Doctorado-ML/bayesclass
1
0
Fork 0
mirror of https://github.com/Doctorado-ML/bayesclass.git synced 2025-08-15 23:55:57 +00:00
Code Issues Packages Projects Releases Wiki Activity

Chcked mutual_info with sklearn

Browse Source
This commit is contained in:
Ricardo Montañana Gómez
2023-06-23 01:21:24 +02:00
parent 0094d500d4
commit 30cc744033
7 changed files with 982 additions and 386 deletions
Download Patch File Download Diff File Expand all files Collapse all files

194
bayesclass/FeatureSelect.cpp
View File

@@ -1,113 +1,121 @@
#include "FeatureSelect.h"
#include <iostream>
namespace features {
// SelectKBestWeighted::SelectKBestWeighted(samples_t& samples, labels_t& labels, weights_t& weights, int k)
// : samples(samples), labels(labels), weights(weights), k(k)
// {
// // if (samples.size() == 0 || samples[0].size() == 0)
// // throw invalid_argument("features must be a non-empty matrix");
// // if (samples.size() != labels.size())
// // throw invalid_argument("number of samples and labels must be equal");
// // if (samples.size() != weights.size())
// // throw invalid_argument("number of samples and weights must be equal");
// // if (k < 1 || k > static_cast<int>(samples[0].size()))
// // throw invalid_argument("k must be between 1 and number of features");
// numFeatures = 0;
// numClasses = 0;
// numSamples = 0;
// fitted = false;
// }
SelectKBestWeighted::SelectKBestWeighted(samples_t& samples) : samples(samples) {}
SelectKBestWeighted::SelectKBestWeighted(samples_t& samples, labels_t& labels, weights_t& weights, int k, bool nat)
: samples(samples), labels(labels), weights(weights), k(k), nat(nat)
{
if (samples.size() == 0 || samples[0].size() == 0)
throw invalid_argument("features must be a non-empty matrix");
if (samples.size() != labels.size())
throw invalid_argument("number of samples and labels must be equal");
if (samples.size() != weights.size())
throw invalid_argument("number of samples and weights must be equal");
if (k < 1 || k > static_cast<int>(samples[0].size()))
throw invalid_argument("k must be between 1 and number of features");
numFeatures = 0;
numClasses = 0;
numSamples = 0;
fitted = false;
}
void SelectKBestWeighted::SelectKBestWeighted::fit()
{
// auto labelsCopy = labels;
numFeatures = 0;//samples[0].size();
auto labelsCopy = labels;
numFeatures = samples[0].size();
numSamples = samples.size();
// sort(labelsCopy.begin(), labelsCopy.end());
// auto last = unique(labelsCopy.begin(), labelsCopy.end());
// labelsCopy.erase(last, labelsCopy.end());
// numClasses = labelsCopy.size();
// score.reserve(numFeatures);
// for (int i = 0; i < numFeatures; ++i) {
// score.push_back(MutualInformation(i));
// }
sort(labelsCopy.begin(), labelsCopy.end());
auto last = unique(labelsCopy.begin(), labelsCopy.end());
labelsCopy.erase(last, labelsCopy.end());
numClasses = labelsCopy.size();
score.reserve(numFeatures);
for (int i = 0; i < numFeatures; ++i) {
score.push_back(MutualInformation(i));
}
outputValues();
fitted = true;
}
void SelectKBestWeighted::outputValues()
{
cout << "numFeatures: " << numFeatures << endl;
// cout << "numClasses: " << numClasses << endl;
cout << "numClasses: " << numClasses << endl;
cout << "numSamples: " << numSamples << endl;
// cout << "k: " << k << endl;
// cout << "weights: ";
// for (auto item : weights)
// cout << item << ", ";
// cout << "end." << endl;
// cout << "labels: ";
// for (auto item : labels)
// cout << item << ", ";
// cout << "end." << endl;
cout << "samples: ";
for (auto item : samples) {
// for (auto item2 : item)
// cout << item2 << ", ";
// cout << "end." << endl;
cout << "k: " << k << endl;
cout << "weights: ";
for (auto item : weights)
cout << item << ", ";
cout << "end." << endl;
cout << "labels: ";
for (auto item : labels)
cout << item << ", ";
cout << "end." << endl;
cout << "samples: " << endl;
for (auto item : samples) {
for (auto item2 : item)
cout << item2 << ", ";
cout << "end." << endl;
}
cout << "end." << endl;
}
// precision_t SelectKBestWeighted::entropyLabel()
// {
// return entropy(labels);
// }
// precision_t SelectKBestWeighted::entropy(const sample_t& data)
// {
// precision_t p;
// precision_t ventropy = 0, totalWeight = 0;
// score_t counts(numClasses + 1, 0);
// for (auto i = 0; i < data.size(); ++i) {
// counts[data[i]] += weights[i];
// totalWeight += weights[i];
// }
// for (auto count : counts) {
// p = count / totalWeight;
// ventropy -= p * log2(p);
// }
// return ventropy;
// }
// // H(Y|X) = sum_{x in X} p(x) H(Y|X=x)
// precision_t SelectKBestWeighted::conditionalEntropy(const int feature)
// {
// unordered_map<value_t, precision_t> featureCounts;
// unordered_map<value_t, unordered_map<value_t, precision_t>> jointCounts;
// featureCounts.clear();
// jointCounts.clear();
// auto totalWeight = 0;
// for (auto i = 0; i < numSamples; i++) {
// featureCounts[samples[feature][i]] += weights[i];
// jointCounts[samples[feature][i]][labels[i]] += weights[i];
// totalWeight += weights[i];
// }
// precision_t entropy = 0;
// for (auto& [f, count] : featureCounts) {
// auto p_f = count / totalWeight;
// precision_t entropy_f = 0;
// for (auto& [l, jointCount] : jointCounts[f]) {
// auto p_l_f = jointCount / totalWeight;
// entropy_f -= p_l_f * log2(p_l_f);
// }
// entropy += p_f * entropy_f;
// }
// return entropy;
// }
// // I(X;Y) = H(Y) - H(Y|X)
// precision_t SelectKBestWeighted::MutualInformation(const int i)
// {
// // return entropyLabel() - conditionalEntropy(i);
// return 25 / (i + 1);
// }
precision_t SelectKBestWeighted::entropyLabel()
{
return entropy(labels);
}
precision_t SelectKBestWeighted::entropy(const sample_t& data)
{
precision_t p;
precision_t ventropy = 0, totalWeight = 0;
score_t counts(numClasses + 1, 0);
for (auto i = 0; i < data.size(); ++i) {
counts[data[i]] += weights[i];
totalWeight += weights[i];
}
for (auto count : counts) {
p = count / totalWeight;
if (p > 0)
if (nat)
ventropy -= p * log(p);
else
ventropy -= p * log2(p);
}
return ventropy;
}
// H(Y|X) = sum_{x in X} p(x) H(Y|X=x)
precision_t SelectKBestWeighted::conditionalEntropy(const int feature)
{
unordered_map<value_t, precision_t> featureCounts;
unordered_map<value_t, unordered_map<value_t, precision_t>> jointCounts;
featureCounts.clear();
jointCounts.clear();
precision_t totalWeight = 0;
for (auto i = 0; i < numSamples; i++) {
featureCounts[samples[i][feature]] += weights[i];
jointCounts[samples[i][feature]][labels[i]] += weights[i];
totalWeight += weights[i];
}
if (totalWeight == 0)
throw invalid_argument("Total weight should not be zero");
precision_t entropy = 0;
for (auto& [feat, count] : featureCounts) {
auto p_f = count / totalWeight;
precision_t entropy_f = 0;
for (auto& [label, jointCount] : jointCounts[feat]) {
auto p_l_f = jointCount / count;
if (p_l_f > 0) {
double epsilon = 1e-9;
if (nat)
entropy_f -= p_l_f * log(p_l_f + epsilon);
else
entropy_f -= p_l_f * log2(p_l_f + epsilon);
}
}
entropy += p_f * entropy_f;
}
return entropy;
}
// I(X;Y) = H(Y) - H(Y|X)
precision_t SelectKBestWeighted::MutualInformation(const int i)
{
return entropyLabel() - conditionalEntropy(i);
}
score_t SelectKBestWeighted::getScore() const
{
if (!fitted)

