From c9a61689e92db115e2afadb7addd0a6855b6227d Mon Sep 17 00:00:00 2001
From: Volkan Kumtepeli <volkan.kumtepeli@gmail.com>
Date: Sun, 17 Mar 2024 03:14:11 +0000
Subject: [PATCH 01/15] [0, Nc) indexing and silhouette score

---
 dtwc/Problem.cpp        | 42 ++++++++++++++-----------------
 dtwc/Problem.hpp        |  4 ++-
 dtwc/Problem_IO.cpp     | 18 +++++++-------
 dtwc/dtwc_cl.cpp        | 28 +++++++++++----------
 dtwc/initialisation.cpp |  2 ++
 dtwc/initialisation.hpp | 17 +++++++------
 dtwc/mip/mip_Gurobi.cpp | 27 ++++++++++----------
 dtwc/mip/mip_Highs.cpp  | 18 +++++++-------
 dtwc/scores.cpp         | 55 +++++++++++++++++++++++------------------
 9 files changed, 110 insertions(+), 101 deletions(-)

diff --git a/dtwc/Problem.cpp b/dtwc/Problem.cpp
index 61efeec..a00a8f8 100644
--- a/dtwc/Problem.cpp
+++ b/dtwc/Problem.cpp
@@ -40,7 +40,7 @@ namespace dtwc {
 void Problem::resize()
 {
   clusters_ind.resize(size());
-  centroids_ind.resize(Nc);
+  centroids_ind.resize(cluster_size());
 }
 
 /**
@@ -195,12 +195,12 @@ void Problem::cluster_by_MIP()
  */
 void Problem::assignClusters()
 {
-  auto assignClustersTask = [this](int i_p) // i_p and i_c in [0, Np]
+  auto assignClustersTask = [this](int i_p) //!< i_p  and i_c in [0, Np)
   {
-    clusters_ind[i_p] = *std::min_element(
-      centroids_ind.begin(),
-      centroids_ind.end(),
-      [this, i_p](int i_c1, int ic2) { return distByInd(i_p, i_c1) < distByInd(i_p, ic2); });
+    auto minIt = std::min_element(centroids_ind.begin(), centroids_ind.end(), [this, i_p](int ic_1, int ic_2) {
+      return distByInd(i_p, ic_1) < distByInd(i_p, ic_2);
+    });
+    clusters_ind[i_p] = std::distance(centroids_ind.begin(), minIt);
   };
 
   clusters_ind.resize(data.size()); // Resize before assigning.
@@ -231,15 +231,14 @@ void Problem::distanceInClusters()
  */
 void Problem::calculateMedoids()
 {
-  std::vector<double> pointCosts(size());
+  static std::vector<double> pointCosts(size()), clusterCosts(cluster_size());
+  pointCosts.resize(size());
 
   auto findBetterMedoidTask = [&](int i_p) // i_p is point index.
   {
-    const auto clusterNo = clusters_ind[i_p];
     double sum{ 0 };
-
     for (const auto i : Range(size()))
-      if (clusters_ind[i] == clusterNo)
+      if (clusters_ind[i] == clusters_ind[i_p]) // If they are in the same cluster
         sum += distByInd(i_p, i);
 
     pointCosts[i_p] = sum;
@@ -247,17 +246,12 @@ void Problem::calculateMedoids()
 
   run(findBetterMedoidTask, size());
 
-  for (const auto i : Range(Nc)) {
-    const auto centroidNow = centroids_ind[i];
-    double minCost{ std::numeric_limits<double>::max() };
-    int minInd{};
-
-    for (const auto i_p : Range(size()))
-      if ((clusters_ind[i_p] == centroidNow) && (pointCosts[i_p] < minCost))
-        std::tie(minCost, minInd) = std::tie(pointCosts[i_p], i_p);
-
-    centroids_ind[i] = minInd;
-  }
+  clusterCosts.assign(cluster_size(), std::numeric_limits<double>::max());
+  for (const auto i : Range(size()))
+    if (pointCosts[i] < clusterCosts[clusters_ind[i]]) {
+      clusterCosts[clusters_ind[i]] = pointCosts[i];
+      centroids_ind[clusters_ind[i]] = i;
+    }
 }
 
 /**
@@ -351,12 +345,12 @@ std::pair<int, double> Problem::cluster_by_kMedoidsPAM_single(int rep)
 double Problem::findTotalCost()
 {
   double sum = 0;
-  for (int i = 0; i < data.size(); i++) {
+  for (int i : Range(size())) {
     if constexpr (settings::isDebug)
       std::cout << "Distance between " << i << " and closest cluster " << clusters_ind[i]
-                << " which is: " << distByInd(i, clusters_ind[i]) << "\n";
+                << " which is: " << distByInd(i, centroid_of(i)) << "\n";
 
-    sum += distByInd(i, clusters_ind[i]); // #TODO should cost be square or like this?
+    sum += distByInd(i, centroid_of(i)); // #TODO should cost be square or like this?
   }
 
   return sum;
diff --git a/dtwc/Problem.hpp b/dtwc/Problem.hpp
index 897b48a..16590d6 100644
--- a/dtwc/Problem.hpp
+++ b/dtwc/Problem.hpp
@@ -73,7 +73,7 @@ class Problem
   std::string name{};                            /*!< Problem name. */
   Data data;                                     /*!< Data associated with the problem. */
 
-  std::vector<int> clusters_ind;  //!< Indices of which point belongs to which cluster. [0,Np]
+  std::vector<int> clusters_ind;  //!< Indices of which point belongs to which cluster. [0,Nc)
   std::vector<int> centroids_ind; //!< indices of cluster centroids. [0, Np)
 
   // Constructors:
@@ -97,6 +97,8 @@ class Problem
   void resize();
 
   // Getters and setters:
+  int centroid_of(int i_p) const { return centroids_ind[clusters_ind[i_p]]; } // [0, Np) Get the centroid of the cluster of i_p
+
   void readDistanceMatrix(const fs::path &distMat_path);
   void set_numberOfClusters(int Nc_);
   void set_clusters(std::vector<int> &candidate_centroids);
diff --git a/dtwc/Problem_IO.cpp b/dtwc/Problem_IO.cpp
index afadee8..9369ac1 100644
--- a/dtwc/Problem_IO.cpp
+++ b/dtwc/Problem_IO.cpp
@@ -58,7 +58,7 @@ void Problem::writeMedoids(std::vector<std::vector<int>> &centroids_all, int rep
  */
 void Problem::printClusters() const
 {
-  std::cout << "Clusters: ";
+  std::cout << "Clusters centroids: ";
   for (auto ind : centroids_ind)
     std::cout << get_name(ind) << ' ';
 
@@ -68,7 +68,7 @@ void Problem::printClusters() const
     std::cout << get_name(centroids_ind[i_c]) << " has: ";
 
     for (const auto i_p : Range(size()))
-      if (clusters_ind[i_p] == centroids_ind[i_c])
+      if (clusters_ind[i_p] == i_c)
         std::cout << get_name(i_p) << " ";
 
     std::cout << '\n';
@@ -85,7 +85,7 @@ void Problem::writeClusters()
 
   std::ofstream myFile(output_folder / file_name, std::ios_base::out);
 
