) == "Number of models: 14");
// auto score = clf.score(raw.X_test, raw.y_test);
// auto scoret = clf.score(raw.X_test, raw.y_test);
// REQUIRE(score == Catch::Approx(0.991666675f).epsilon(raw.epsilon));
// REQUIRE(scoret == Catch::Approx(0.991666675f).epsilon(raw.epsilon));
// }
// TEST_CASE("Bisection Best vs Last", "[XBAODE]")
// {
// auto raw = RawDatasets("kdd_JapaneseVowels", true, 1500, true, false);
// auto clf = bayesnet::XBAODE(true);
// auto hyperparameters = nlohmann::json{
// {"bisection", true},
// {"maxTolerance", 3},
// {"convergence", true},
// {"convergence_best", true},
// };
// clf.setHyperparameters(hyperparameters);
// clf.fit(raw.X_train, raw.y_train, raw.features, raw.className, raw.states, raw.smoothing);
// auto score_best = clf.score(raw.X_test, raw.y_test);
// REQUIRE(score_best == Catch::Approx(0.980000019f).epsilon(raw.epsilon));
// // Now we will set the hyperparameter to use the last accuracy
// hyperparameters["convergence_best"] = false;
// clf.setHyperparameters(hyperparameters);
// clf.fit(raw.X_train, raw.y_train, raw.features, raw.className, raw.states, raw.smoothing);
// auto score_last = clf.score(raw.X_test, raw.y_test);
// REQUIRE(score_last == Catch::Approx(0.976666689f).epsilon(raw.epsilon));
// }
// TEST_CASE("Block Update", "[XBAODE]")
// {
// auto clf = bayesnet::XBAODE();
// auto raw = RawDatasets("mfeat-factors", true, 500);
// clf.setHyperparameters({
// {"bisection", true},
// {"block_update", true},
// {"maxTolerance", 3},
// {"convergence", true},
// });
// clf.fit(raw.X_train, raw.y_train, raw.features, raw.className, raw.states, raw.smoothing);
// REQUIRE(clf.getNumberOfNodes() == 868);
// REQUIRE(clf.getNumberOfEdges() == 1724);
// REQUIRE(clf.getNotes().size() == 3);
// REQUIRE(clf.getNotes()[0] == "Convergence threshold reached & 15 models eliminated");
// REQUIRE(clf.getNotes()[1] == "Used features in train: 19 of 216");
// REQUIRE(clf.getNotes()[2] == "Number of models: 4");
// auto score = clf.score(raw.X_test, raw.y_test);
// auto scoret = clf.score(raw.X_test, raw.y_test);
// REQUIRE(score == Catch::Approx(0.99f).epsilon(raw.epsilon));
// REQUIRE(scoret == Catch::Approx(0.99f).epsilon(raw.epsilon));
// //
// // std::cout << "Number of nodes " << clf.getNumberOfNodes() << std::endl;
// // std::cout << "Number of edges " << clf.getNumberOfEdges() << std::endl;
// // std::cout << "Notes size " << clf.getNotes().size() << std::endl;
// // for (auto note : clf.getNotes()) {
// // std::cout << note << std::endl;
// // }
// // std::cout << "Score " << score << std::endl;
// }
// TEST_CASE("Alphablock", "[XBAODE]")
// {
// auto clf_alpha = bayesnet::XBAODE();
// auto clf_no_alpha = bayesnet::XBAODE();
// auto raw = RawDatasets("diabetes", true);
// clf_alpha.setHyperparameters({
// {"alpha_block", true},
// });
// clf_alpha.fit(raw.X_train, raw.y_train, raw.features, raw.className, raw.states, raw.smoothing);
// clf_no_alpha.fit(raw.X_train, raw.y_train, raw.features, raw.className, raw.states, raw.smoothing);
// auto score_alpha = clf_alpha.score(raw.X_test, raw.y_test);
// auto score_no_alpha = clf_no_alpha.score(raw.X_test, raw.y_test);
// REQUIRE(score_alpha == Catch::Approx(0.720779f).epsilon(raw.epsilon));
// REQUIRE(score_no_alpha == Catch::Approx(0.733766f).epsilon(raw.epsilon));
// }