My code from the last meetup

04-Clustering/c++/clusters.dat

@@ -0,0 +1,150 @@
+6.3 2.8 5.1 1.5 1
+6.3 3.4 5.6 2.4 1
+5 3.4 1.5 0.2 2
+4.9 2.5 4.5 1.7 0
+5 3.6 1.4 0.2 2
+5.4 3.9 1.7 0.4 2
+4.6 3.4 1.4 0.3 2
+4.7 3.2 1.3 0.2 2
+4.4 2.9 1.4 0.2 2
+4.9 3.1 1.5 0.1 2
+5.4 3.7 1.5 0.2 2
+4.8 3.4 1.6 0.2 2
+4.8 3 1.4 0.1 2
+4.3 3 1.1 0.1 2
+5.8 4 1.2 0.2 2
+5.7 4.4 1.5 0.4 2
+5.4 3.9 1.3 0.4 2
+5.1 3.5 1.4 0.3 2
+5.7 3.8 1.7 0.3 2
+5.1 3.8 1.5 0.3 2
+5.4 3.4 1.7 0.2 2
+5.1 3.7 1.5 0.4 2
+4.6 3.6 1 0.2 2
+5.1 3.3 1.7 0.5 2
+4.8 3.4 1.9 0.2 2
+5 3 1.6 0.2 2
+5 3.4 1.6 0.4 2
+5.2 3.5 1.5 0.2 2
+5.2 3.4 1.4 0.2 2
+4.7 3.2 1.6 0.2 2
+4.8 3.1 1.6 0.2 2
+5.4 3.4 1.5 0.4 2
+5.2 4.1 1.5 0.1 2
+5.5 4.2 1.4 0.2 2
+4.9 3.1 1.5 0.1 2
+5 3.2 1.2 0.2 2
+5.5 3.5 1.3 0.2 2
+4.9 3.1 1.5 0.1 2
+4.4 3 1.3 0.2 2
+5.1 3.4 1.5 0.2 2
+5 3.5 1.3 0.3 2
+4.5 2.3 1.3 0.3 2
+4.4 3.2 1.3 0.2 2
+5 3.5 1.6 0.6 2
+5.1 3.8 1.9 0.4 2
+4.8 3 1.4 0.3 2
+5.1 3.8 1.6 0.2 2
+4.6 3.2 1.4 0.2 2
+5.3 3.7 1.5 0.2 2
+5 3.3 1.4 0.2 2
+7 3.2 4.7 1.4 1
+6.4 3.2 4.5 1.5 0
+6.9 3.1 4.9 1.5 1
+5.5 2.3 4 1.3 0
+6.5 2.8 4.6 1.5 0
+5.7 2.8 4.5 1.3 0
+6.3 3.3 4.7 1.6 0
+4.9 2.4 3.3 1 3
+6.6 2.9 4.6 1.3 0
+5.2 2.7 3.9 1.4 0
+5 2 3.5 1 3
+5.9 3 4.2 1.5 0
+6 2.2 4 1 0
+6.1 2.9 4.7 1.4 0
+5.6 2.9 3.6 1.3 0
+6.7 3.1 4.4 1.4 0
+5.6 3 4.5 1.5 0
+5.8 2.7 4.1 1 0
+6.2 2.2 4.5 1.5 0
+5.6 2.5 3.9 1.1 0
+5.9 3.2 4.8 1.8 0
+6.1 2.8 4 1.3 0