25
bayesclass/FeatureSelect.h
View File

@@ -5,31 +5,30 @@
#include <string>
using namespace std;
namespace features {
typedef float precision_t;
typedef double precision_t;
typedef int value_t;
typedef vector<value_t> sample_t;
// typedef vector<sample_t> samples_t;
typedef vector<value_t> samples_t;
typedef vector<sample_t> samples_t;
typedef vector<value_t> labels_t;
typedef vector<precision_t> score_t, weights_t;
class SelectKBestWeighted {
private:
samples_t& samples;
// const labels_t& labels;
// const weights_t& weights;
// const int k;
const samples_t samples;
const labels_t labels;
const weights_t weights;
const int k;
bool nat; // use natural log or log2
int numFeatures, numClasses, numSamples;
bool fitted;
score_t score;
// precision_t entropyLabel();
// precision_t entropy(const sample_t&);
// precision_t conditionalEntropy(const int);
// precision_t MutualInformation(const int);
precision_t entropyLabel();
precision_t entropy(const sample_t&);
precision_t conditionalEntropy(const int);
precision_t MutualInformation(const int);
void outputValues();
public:
// SelectKBestWeighted(samples_t&, labels_t&, weights_t&, int);
SelectKBestWeighted(samples_t&);
SelectKBestWeighted(samples_t&, labels_t&, weights_t&, int, bool);
void fit();
score_t getScore() const;
static inline string version() { return "0.1.0"; };