-  myFile << "Clusters:\n";
+  myFile << "Cluster centroids:\n";
 
   for (int i{ 0 }; i < Nc; i++) {
     if (i != 0) myFile << ',';
@@ -96,8 +96,8 @@ void Problem::writeClusters()
   myFile << "\n\n"
          << "Data" << ',' << "its cluster\n";
 
-  for (int i{ 0 }; i < data.size(); i++)
-    myFile << get_name(i) << ',' << get_name(clusters_ind[i]) << '\n';
+  for (int i : Range(size()))
+    myFile << get_name(i) << ',' << get_name(centroid_of(i)) << '\n';
 
   myFile << "Procedure is completed with cost: " << findTotalCost() << '\n';
 
@@ -114,12 +114,12 @@ void Problem::writeSilhouettes()
 
   std::string silhouette_name{ name + "_silhouettes_Nc_" };
 
-  silhouette_name += std::to_string(Nc) + ".csv";
+  silhouette_name += std::to_string(cluster_size()) + ".csv";
 
   std::ofstream myFile(output_folder / silhouette_name, std::ios_base::out);
 
   myFile << "Silhouettes:\n";
-  for (int i{ 0 }; i < data.size(); i++)
+  for (auto i : Range(size()))
     myFile << get_name(i) << ',' << silhouettes[i] << '\n';
 
   myFile.close();
@@ -136,9 +136,9 @@ void Problem::writeMedoidMembers(int iter, int rep) const
                                   + std::to_string(rep) + "_iter_" + std::to_string(iter) + ".csv";
 
   std::ofstream medoidMembers(output_folder / medoid_name, std::ios_base::out);
-  for (const auto i_c : Range(Nc)) {
+  for (const auto i_c : Range(cluster_size())) {
     for (const auto i_p : Range(size()))
-      if (clusters_ind[i_p] == centroids_ind[i_c])
+      if (clusters_ind[i_p] == i_c)
         medoidMembers << get_name(i_p) << ',';
 
     medoidMembers << '\n';
diff --git a/dtwc/dtwc_cl.cpp b/dtwc/dtwc_cl.cpp
index e50f9d9..3c981f4 100644
--- a/dtwc/dtwc_cl.cpp
+++ b/dtwc/dtwc_cl.cpp
@@ -1,10 +1,12 @@
-/*
- * dtwc_cl.cpp
+/**
+ * @file dtwc_cl.cpp
+ * @brief Command line interface for DTWC++
  *
- * Command line interface for DTWC++
+ * This file contains the implementation of the command line interface for DTWC++.
+ * It provides a command line interface for users to interact with the DTWC++ library.
  *
- * Created on: 11 Dec 2023
- *  Author(s): Volkan Kumtepeli, Becky Perriment
+ * @date 11 Dec 2023
+ * @authors Volkan Kumtepeli, Becky Perriment
  */
 
 #include "dtwc.hpp"
@@ -31,9 +33,9 @@ int main(int argc, char **argv)
   std::string distMatPath{ "" };
 
   int maxIter{ dtwc::settings::DEFAULT_MAX_ITER };
-  int skipRows{0}, skipCols{0};
+  int skipRows{ 0 }, skipCols{ 0 };
   int N_repetition{ 1 };
-  int bandWidth{-1};
+  int bandWidth{ -1 };
 
   CLI::App app{ app_description };
 
@@ -60,8 +62,8 @@ int main(int argc, char **argv)
 
   std::cout << "Arguments are parsed." << std::endl;
 
-  auto Nc = str_to_range(Nc_str); // dtwc::Range(3,5); 
-  dtwc::Clock clk; // Create a clock object
+  auto Nc = str_to_range(Nc_str); //!< dtwc::Range(3,5);
+  dtwc::Clock clk;                //!< Create a clock object
 
   std::cout << "Nc_str : " << Nc_str << '\n';
   std::cout << "name : " << probName << '\n';
@@ -72,9 +74,9 @@ int main(int argc, char **argv)
   std::cout << "Max iteration : " << maxIter << std::endl;
 
   dtwc::DataLoader dl{ inputPath };
-  dl.startColumn(skipCols).startRow(skipRows); // Since dummy files are in Pandas format skip first row/column.
+  dl.startColumn(skipCols).startRow(skipRows); //!< Since dummy files are in Pandas format skip first row/column.
 
-  dtwc::Problem prob{ probName, dl }; // Create a problem.
+  dtwc::Problem prob{ probName, dl }; //!< Create a problem.
   std::cout << "Data loading finished at " << clk << "\n";
 
   prob.maxIter = maxIter;
@@ -82,7 +84,7 @@ int main(int argc, char **argv)
   prob.output_folder = outPath;
   prob.band = bandWidth;
   try {
-    if(distMatPath != "")
+    if (distMatPath != "")
       prob.readDistanceMatrix(distMatPath);
   } catch (const std::exception &e) {
     std::cout << "Distance matrix could not be read! Continuing without matrix!" << std::endl;
@@ -106,7 +108,7 @@ int main(int argc, char **argv)
 
   for (auto nc : Nc) {
     std::cout << "\n\nClustering by " << method << " for Number of clusters : " << nc << std::endl;
-    prob.set_numberOfClusters(nc); // Nc = number of clusters.
+    prob.set_numberOfClusters(nc); //!< Nc = number of clusters.
     prob.cluster_and_process();
   }
 
diff --git a/dtwc/initialisation.cpp b/dtwc/initialisation.cpp
index f9ab225..d763581 100644
--- a/dtwc/initialisation.cpp
+++ b/dtwc/initialisation.cpp
@@ -1,3 +1,5 @@
+#include <stdexcept> // Add missing include directive for the <stdexcept> header file
+
 /**
  * @file initialisation.cpp
  *
diff --git a/dtwc/initialisation.hpp b/dtwc/initialisation.hpp
index e433cd2..ba6850c 100644
--- a/dtwc/initialisation.hpp
+++ b/dtwc/initialisation.hpp
@@ -1,10 +1,11 @@
-/*
- * initialisation.hpp
+/**
+ * @file initialisation.hpp
+ * @brief Header file for initialisation functions.
  *
- * Header file for initialisation functions.
-
- *  Created on: 19 Jan 2021
- *   Author(s): Volkan Kumtepeli, Becky Perriment
+ * This file contains the declarations of initialisation functions for the dtwc namespace.
+ *
+ * @date 19 Jan 2021
+ * @authors Volkan Kumtepeli, Becky Perriment
  */
 