+6.3 2.5 4.9 1.5 0
+6.1 2.8 4.7 1.2 0
+6.4 2.9 4.3 1.3 0
+6.6 3 4.4 1.4 0
+6.8 2.8 4.8 1.4 1
+6.7 3 5 1.7 1
+6 2.9 4.5 1.5 0
+5.7 2.6 3.5 1 0
+5.5 2.4 3.8 1.1 0
+5.5 2.4 3.7 1 0
+5.8 2.7 3.9 1.2 0
+6 2.7 5.1 1.6 0
+5.4 3 4.5 1.5 0
+6 3.4 4.5 1.6 0
+6.7 3.1 4.7 1.5 1
+6.3 2.3 4.4 1.3 0
+5.6 3 4.1 1.3 0
+5.5 2.5 4 1.3 0
+5.5 2.6 4.4 1.2 0
+6.1 3 4.6 1.4 0
+5.8 2.6 4 1.2 0
+5 2.3 3.3 1 3
+5.6 2.7 4.2 1.3 0
+5.7 3 4.2 1.2 0
+5.7 2.9 4.2 1.3 0
+6.2 2.9 4.3 1.3 0
+5.1 2.5 3 1.1 3
+5.7 2.8 4.1 1.3 0
+6.3 3.3 6 2.5 1
+5.8 2.7 5.1 1.9 0
+7.1 3 5.9 2.1 1
+6.3 2.9 5.6 1.8 1
+6.5 3 5.8 2.2 1
+7.6 3 6.6 2.1 1
+4.6 3.1 1.5 0.2 2
+7.3 2.9 6.3 1.8 1
+6.7 2.5 5.8 1.8 1
+7.2 3.6 6.1 2.5 1
+6.5 3.2 5.1 2 1
+6.4 2.7 5.3 1.9 1
+6.8 3 5.5 2.1 1
+5.7 2.5 5 2 0
+5.8 2.8 5.1 2.4 1
+6.4 3.2 5.3 2.3 1
+6.5 3 5.5 1.8 1
+7.7 3.8 6.7 2.2 1
+7.7 2.6 6.9 2.3 1
+6 2.2 5 1.5 0
+6.9 3.2 5.7 2.3 1
+5.6 2.8 4.9 2 0
+7.7 2.8 6.7 2 1
+6.3 2.7 4.9 1.8 1
+6.7 3.3 5.7 2.1 1
+7.2 3.2 6 1.8 1
+6.2 2.8 4.8 1.8 0
+6.1 3 4.9 1.8 1
+6.4 2.8 5.6 2.1 1
+7.2 3 5.8 1.6 1
+7.4 2.8 6.1 1.9 1
+7.9 3.8 6.4 2 1
+6.4 2.8 5.6 2.2 1
+5.1 3.5 1.4 0.2 2
+6.1 2.6 5.6 1.4 1
+7.7 3 6.1 2.3 1
+4.9 3 1.4 0.2 2
+6.4 3.1 5.5 1.8 1
+6 3 4.8 1.8 0
+6.9 3.1 5.4 2.1 1
+6.7 3.1 5.6 2.4 1
+6.9 3.1 5.1 2.3 1
+5.8 2.7 5.1 1.9 0
+6.8 3.2 5.9 2.3 1
+6.7 3.3 5.7 2.5 1
+6.7 3 5.2 2.3 1
+6.3 2.5 5 1.9 1
+6.5 3 5.2 2 1
+6.2 3.4 5.4 2.3 1
+5.9 3 5.1 1.8 1