37
bayesclass/FeatureTest.cpp
View File

@@ -1,37 +0,0 @@
#include "FeatureTest.h"
#include <iostream>
namespace featuresTest {
SelectKBest::SelectKBest(vector<int>& samples) : samples(samples) {}
SelectKBest::SelectKBest() = default;
SelectKBest::~SelectKBest() = default;
void SelectKBest::SelectKBest::fit()
{
numFeatures = 0;
numSamples = samples.size();
outputValues();
fitted = true;
}
void SelectKBest::outputValues()
{
cout << "numFeatures: " << numFeatures << endl;
// cout << "numClasses: " << numClasses << endl;
cout << "numSamples: " << numSamples << endl;
// cout << "k: " << k << endl;
// cout << "weights: ";
// for (auto item : weights)
// cout << item << ", ";
// cout << "end." << endl;
// cout << "labels: ";
// for (auto item : labels)
// cout << item << ", ";
// cout << "end." << endl;
cout << "samples: ";
for (auto item : samples) {
// for (auto item2 : item)
// cout << item2 << ", ";
// cout << "end." << endl;
cout << item << ", ";
}
cout << "end." << endl;
}
}

30
bayesclass/FeatureTest.h
View File

@@ -1,30 +0,0 @@
#ifndef SELECT_K_BEST_TEST_H
#define SELECT_K_BEST_TEST_H
#include <map>
#include <vector>
#include <string>
using namespace std;
namespace featuresTest {
typedef float precision_t;
typedef int value_t;
typedef vector<value_t> sample_t;
// typedef vector<sample_t> samples_t;
typedef vector<value_t> samples_t;
typedef vector<value_t> labels_t;
typedef vector<precision_t> score_t, weights_t;
class SelectKBest {
private:
vector<int>& samples;
int numFeatures, numClasses, numSamples;
bool fitted;
void outputValues();
public:
SelectKBest();
SelectKBest(vector<int>&);
~SelectKBest();
void fit();
static inline string version() { return "0.1.0"; };
};
}
#endif

1021
bayesclass/cSelectFeatures.cpp
View File

File diff suppressed because it is too large Load Diff

49
bayesclass/cSelectFeatures.pyx
View File

@@ -2,50 +2,29 @@
# cython: language_level = 3
from libcpp.vector cimport vector
from libcpp.string cimport string
from libcpp cimport bool
cdef extern from "FeatureTest.h" namespace "featuresTest":
ctypedef float precision_t
cdef cppclass SelectKBest:
SelectKBest(vector[int]&) except +
cdef extern from "FeatureSelect.h" namespace "features":
ctypedef double precision_t
cdef cppclass SelectKBestWeighted:
SelectKBestWeighted(vector[vector[int]]&, vector[int]&, vector[precision_t]&, int, bool) except +
void fit()
string version()
vector[precision_t] getScore()
cdef class CSelectKBest:
cdef SelectKBest *thisptr
def __cinit__(self, X):
self.thisptr = new SelectKBest(X)
cdef class CSelectKBestWeighted:
cdef SelectKBestWeighted *thisptr
def __cinit__(self, X, y, weights, k, natural=False): # log or log2
self.thisptr = new SelectKBestWeighted(X, y, weights, k, natural)
def __dealloc__(self):
del self.thisptr
def fit(self,):
self.thisptr.fit()
return self
def get_score(self):
return self.thisptr.getScore()
def get_version(self):
return self.thisptr.version()
def __reduce__(self):
return (CSelectKBest, ())
# cdef extern from "FeatureSelect.h" namespace "features":
# ctypedef float precision_t
# cdef cppclass SelectKBestWeighted:
# SelectKBestWeighted(vector[int]&) except +
# # SelectKBestWeighted(vector[int]&, vector[int]&, vector[precision_t]&, int) except +
# void fit()
# string version()
# vector[precision_t] getScore()
# cdef class CSelectKBestWeighted:
# cdef SelectKBestWeighted *thisptr
# def __cinit__(self, X, y, weights, k):
# # self.thisptr = new SelectKBestWeighted(X, y, weights, k)
# self.thisptr = new SelectKBestWeighted(X)
# def __dealloc__(self):
# del self.thisptr
# def fit(self,):
# self.thisptr.fit()
# return self
# def get_score(self):
# return self.thisptr.getScore()
# def get_version(self):
# return self.thisptr.version()
# def __reduce__(self):
# return (CSelectKBestWeighted, ())
return (CSelectKBestWeighted, ())

10
test.py Normal file
View File

@@ -0,0 +1,10 @@
from bayesclass.cppSelectFeatures import CSelectKBestWeighted
X = [[x for x in range(i, i + 3)] for i in range(1, 30, 3)]
weights = [25 / (i + 1) for i in range(10)]
labels = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
test = CSelectKBestWeighted(X, labels, weights, 3)
test.fit()
for item in test.get_score():
print(item)