 #pragma once
@@ -12,7 +13,7 @@
 namespace dtwc {
 class Problem;
 namespace init {
-  void random(Problem &prob);   // Random centroids initialisation
-  void Kmeanspp(Problem &prob); // Kmeanspp centroids initialisation
+  void random(Problem &prob);   //!< This function initializes the centroids randomly.
+  void Kmeanspp(Problem &prob); //!< This function initializes the centroids using the K-means++ algorithm.
 } // namespace init
 } // namespace dtwc
\ No newline at end of file
diff --git a/dtwc/mip/mip_Gurobi.cpp b/dtwc/mip/mip_Gurobi.cpp
index 270d052..d88e348 100644
--- a/dtwc/mip/mip_Gurobi.cpp
+++ b/dtwc/mip/mip_Gurobi.cpp
@@ -11,6 +11,8 @@
 #include "mip.hpp"
 #include "../Problem.hpp"
 #include "../settings.hpp"
+#include "../types/types.hpp" // for Range
+
 
 #include <vector>
 #include <string_view>
@@ -30,8 +32,7 @@ void MIP_clustering_byGurobi(Problem &prob)
 {
 #ifdef DTWC_ENABLE_GUROBI
 
-  const auto Nb = prob.data.size();
-  const auto Nc = prob.cluster_size();
+  const auto Nb(prob.size()), Nc(prob.cluster_size());
   prob.centroids_ind.clear();
 
   try {
@@ -41,22 +42,22 @@ void MIP_clustering_byGurobi(Problem &prob)
     // Create variables
     std::unique_ptr<GRBVar[]> w{ model.addVars(Nb * Nb, GRB_BINARY) };
 
-    for (int i{ 0 }; i < Nb; i++) {
+    for (auto i : Range(Nb)) {
       GRBLinExpr lhs = 0;
-      for (int j{ 0 }; j < Nb; j++)
+      for (auto j : Range(Nb))
         lhs += w[j + i * Nb];
 
       model.addConstr(lhs, '=', 1.0);
     }
 
 
-    for (int j{ 0 }; j < Nb; j++)
-      for (int i{ 0 }; i < Nb; i++)
+    for (auto j : Range(Nb))
+      for (auto i : Range(Nb))
         model.addConstr(w[i + j * Nb] <= w[i * (Nb + 1)]);
 
     {
       GRBLinExpr lhs = 0;
-      for (int i{ 0 }; i < Nb; i++)
+      for (auto i : Range(Nb))
         lhs += w[i * (Nb + 1)];
 
       model.addConstr(lhs == Nc); // There should be Nc clusters.
@@ -66,8 +67,8 @@ void MIP_clustering_byGurobi(Problem &prob)
     const auto scaling_factor = std::max(prob.maxDistance() / 2.0, 1.0);
     // Set objective
     GRBLinExpr obj = 0;
-    for (int j{ 0 }; j < Nb; j++)
-      for (int i{ 0 }; i < Nb; i++)
+    for (auto j : Range(Nb))
+      for (auto i : Range(Nb))
         obj += w[i + j * Nb] * prob.distByInd(i, j) / scaling_factor;
 
     model.setObjective(obj, GRB_MINIMIZE);
@@ -79,15 +80,15 @@ void MIP_clustering_byGurobi(Problem &prob)
 
     model.optimize();
 
-    for (int i{ 0 }; i < Nb; i++)
+    for (auto i : Range(Nb))
       if (w[i * (Nb + 1)].get(GRB_DoubleAttr_X) > 0.5)
         prob.centroids_ind.push_back(i);
 
     prob.clusters_ind.resize(Nb);
 
-    for (auto i : prob.centroids_ind)
-      for (int j{ 0 }; j < Nb; j++)
-        if (w[i + j * Nb].get(GRB_DoubleAttr_X) > 0.5)
+    for (auto i : Range(prob.cluster_size()))
+      for (auto j : Range(Nb))
+        if (w[prob.centroids_ind[i] + j * Nb].get(GRB_DoubleAttr_X) > 0.5)
           prob.clusters_ind[j] = i;
 
   } catch (GRBException &e) {
diff --git a/dtwc/mip/mip_Highs.cpp b/dtwc/mip/mip_Highs.cpp
index 3fe469d..bbb92c0 100644
--- a/dtwc/mip/mip_Highs.cpp
+++ b/dtwc/mip/mip_Highs.cpp
@@ -33,15 +33,15 @@ void extract_mip_solution(Problem &prob, const T &solution)
   prob.centroids_ind.clear();
   const auto Nb = prob.data.size();
 
-  for (int i{ 0 }; i < Nb; i++)
+  for (auto i : Range(Nb))
     if (solution[i * (Nb + 1)] > 0.5)
       prob.centroids_ind.push_back(i);
 
   prob.clusters_ind.resize(Nb);
 
-  for (auto i : prob.centroids_ind) 
-    for (int j{ 0 }; j < Nb; j++)
-      if (solution[i * Nb + j] > 0.5)
+  for (auto i : Range(prob.cluster_size()))
+    for (auto j : Range(Nb))
+      if (solution[prob.centroids_ind[i] * Nb + j] > 0.5)
         prob.clusters_ind[j] = i;
 }
 
@@ -173,11 +173,11 @@ void MIP_clustering_byHiGHS(Problem &prob)
 
   // Get the solution information
   const HighsInfo &info = highs.getInfo();
-  std::cout << "Simplex iteration count: " << info.simplex_iteration_count << '\n';
-  std::cout << "Objective function value: " << info.objective_function_value << '\n';
-  std::cout << "Primal  solution status: " << highs.solutionStatusToString(info.primal_solution_status) << '\n';
-  std::cout << "Dual    solution status: " << highs.solutionStatusToString(info.dual_solution_status) << '\n';
-  std::cout << "Basis: " << highs.basisValidityToString(info.basis_validity) << '\n';
+  std::cout << "Simplex iteration count: " << info.simplex_iteration_count << '\n'
+            << "Objective function value: " << info.objective_function_value << '\n'
+            << "Primal  solution status: " << highs.solutionStatusToString(info.primal_solution_status) << '\n'
+            << "Dual    solution status: " << highs.solutionStatusToString(info.dual_solution_status) << '\n'
+            << "Basis: " << highs.basisValidityToString(info.basis_validity) << '\n';
 
   // Get the solution values
   extract_mip_solution(prob, highs.getSolution().col_value);
diff --git a/dtwc/scores.cpp b/dtwc/scores.cpp
index 6eda4e1..6b78cf6 100644
--- a/dtwc/scores.cpp
+++ b/dtwc/scores.cpp
@@ -18,6 +18,7 @@
 #include <iostream>
 #include <vector>
 #include <cstddef>
+#include <utility> // for pair
 
 namespace dtwc::scores {
 
@@ -36,41 +37,47 @@ namespace dtwc::scores {
  */
 std::vector<double> silhouette(Problem &prob)
 {
-  const auto Nb = prob.data.size();    //!< Number of profiles
+  const auto Nb = prob.size();         //!< Number of profiles
   const auto Nc = prob.cluster_size(); //!< Number of clusters
 
-  std::vector<double> silhouettes(Nb);
+  std::vector<double> silhouettes(Nb, -1); //!< Silhouette scores for each profile initialised to -1
 
   if (prob.centroids_ind.empty()) {
     std::cout << "Please cluster the data before calculating silhouette!" << std::endl;
     return silhouettes;
   }
 