04-Clustering/c++/kmeans.cpp

@@ -0,0 +1,174 @@
+#include <iostream>
+#include <fstream>
+#include <string>
+#include <vector>
+#include <math.h>
+#include <time.h>
+
+#include <boost/algorithm/string.hpp>
+
+// K-means clustering using Lloyd's algorithm
+
+using namespace std;
+
+// number of clusters
+size_t K=4;
+
+
+struct Point {
+    Point(double x1,double x2, double x3, double x4) 
+        : x1_(x1),
+          x2_(x2),
+          x3_(x3),
+          x4_(x4) 
+    { cluster_ = -1; }
+
+    Point& operator+(const Point &p) {
+        x1_ += p.x1_;
+        x2_ += p.x2_;
+        x3_ += p.x3_;
+        x4_ += p.x4_;
+        cluster_ += p.cluster_;
+        return *this;
+    }
+
+    double x1_,x2_,x3_,x4_;
+    int cluster_;
+};
+
+typedef std::vector<Point> DataVec;
+
+std::ostream& operator << (std::ostream &o, const Point& p){
+      return o << p.x1_ << " " << p.x2_ << " " << p.x3_ << " " << p.x4_ << " " << p.cluster_;
+}
+
+double dist(const Point& p1, const Point& p2) {
+    return sqrt(pow((p1.x1_-p2.x1_),2.0) + 
+                pow((p1.x2_-p2.x2_),2.0) + 
+                pow((p1.x3_-p2.x3_),2.0) + 
+                pow((p1.x4_-p2.x4_),2.0));
+}
+
+// Fisher-Yates shuffle
+template<class fwditer>
+fwditer random_unique(fwditer begin, fwditer end, size_t num_random) {
+    size_t left = std::distance(begin, end);
+    while (num_random--) {
+        fwditer r = begin;
+        std::advance(r, rand()%left);
+        std::swap(*begin, *r);
+        ++begin;
+        --left;
+    }
+    return begin;
+}
+
+bool getCentroid(const DataVec& data, const int cluster, Point& centroid) {
+    size_t num(1);
+    Point new_centroid(0.0,0.0,0.0,0.0);
+    for (DataVec::const_iterator ct=data.begin();ct!=data.end();++ct) {
+        if (ct->cluster_ == cluster) {
+            new_centroid.x1_ += ct->x1_;
+            new_centroid.x2_ += ct->x2_;
+            new_centroid.x3_ += ct->x3_;
+            new_centroid.x4_ += ct->x4_;
+            ++new_centroid.cluster_;
+            ++num;
+        }
+    }
+    new_centroid.x1_ /= num;
+    new_centroid.x2_ /= num;
+    new_centroid.x3_ /= num;
+    new_centroid.x4_ /= num;
+
+    double d = dist(centroid,new_centroid);
+    std::cout << "getCentroid: d=" << d << "\n";
+    bool changed = d>0.05 ? true : false;
+    centroid = new_centroid;
+    return changed;
+}
+
+bool fit(DataVec& data, DataVec& centroids) {
+    bool converged(true);
+
+    // assign points to closest centroid
+    for (DataVec::iterator it1 = data.begin(); it1!=data.end(); ++it1) {
+        double min_dist = std::numeric_limits<double>::max();
+        int min_clust   = -1;
+        for (DataVec::iterator it2 = centroids.begin(); it2!=centroids.end(); ++it2) {
+            double d = dist(*it1,*it2);
+            if (d < min_dist) {
+                min_dist = d;
+                min_clust = it2-centroids.begin();
+            }
+        }
+        //std::cout << "Point " << *it1 << "\n";
+        //std::cout << "min_dist=" << min_dist << " min_clust=" << min_clust << "\n";
+        it1->cluster_ = min_clust;
+    }
+
+    // re-estimate centroids
+    for (size_t i=0;i<K;++i) {
+        std::cout << "Centroid at " << i << " was " << centroids[i] << "\n";
+        bool centroidUpdated = getCentroid(data,i,centroids[i]);
+        if (centroidUpdated) converged = false;
+        std::cout << "Centroid at " << i << " is now " << centroids[i] << "\n";
+    }
+
+    return converged;
+}
+
+
+int main(int argc, char** argv) {
+
+    // seed random generator
+    //srand(time(NULL));
+
+    ifstream infile("../iris.data");
+    string line;
+    DataVec data;
+    while (std::getline(infile, line))
+    {
+        std::vector<std::string> fields;
+        boost::split(fields,line, boost::is_any_of(","));
+        assert(fields.size() == 5);
+
+        double x1 = atof(fields[0].c_str());
+        double x2 = atof(fields[1].c_str());
+        double x3 = atof(fields[2].c_str());
+        double x4 = atof(fields[3].c_str());
+
+        Point p(x1,x2,x3,x4);
+        //std::cout << p << std::endl;
+        data.push_back(p);
+    }
+    std::cout << "Collected " << data.size() << " points. " << std::endl;
+    assert(data.size()>K);
+
+    // init centroids to random points
+    DataVec centroids;
+    //centroids.reserve(K);
+    DataVec::iterator dataBegin = random_unique(data.begin(),data.end(),K);
+    std::cout << K << " random points " << std::endl;
+    for (size_t i=0;i<K;++i) {
+        std::cout << data[i] << "\n";
+        centroids.push_back(data[i]);
+    }
+    std::cout << centroids.size() << " centroids " << std::endl;
+
+    // Lloyd's algorithm to iteratively fit the cluster centroids.
+    bool done = fit(data,centroids);
+    while(!done) {
+        done = fit(data,centroids);
+        std::cout << done << "\n";
+    }
+
+    ofstream of("clusters.dat");
+    for (DataVec::iterator it = data.begin(); it!=data.end(); ++it) {
+        of << *it << std::endl;
+    }
+    of.close();
+
+    return EXIT_SUCCESS;
+}
+

04-Clustering/c++/plot.R

@@ -0,0 +1,5 @@
+x <- read.table("clusters.dat");
+x[,5] <- as.factor(x[,5])
+png(file="cluster.png",width=600,height=600);
+plot(x[,1],x[,2],col=x[,5])
+dev.off();