From 7f4b09d2d63b7db0db9156b6b4aec2750ab28a84 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ricardo=20Monta=C3=B1ana?= <rmontanana@gmail.com>
Date: Wed, 7 Dec 2022 01:26:38 +0100
Subject: [PATCH] Add good_cut filter

---
 fimdlp/CPPFImdlp.cpp | 34 +++++++++++++++++++++++++---------
 fimdlp/CPPFImdlp.h   |  1 +
 fimdlp/main.cpp      | 21 ++++++++++++++++++---
 3 files changed, 44 insertions(+), 12 deletions(-)

diff --git a/fimdlp/CPPFImdlp.cpp b/fimdlp/CPPFImdlp.cpp
index b7ae995..be5819f 100644
--- a/fimdlp/CPPFImdlp.cpp
+++ b/fimdlp/CPPFImdlp.cpp
@@ -46,9 +46,9 @@ namespace mdlp {
         if (X.size() == 0 || y.size() == 0) {
             throw invalid_argument("X and y must have at least one element");
         }
-        this->indices = sortIndices(X_);
-        this->xDiscretized = labels(X.size(), -1);
-        this->numClasses = Metrics::numClasses(y, indices, 0, X.size());
+        indices = sortIndices(X_);
+        xDiscretized = labels(X.size(), -1);
+        numClasses = Metrics::numClasses(y, indices, 0, X.size());
 
         if (proposal) {
             computeCutPointsProposal();
@@ -168,9 +168,9 @@ namespace mdlp {
             }
             while (idx < numElements && xCur == xPivot);
             // Check if the class changed and there are more than 1 element
-            if ((idx - start > 1) && (yPivot == -1 || yPrev != yCur)) {
+            if ((idx - start > 1) && (yPivot == -1 || yPrev != yCur) && goodCut(start, idx, numElements + 1)) {
                 // Must we add the entropy criteria here?
-                // if (totalEntropy - (entropyLeft + entropyRight) < 0) { Accept cut point }
+                // if (totalEntropy - (entropyLeft + entropyRight) > 0) { Accept cut point }
                 cutPoint.start = start;
                 cutPoint.end = idx;
                 start = idx;
@@ -211,16 +211,17 @@ namespace mdlp {
         int yPrev;
         bool first = true;
         // idxPrev is the index of the init instance of the cutPoint
-        size_t index, idxPrev = 0, idx = indices[0];
+        size_t index, idxPrev = 0, last, idx = indices[0];
         xPrev = X[idx];
         yPrev = y[idx];
-        for (index = 0; index < size_t(indices.size()) - 1; index++) {
+        last = indices.size() - 1;
+        for (index = 0; index < last; index++) {
             idx = indices[index];
             // Definition 2 Cut points are always on class boundaries && 
             // there are more than 1 items in the interval
-            if (y[idx] != yPrev && xPrev < X[idx] && idxPrev != index - 1) {
+             // if (entropy of interval) > (entropyLeft + entropyRight)) { Accept cut point } (goodCut)
+            if (y[idx] != yPrev && xPrev < X[idx] && idxPrev != index - 1 && goodCut(idxPrev, idx, last + 1)) {
                 // Must we add the entropy criteria here?
-                // if (totalEntropy - (entropyLeft + entropyRight) < 0) { Accept cut point }
                 if (first) {
                     first = false;
                     cutPoint.fromValue = numeric_limits<float>::lowest();
@@ -253,6 +254,21 @@ namespace mdlp {
         }
         cutPoints = cutPts;
     }
+    bool CPPFImdlp::goodCut(size_t start, size_t cut, size_t end)
+    {
+        /*
+        Meter las entropías en una matríz cuadrada dispersa (samples, samples) M[start, end] iniciada a -1 y si no se ha calculado calcularla y almacenarla
+
+
+        */
+        float entropyLeft = Metrics::entropy(y, indices, start, cut, numClasses);
+        float entropyRight = Metrics::entropy(y, indices, cut, end, numClasses);
+        float entropyInterval = Metrics::entropy(y, indices, start, end, numClasses);
+        if (debug)
+            printf("Entropy L, R, T: L(%5.3g) + R(%5.3g) - T(%5.3g) \t", entropyLeft, entropyRight, entropyInterval);
+        //return (entropyInterval - (entropyLeft + entropyRight) > 0);
+        return true;
+    }
     // Argsort from https://stackoverflow.com/questions/1577475/c-sorting-and-keeping-track-of-indexes
     indices_t CPPFImdlp::sortIndices(samples& X_)
     {
diff --git a/fimdlp/CPPFImdlp.h b/fimdlp/CPPFImdlp.h
index 8965092..926b6e6 100644
--- a/fimdlp/CPPFImdlp.h
+++ b/fimdlp/CPPFImdlp.h
@@ -22,6 +22,7 @@ namespace mdlp {
         void computeCutPointsProposal();
         bool evaluateCutPoint(cutPoint_t, cutPoint_t);
         void filterCutPoints();
+        bool goodCut(size_t, size_t, size_t); // if the cut candidate reduces entropy
 
     public:
         CPPFImdlp();
diff --git a/fimdlp/main.cpp b/fimdlp/main.cpp
index 8c918fd..7c1c4a1 100644
--- a/fimdlp/main.cpp
+++ b/fimdlp/main.cpp
@@ -19,19 +19,34 @@ int main()
 
     // Read the Data from the file
     // as String Vector
+    size_t col;
     vector<string> row;
     string line, word;
+    vector<vector<float>> dataset = vector<vector<float>>(15, vector<float>());
     while (getline(fin, line)) {
         if (count++ > 215) {
-            row.clear();
             stringstream ss(line);
+            col = 0;
             while (getline(ss, word, ',')) {
-                row.push_back(word);
-                cout << word << " ";
+                col = col % 15;
+                dataset[col].push_back(stof(word));
+                cout << col << "-" << word << " ";
+                col++;
             }
             cout << endl;
         }
     }
+    labels y = labels(dataset[0].begin(), dataset[0].end());
+    cout << "Column 0 (y): " << y.size() << endl;
+    for (auto item : y) {
+        cout << item << " ";
+    }
+    CPPFImdlp test = CPPFImdlp(false, 6, true);
+    test.fit(dataset[3], y);
+    cout << "Cut points: " << test.getCutPoints().size() << endl;
+    for (auto item : test.getCutPoints()) {
+        cout << item << " ";
+    }
     fin.close();
     return 0;
 }
\ No newline at end of file