+  prob.fillDistanceMatrix(); //!< We need all pairwise distance for silhouette score.
+
   auto oneTask = [&](size_t i_b) {
-    // auto i_c = prob.clusters_ind[i_b];
-
-    // if (prob.cluster_members[i_c].size() == 1)
-    //   silhouettes[i_b] = 0;
-    // else {
-    //   thread_local std::vector<double> mean_distances(Nc);
-
-    //   for (int i = 0; i < Nb; i++)
-    //     mean_distances[prob.clusters_ind[i]] += prob.distByInd(i, i_b);
-
-    //   auto min = std::numeric_limits<double>::max();
-    //   for (int i = 0; i < Nc; i++) // Finding means:
-    //     if (i == i_c)
-    //       mean_distances[i] /= (prob.cluster_members[i].size() - 1);
-    //     else {
-    //       mean_distances[i] /= prob.cluster_members[i].size();
-    //       min = std::min(min, mean_distances[i]);
-    //     }
-
-    //   silhouettes[i_b] = (min - mean_distances[i_c]) / std::max(min, mean_distances[i_c]);
-    // }
+    const auto i_c = prob.clusters_ind[i_b];
+
+    thread_local std::vector<std::pair<int, double>> mean_distances(Nc);
+    mean_distances.assign(Nc, { 0, 0 });
+
+    for (auto i : Range(prob.size())) {
+      mean_distances[prob.clusters_ind[i]].first++;
+      mean_distances[prob.clusters_ind[i]].second += prob.distByInd(i, i_b);
+    }
+
+
+    if (mean_distances[i_c].first == 1) // If the profile is the only member of the cluster
+      silhouettes[i_b] = 0;
+    else {
+      auto min = std::numeric_limits<double>::max();
+      for (int i = 0; i < Nc; i++) // Finding means:
+        if (i == i_c)
+          mean_distances[i].second /= (mean_distances[i].first - 1);
+        else {
+          mean_distances[i].second /= mean_distances[i].first;
+          min = std::min(min, mean_distances[i].second);
+        }
+
+      silhouettes[i_b] = (min - mean_distances[i_c].second) / std::max(min, mean_distances[i_c].second);
+    }
   };
 
-  dtwc::run(oneTask, Nb);
+  dtwc::run(oneTask, prob.size());
 
   return silhouettes;
 }

From 9a24030a6a635f6cfb5cd5974f58d406fe1d6ca3 Mon Sep 17 00:00:00 2001
From: Volkan Kumtepeli <volkan.kumtepeli@gmail.com>
Date: Mon, 18 Mar 2024 03:01:17 +0000
Subject: [PATCH 02/15] python wrapper

---
 .gitignore               |   1 +
 CMakeLists.txt           |  11 ++++
 cmake/Dependencies.cmake |   1 -
 pyproject.toml           |   3 +
 python/__init__.py       |   1 +
 python/py_main.cpp       |  50 ++++++++++++++
 setup.py                 | 139 +++++++++++++++++++++++++++++++++++++++
 7 files changed, 205 insertions(+), 1 deletion(-)
 create mode 100644 pyproject.toml
 create mode 100644 python/__init__.py
 create mode 100644 python/py_main.cpp
 create mode 100644 setup.py

diff --git a/.gitignore b/.gitignore
index d35ed2d..b173473 100644
--- a/.gitignore
+++ b/.gitignore
@@ -49,3 +49,4 @@ cmake-build-debug/*
 cmake-build-relwithdebinfo/*
 build/*
 bin/*
+*.egg-info/*
diff --git a/CMakeLists.txt b/CMakeLists.txt
index 293fafa..023b32d 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -78,4 +78,15 @@ endif()
 
 if(BUILD_BENCHMARK)
   add_subdirectory(benchmark)
+endif()
+
+find_package(pybind11 QUIET)
+
+if(pybind11_FOUND)
+    pybind11_add_module(dtwcpp 
+            python/py_main.cpp)
+
+    target_link_libraries(dtwcpp PRIVATE dtwc++ project_warnings project_options armadillo)
+else()
+    message(STATUS "pybind11 not found. Skipping the dtwcpp module.")
 endif()
\ No newline at end of file
diff --git a/cmake/Dependencies.cmake b/cmake/Dependencies.cmake
index 43f140e..728ee40 100644
--- a/cmake/Dependencies.cmake
+++ b/cmake/Dependencies.cmake
@@ -55,5 +55,4 @@ function(dtwc_setup_dependencies)
     OPTIONS
     "BUILD_SMOKE_TEST OFF"
   )
-
 endfunction()
diff --git a/pyproject.toml b/pyproject.toml
new file mode 100644
index 0000000..beb36de
--- /dev/null
+++ b/pyproject.toml
@@ -0,0 +1,3 @@
+[build-system]
+requires = ["setuptools","prebuilt_binaries", "cmake>=3.21", "pybind11", "wheel", "ninja"]
+build-backend = "setuptools.build_meta"
\ No newline at end of file
diff --git a/python/__init__.py b/python/__init__.py
new file mode 100644
index 0000000..8b13789
--- /dev/null
+++ b/python/__init__.py
@@ -0,0 +1 @@
+
diff --git a/python/py_main.cpp b/python/py_main.cpp
new file mode 100644
index 0000000..f237ade
--- /dev/null
+++ b/python/py_main.cpp
@@ -0,0 +1,50 @@
+#include <pybind11/pybind11.h>
+#include <pybind11/stl.h>
+#include <pybind11/numpy.h>
+
+#include "../dtwc/Problem.hpp"
+
+#include <vector>
+#include <string>
+#include <string_view>
+
+namespace py = pybind11;
+
+using namespace dtwc;
+
+PYBIND11_MODULE(dtwcpp, m)
+{
+  m.doc() = "DTWC++ (Dynamic Time Warping Clustering++ Library)";
+
+  py::class_<Problem>(m, "Problem")
+    .def(py::init<>())
+    .def(py::init<std::string_view>())
+    .def(py::init<std::string_view, DataLoader &>())
+    .def("size", &Problem::size)
+    .def("cluster_size", &Problem::cluster_size)
+    .def("get_name", (const std::string &(Problem::*)(size_t) const) & Problem::get_name, py::return_value_policy::reference)
+    .def("p_vec", (const std::vector<data_t> &(Problem::*)(size_t) const) & Problem::p_vec, py::return_value_policy::reference)
+    .def("refreshDistanceMatrix", &Problem::refreshDistanceMatrix)
+    .def("resize", &Problem::resize)
+    .def("centroid_of", &Problem::centroid_of)
+    .def("readDistanceMatrix", &Problem::readDistanceMatrix)
+    .def("set_numberOfClusters", &Problem::set_numberOfClusters)
+    .def("set_clusters", &Problem::set_clusters)
+    .def("set_solver", &Problem::set_solver)
+    .def("set_data", &Problem::set_data)
+    .def("maxDistance", &Problem::maxDistance)
+    .def("distByInd", &Problem::distByInd)
+    .def("isDistanceMatrixFilled", &Problem::isDistanceMatrixFilled)
+    .def("fillDistanceMatrix", &Problem::fillDistanceMatrix)
+    .def("printDistanceMatrix", &Problem::printDistanceMatrix)
+    .def("writeDistanceMatrix", (void(Problem::*)(const std::string &) const) & Problem::writeDistanceMatrix)
+    .def("writeClusters", &Problem::writeClusters)
+    .def("writeMedoidMembers", &Problem::writeMedoidMembers)
+    .def("writeSilhouettes", &Problem::writeSilhouettes)
+    .def("init", &Problem::init)
+    .def("cluster", &Problem::cluster)
+    .def("cluster_and_process", &Problem::cluster_and_process)
+    .def("findTotalCost", &Problem::findTotalCost)
+    .def("assignClusters", &Problem::assignClusters)
+    .def("calculateMedoids", &Problem::calculateMedoids);
+}
\ No newline at end of file
diff --git a/setup.py b/setup.py
new file mode 100644
index 0000000..d92f128
--- /dev/null
+++ b/setup.py
@@ -0,0 +1,139 @@
+import os
+import re
+import subprocess
+import sys
+from pathlib import Path
+
+from setuptools import Extension, setup
+from setuptools.command.build_ext import build_ext
+
+# Convert distutils Windows platform specifiers to CMake -A arguments
+PLAT_TO_CMAKE = {
+    "win32": "Win32",
+    "win-amd64": "x64",
+    "win-arm32": "ARM",
+    "win-arm64": "ARM64",
+}
+
+# A CMakeExtension needs a sourcedir instead of a file list.
+# The name must be the _single_ output extension from the CMake build.
+# If you need multiple extensions, see scikit-build.
+class CMakeExtension(Extension):
+    def __init__(self, name: str, sourcedir: str = "") -> None:
+        super().__init__(name, sources=[])
+        self.sourcedir = os.fspath(Path(sourcedir).resolve())
+
+
+class CMakeBuild(build_ext):
+    def build_extension(self, ext: CMakeExtension) -> None:
+        # Must be in this form due to bug in .resolve() only fixed in Python 3.10+
+        ext_fullpath = Path.cwd() / self.get_ext_fullpath(ext.name)
+        extdir = ext_fullpath.parent.resolve()
+
+        # Using this requires trailing slash for auto-detection & inclusion of
+        # auxiliary "native" libs
+
+        debug = int(os.environ.get("DEBUG", 0)) if self.debug is None else self.debug
+        cfg = "Debug" if debug else "Release"
+
+        # CMake lets you override the generator - we need to check this.
+        # Can be set with Conda-Build, for example.
+        cmake_generator = os.environ.get("CMAKE_GENERATOR", "")
+
+        # Set Python_EXECUTABLE instead if you use PYBIND11_FINDPYTHON
+        # EXAMPLE_VERSION_INFO shows you how to pass a value into the C++ code
+        # from Python.
+        cmake_args = [
+            f"-DCMAKE_LIBRARY_OUTPUT_DIRECTORY={extdir}{os.sep}",
+            f"-DPYTHON_EXECUTABLE={sys.executable}",
+            f"-DCMAKE_BUILD_TYPE={cfg}",  # not used on MSVC, but no harm
+        ]
+        build_args = []
+        # Adding CMake arguments set as environment variable
+        # (needed e.g. to build for ARM OSx on conda-forge)
+        if "CMAKE_ARGS" in os.environ:
+            cmake_args += [item for item in os.environ["CMAKE_ARGS"].split(" ") if item]
+
+        # In this example, we pass in the version to C++. You might not need to.
+        cmake_args += [f"-DEXAMPLE_VERSION_INFO={self.distribution.get_version()}"]
+
+        if self.compiler.compiler_type != "msvc":
+            # Using Ninja-build since it a) is available as a wheel and b)
+            # multithreads automatically. MSVC would require all variables be
+            # exported for Ninja to pick it up, which is a little tricky to do.
+            # Users can override the generator with CMAKE_GENERATOR in CMake
+            # 3.15+.
+            if not cmake_generator or cmake_generator == "Ninja":
+                try:
+                    import ninja
+
+                    ninja_executable_path = Path(ninja.BIN_DIR) / "ninja"
+                    cmake_args += [
+                        "-GNinja",
+                        f"-DCMAKE_MAKE_PROGRAM:FILEPATH={ninja_executable_path}",
+                    ]
+                except ImportError:
+                    pass
+
+        else:
+            # Single config generators are handled "normally"
+            single_config = any(x in cmake_generator for x in {"NMake", "Ninja"})
+
+            # CMake allows an arch-in-generator style for backward compatibility
+            contains_arch = any(x in cmake_generator for x in {"ARM", "Win64"})
+
+            # Specify the arch if using MSVC generator, but only if it doesn't
+            # contain a backward-compatibility arch spec already in the
+            # generator name.
+            if not single_config and not contains_arch:
+                cmake_args += ["-A", PLAT_TO_CMAKE[self.plat_name]]
+
+            # Multi-config generators have a different way to specify configs
+            if not single_config:
+                cmake_args += [
+                    f"-DCMAKE_LIBRARY_OUTPUT_DIRECTORY_{cfg.upper()}={extdir}"
+                ]
+                build_args += ["--config", cfg]
+
+        if sys.platform.startswith("darwin"):
+            # Cross-compile support for macOS - respect ARCHFLAGS if set
+            archs = re.findall(r"-arch (\S+)", os.environ.get("ARCHFLAGS", ""))
+            if archs:
+                cmake_args += ["-DCMAKE_OSX_ARCHITECTURES={}".format(";".join(archs))]
+
+        # Set CMAKE_BUILD_PARALLEL_LEVEL to control the parallel build level
+        # across all generators.
+        if "CMAKE_BUILD_PARALLEL_LEVEL" not in os.environ:
+            # self.parallel is a Python 3 only way to set parallel jobs by hand
+            # using -j in the build_ext call, not supported by pip or PyPA-build.
+            if hasattr(self, "parallel") and self.parallel:
+                # CMake 3.12+ only.
+                build_args += [f"-j{self.parallel}"]
+
+        build_temp = Path(self.build_temp) / ext.name
+        if not build_temp.exists():
+            build_temp.mkdir(parents=True)
+
+        subprocess.run(
+            ["cmake", ext.sourcedir, *cmake_args], cwd=build_temp, check=True
+        )
+        subprocess.run(
+            ["cmake", "--build", ".", *build_args], cwd=build_temp, check=True
+        )
+
+
+# The information here can also be placed in setup.cfg - better separation of
+# logic and declaration, and simpler if you include description/version in a file.
+setup(
+    name="dtwcpp",
+    version="0.1.0",
+    author="Volkan Kumtepeli",
+    author_email="volkan.kumtepeli@gmail.com",
+    description="A test project using pybind11 and CMake",
+    long_description="",
+    ext_modules=[CMakeExtension("dtwcpp")],
+    cmdclass={"build_ext": CMakeBuild},
+    zip_safe=False,
+    extras_require={"test": ["pytest>=6.0"]},
+    python_requires=">=3.7",
+)
\ No newline at end of file

From 9eb2bf7c225dba990287c7cbcca654b57fa67c82 Mon Sep 17 00:00:00 2001
From: Volkan Kumtepeli <volkan.kumtepeli@gmail.com>
Date: Tue, 19 Mar 2024 01:43:16 +0000
Subject: [PATCH 03/15] working on DataLoader and other classes.

---
 dtwc/scores.cpp    | 69 ++++++++++++++++++++++++++++++++++++++++++++++
 python/py_main.cpp | 21 +++++++++++++-
 test.py            |  6 ++++
 3 files changed, 95 insertions(+), 1 deletion(-)
 create mode 100644 test.py

diff --git a/dtwc/scores.cpp b/dtwc/scores.cpp
index 6b78cf6..2d1d418 100644
--- a/dtwc/scores.cpp
+++ b/dtwc/scores.cpp
@@ -82,4 +82,73 @@ std::vector<double> silhouette(Problem &prob)
   return silhouettes;
 }
 
+/**
+ * @brief Calculates the Davies-Bouldin index for a given clustering problem.
+ *
+ * The Davies-Bouldin index is a measure of the average similarity between clusters and the
+ * dissimilarity between clusters. It is used to evaluate the quality of a clustering solution,
+ * with a lower value indicating better separation between clusters.
+ *
+ * @param prob The clustering problem instance, which contains the data points, cluster indices, and centroids.
+ * @return double The Davies-Bouldin index.
+ *
+ * @note Requires that the data has already been clustered; if not, it will prompt the user to cluster the data first.
+ * @see https://en.wikipedia.org/wiki/Davies%E2%80%93Bouldin_index for more information on the Davies-Bouldin index.
+ */
+// double daviesBouldinIndex(Problem &prob)
+// {
+//   const auto Nc = prob.cluster_size(); //!< Number of clusters
+
+//   if (prob.centroids_ind.empty()) {
+//     std::cout << "Please cluster the data before calculating the Davies-Bouldin index!" << std::endl;
+//     return 0.0;
+//   }
+
+//   prob.fillDistanceMatrix(); //!< We need all pairwise distances for the Davies-Bouldin index.
+
+//   std::vector<double> clusterSimilarities(Nc, 0.0);    //!< Similarities between clusters
+//   std::vector<double> clusterDissimilarities(Nc, 0.0); //!< Dissimilarities between clusters
+
+//   // Calculate the similarity and dissimilarity for each cluster
+//   for (int i = 0; i < Nc; i++) {
+//     double maxSimilarity = std::numeric_limits<double>::lowest();
+
+//     for (int j = 0; j < Nc; j++) {
+//       if (i != j) {
+//         double similarity = (prob.distByInd(prob.centroids_ind[i], prob.centroids_ind[i]) + prob.distByInd(prob.centroids_ind[j], prob.centroids_ind[j])) / prob.distByInd(prob.centroids_ind[i], prob.centroids_ind[j]);
+
+//         if (similarity > maxSimilarity) {
+//           maxSimilarity = similarity;
+//         }
+//       }
+//     }
+
+//     clusterSimilarities[i] = maxSimilarity;
+//   }
+
+//   // Calculate the dissimilarity for each cluster
+//   for (int i = 0; i < Nc; i++) {
+//     double sumDissimilarity = 0.0;
+
+//     for (int j = 0; j < Nc; j++) {
+//       if (i != j) {
+//         sumDissimilarity += prob.distByInd(prob.centroids_ind[i], prob.centroids_ind[j]);
+//       }
+//     }
+
+//     clusterDissimilarities[i] = sumDissimilarity / (Nc - 1);
+//   }
+
+//   // Calculate the Davies-Bouldin index
+//   double daviesBouldinIndex = 0.0;
+
+//   for (int i = 0; i < Nc; i++) {
+//     daviesBouldinIndex += clusterSimilarities[i] + clusterDissimilarities[i];
+//   }
+
+//   daviesBouldinIndex /= Nc;
+
+//   return daviesBouldinIndex;
+// }
+
 } // namespace dtwc::scores
diff --git a/python/py_main.cpp b/python/py_main.cpp
index f237ade..f99601b 100644
--- a/python/py_main.cpp
+++ b/python/py_main.cpp
@@ -2,7 +2,7 @@
 #include <pybind11/stl.h>
 #include <pybind11/numpy.h>
 
-#include "../dtwc/Problem.hpp"
+#include "../dtwc/dtwc.hpp"
 
 #include <vector>
 #include <string>
@@ -47,4 +47,23 @@ PYBIND11_MODULE(dtwcpp, m)
     .def("findTotalCost", &Problem::findTotalCost)
     .def("assignClusters", &Problem::assignClusters)
     .def("calculateMedoids", &Problem::calculateMedoids);
+
+
+  py::class_<DataLoader>(m, "DataLoader")
+    .def(py::init<>())
+    .def(py::init<const std::filesystem::path &>())
+    .def(py::init<const std::filesystem::path &, int>())
+    .def("startColumn", (int(DataLoader::*)()) & DataLoader::startColumn)
+    .def("startRow", (int(DataLoader::*)()) & DataLoader::startRow)
+    .def("n_data", (int(DataLoader::*)()) & DataLoader::n_data)
+    .def("delimiter", (char(DataLoader::*)()) & DataLoader::delimiter)
+    .def("path", (std::filesystem::path(DataLoader::*)()) & DataLoader::path)
+    .def("verbosity", (int(DataLoader::*)()) & DataLoader::verbosity)
+    .def("startColumn", (DataLoader & (DataLoader::*)(int)) & DataLoader::startColumn)
+    .def("startRow", (DataLoader & (DataLoader::*)(int)) & DataLoader::startRow) // Added missing function signature
+    .def("n_data", &DataLoader::n_data)
+    .def("delimiter", &DataLoader::delimiter)
+    .def("path", &DataLoader::path)
+    .def("verbosity", &DataLoader::verbosity)
+    .def("load", &DataLoader::load);
 }
\ No newline at end of file
diff --git a/test.py b/test.py
new file mode 100644
index 0000000..4c97be3
--- /dev/null
+++ b/test.py
@@ -0,0 +1,6 @@
+import dtwcpp as dtwc
+
+
+a = dtwc.Problem()
+
+print(a.size())
\ No newline at end of file

From d627e7309fca4485c442169079362ef85fea00c0 Mon Sep 17 00:00:00 2001
From: Volkan Kumtepeli <volkan.kumtepeli@gmail.com>
Date: Tue, 26 Mar 2024 03:58:28 +0000
Subject: [PATCH 04/15] Update py_main.cpp

---
 python/py_main.cpp | 28 ++++++++++++++--------------
 1 file changed, 14 insertions(+), 14 deletions(-)

diff --git a/python/py_main.cpp b/python/py_main.cpp
index f99601b..2999791 100644
--- a/python/py_main.cpp
+++ b/python/py_main.cpp
@@ -52,18 +52,18 @@ PYBIND11_MODULE(dtwcpp, m)
   py::class_<DataLoader>(m, "DataLoader")
     .def(py::init<>())
     .def(py::init<const std::filesystem::path &>())
-    .def(py::init<const std::filesystem::path &, int>())
-    .def("startColumn", (int(DataLoader::*)()) & DataLoader::startColumn)
-    .def("startRow", (int(DataLoader::*)()) & DataLoader::startRow)
-    .def("n_data", (int(DataLoader::*)()) & DataLoader::n_data)
-    .def("delimiter", (char(DataLoader::*)()) & DataLoader::delimiter)
-    .def("path", (std::filesystem::path(DataLoader::*)()) & DataLoader::path)
-    .def("verbosity", (int(DataLoader::*)()) & DataLoader::verbosity)
-    .def("startColumn", (DataLoader & (DataLoader::*)(int)) & DataLoader::startColumn)
-    .def("startRow", (DataLoader & (DataLoader::*)(int)) & DataLoader::startRow) // Added missing function signature
-    .def("n_data", &DataLoader::n_data)
-    .def("delimiter", &DataLoader::delimiter)
-    .def("path", &DataLoader::path)
-    .def("verbosity", &DataLoader::verbosity)
-    .def("load", &DataLoader::load);
+    .def(py::init<const std::filesystem::path &, int>());
+  // .def("startColumn", (int(DataLoader::*)()) & DataLoader::startColumn)
+  // .def("startRow", (int(DataLoader::*)()) & DataLoader::startRow)
+  // .def("n_data", (int(DataLoader::*)()) & DataLoader::n_data)
+  // .def("delimiter", (char(DataLoader::*)()) & DataLoader::delimiter)
+  // .def("path", (std::filesystem::path(DataLoader::*)()) & DataLoader::path)
+  // .def("verbosity", (int(DataLoader::*)()) & DataLoader::verbosity)
+  // .def("startColumn", (DataLoader & (DataLoader::*)(int)) & DataLoader::startColumn)
+  // .def("startRow", (DataLoader & (DataLoader::*)(int)) & DataLoader::startRow)
+  // .def("n_data", &DataLoader::n_data)
+  // .def("delimiter", &DataLoader::delimiter)
+  // .def("path", &DataLoader::path)
+  // .def("verbosity", &DataLoader::verbosity)
+  // .def("load", &DataLoader::load);
 }
\ No newline at end of file

From e8adcbe96fb4c562b73a8b9865d71ade6dfb5051 Mon Sep 17 00:00:00 2001
From: Volkan Kumtepeli <volkan.kumtepeli@gmail.com>
Date: Sun, 28 Apr 2024 22:20:58 +0100
Subject: [PATCH 05/15] change actions v2 -> v4

---
 .github/workflows/documentation.yml |  2 +-
 .github/workflows/draft-pdf.yml     |  2 +-
 .github/workflows/macos-unit.yml    |  2 +-
 .github/workflows/ubuntu-unit.yml   | 11 ++++-------
 .github/workflows/windows-unit.yml  |  2 +-
 5 files changed, 8 insertions(+), 11 deletions(-)

diff --git a/.github/workflows/documentation.yml b/.github/workflows/documentation.yml
index e880a7c..f43e52c 100644
--- a/.github/workflows/documentation.yml
+++ b/.github/workflows/documentation.yml
@@ -18,7 +18,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Checkout current branch
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
       - name: Setup Pages
         uses: actions/configure-pages@v2
       - name: Build master with Jekyll
diff --git a/.github/workflows/draft-pdf.yml b/.github/workflows/draft-pdf.yml
index 588b9b0..f29083b 100644
--- a/.github/workflows/draft-pdf.yml
+++ b/.github/workflows/draft-pdf.yml
@@ -13,7 +13,7 @@ jobs:
     name: Paper Draft
     steps:
       - name: Checkout
-        uses: actions/checkout@v2
+        uses: actions/checkout@v4
       - name: Build draft PDF
         uses: openjournals/openjournals-draft-action@master
         with:
diff --git a/.github/workflows/macos-unit.yml b/.github/workflows/macos-unit.yml
index f25dc64..d11f730 100644
--- a/.github/workflows/macos-unit.yml
+++ b/.github/workflows/macos-unit.yml
@@ -16,7 +16,7 @@ jobs:
     runs-on: macos-latest
 
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       
       - name: install openmp
         run: brew reinstall libomp & brew link libomp --force
diff --git a/.github/workflows/ubuntu-unit.yml b/.github/workflows/ubuntu-unit.yml
index ee867cf..0b68764 100644
--- a/.github/workflows/ubuntu-unit.yml
+++ b/.github/workflows/ubuntu-unit.yml
@@ -42,14 +42,11 @@ jobs:
           - CC: clang-15
             CXX: clang++-15
             compiler: clang-15 libomp-15-dev
-          # - CC: clang-16  -> we don't have these as package. 
-          #   CXX: clang++-16
-          #   compiler: clang-16
-          # - CC: clang-17
-          #   CXX: clang++-17
-          #   compiler: clang-17
+          - CC: clang-16
+            CXX: clang++-16
+            compiler: clang-16 libomp-16-dev
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
 
       - name: update
         run: sudo apt update
diff --git a/.github/workflows/windows-unit.yml b/.github/workflows/windows-unit.yml
index c9b0da4..a4278ea 100644
--- a/.github/workflows/windows-unit.yml
+++ b/.github/workflows/windows-unit.yml
@@ -16,7 +16,7 @@ jobs:
     runs-on: windows-latest
 
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
 
       - name: make build directory
         run: mkdir build_dir

From 1e1bc3752f695ebdfdad5d30583e086f2a5728d6 Mon Sep 17 00:00:00 2001
From: Volkan Kumtepeli <volkan.kumtepeli@gmail.com>
Date: Sun, 28 Apr 2024 22:24:26 +0100
Subject: [PATCH 06/15] Change versioning

---
 CHANGELOG.md           | 11 +++++------
 develop/TODO.md        | 10 ++++++----
 develop/conventions.md |  9 +++++++++
 3 files changed, 20 insertions(+), 10 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 8bb963d..02951a3 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -6,13 +6,12 @@ This changelog contains a non-exhaustive list of new features and notable bug-fi
 
 
 <br/><br/>
-# DTWC v0.1.0
+# DTWC v1.0.0
 
 ## New features
-* HiGHS solver is added for open-source alternative to Gurobi. 
+* HiGHS solver is added for open-source alternative to Gurobi (which is now not necessary for compilation and can be enabled by necessary flags). 
 * Command line interface is added. 
-* Documentation is improved.
-
+* Documentation is improved (Doxygen website).
 
 
 ## Notable Bug-fixes
@@ -44,9 +43,9 @@ This changelog contains a non-exhaustive list of new features and notable bug-fi
 * `std::filesystem::path operator+` was unnecessary and removed. 
 
 <br/><br/>
-# DTWC v0.0.2
+# DTWC v0.1.0
 
-DTWC with a better interface for other people's use. 
+A user interface is created for other people's use. 
 
 ## New features / updates
 - Scores file with silhouette score is added. 
diff --git a/develop/TODO.md b/develop/TODO.md
index 58b1d55..a3524af 100644
--- a/develop/TODO.md
+++ b/develop/TODO.md
@@ -8,13 +8,15 @@
 
 ### Current priority: 
 
+- [ ] Numpy-like arrays for multi-dimensional DTW.
 - [ ] If you accidentally compile with amd64_x86 you get bad_alloc due to the insufficient RAM in 32-bit computer. So, put a warning for that. 
 - [ ] Modernise `*.csv` reading interface, allow for rowwise or columnwise data. 
 - [x] CMake files arranging:
   - [x] HiGHS 1.6.0 does not always work. Also it does not work in debug mode due to iterator level. 
 - [ ] Commenting
 - [ ] MATLAB / Python integration
-- [ ] JOSS paper  
+ - [ ] Pybind11 usage for Python integration.
+- [x] JOSS paper  
 - [ ] Speed up the code
 - [ ] GPU programming 
 - [ ] Lighter and faster DTW cost calculation + make the band from long side so it is more accurate. 
@@ -26,8 +28,8 @@
 - [ ] Give error message if data could not be loaded. 
 - Benchmarking:
   - [x] UCR_test_2018 is continuing. 
-  - [ ] USR_dtai.py
-  - [ ] TSlearn 
+  - [x] USR_dtai.py
+  - [x] TSlearn 
   - [ ] dtwclust in R
 - [ ] Encapsulating Data and related functions in one folder. 
 - [ ] Open-source solver addition. 
@@ -38,7 +40,7 @@
 - [x] Doxygen website? 
 - [x] Remove unnecessary warping functions. 
 - [ ] Fix warnings. Especially, we should not get warnings from external libraries. 
-- [ ] Make Gurobi dependency optional. (Now it cannot be disabled. )
+- [x] Make Gurobi dependency optional. (Now it cannot be disabled. )
 - [x] Consider including Eigen library for matrix operations / linear system solution. 
 - [ ] w based DTW. 
 - [ ] Reading memoisation matrix distMat from file instead of re-calculating DTW every time. 
diff --git a/develop/conventions.md b/develop/conventions.md
index 8331cd7..e32c2e3 100644
--- a/develop/conventions.md
+++ b/develop/conventions.md
@@ -3,6 +3,15 @@
 - Do not use "using" in global namespace.
 - First include standard headers then our headers.
 
+## Versioning: 
+
+We use semantic versioning with MAJOR.MINOR.PATCH where:
+
+- PATCH: Backwards compatible bug fix without introducing a new feature. 
+- MINOR: Backwards compatible new feature. 
+- MAJOR: Incompatible changes, especially on API. 
+
+See [this link](https://www.youtube.com/watch?v=xvPiZyx0cDc) for a detailed tutorial on versioning. 
 
 ### Commenting: 
 

From c88a2b98b411b7930955ee5acc6db64a4419865c Mon Sep 17 00:00:00 2001
From: Volkan Kumtepeli <volkan.kumtepeli@gmail.com>
Date: Sun, 28 Apr 2024 22:40:43 +0100
Subject: [PATCH 07/15] Update ubuntu-unit.yml

---
 .github/workflows/ubuntu-unit.yml | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/.github/workflows/ubuntu-unit.yml b/.github/workflows/ubuntu-unit.yml
index 0b68764..d7e9cfe 100644
--- a/.github/workflows/ubuntu-unit.yml
+++ b/.github/workflows/ubuntu-unit.yml
@@ -1,16 +1,14 @@
-name: Ubuntu unit
-# adapted from: https://github.com/trase-cpp/trase/tree/master/.github/workflows
+name: Ubuntu unit # adapted from: https://github.com/trase-cpp/trase/tree/master/.github/workflows
 on:
   push:
     branches:
-   #   - main
+      # - main
       - develop
   pull_request:
     branches:
       - '**'
 
 jobs:
-
   build-and-test:
     name: Ubuntu unit ${{matrix.CC}}
     runs-on: ubuntu-latest
@@ -45,12 +43,22 @@ jobs:
           - CC: clang-16
             CXX: clang++-16
             compiler: clang-16 libomp-16-dev
+          - CC: clang-17
+            CXX: clang++-17
+            compiler: clang-17 libomp-17-dev
     steps:
       - uses: actions/checkout@v4
 
       - name: update
         run: sudo apt update
 
+      - name: Install Clang 16 or 17
+        if: matrix.CC == 'clang-16' || matrix.CC == 'clang-17'
+        run: |
+          wget https://apt.llvm.org/llvm.sh
+          chmod +x llvm.sh
+          sudo ./llvm.sh $(echo "${{ matrix.CC }}" | cut -d '-' -f 2)
+
       - name: install compiler
         run: sudo apt install ${{matrix.compiler}}
 

From 451f6dc2c4f587178cd4b2039df082cfc8148e9f Mon Sep 17 00:00:00 2001
From: Volkan Kumtepeli <volkan.kumtepeli@gmail.com>
Date: Sun, 28 Apr 2024 23:11:37 +0100
Subject: [PATCH 08/15] Remove force link from macOS as unit test does not
 work.

---
 .github/workflows/macos-unit.yml | 22 ++++++++++++----------
 1 file changed, 12 insertions(+), 10 deletions(-)

diff --git a/.github/workflows/macos-unit.yml b/.github/workflows/macos-unit.yml
index d11f730..493cb9b 100644
--- a/.github/workflows/macos-unit.yml
+++ b/.github/workflows/macos-unit.yml
@@ -3,14 +3,13 @@ name: macOS unit
 on:
   push:
     branches:
-    #  - main
+      # - main
       - develop
   pull_request:
     branches:
       - '**'
 
 jobs:
-
   build-and-test:
     name: macOS unit
     runs-on: macos-latest
@@ -18,20 +17,23 @@ jobs:
     steps:
       - uses: actions/checkout@v4
       
-      - name: install openmp
-        run: brew reinstall libomp & brew link libomp --force
-      
-      - name: make build directory
+      - name: Install libomp without linking
+        run: |
+          brew reinstall libomp
+          echo "export LDFLAGS=-L/opt/homebrew/opt/libomp/lib" >> $GITHUB_ENV
+          echo "export CPPFLAGS=-I/opt/homebrew/opt/libomp/include" >> $GITHUB_ENV
+
+      - name: Make build directory
         run: mkdir build_dir
 
-      - name: cmake configure
+      - name: CMake configure
         run: cmake .. -DCMAKE_BUILD_TYPE=Debug -DDTWC_BUILD_TESTING=ON
         working-directory: build_dir
 
-      - name: cmake build
+      - name: CMake build
         run: cmake --build . --parallel 2
         working-directory: build_dir
 
-      - name: cmake test
+      - name: CMake test
         run: ctest -j2 -C Debug --output-on-failure
-        working-directory: build_dir
+        working-directory: build_dir
\ No newline at end of file

From 6e6834535786f9666d00df1ac4df708657821a7f Mon Sep 17 00:00:00 2001
From: Volkan Kumtepeli <volkan.kumtepeli@gmail.com>
Date: Sun, 28 Apr 2024 23:17:12 +0100
Subject: [PATCH 09/15] export omp flags

---
 .github/workflows/macos-unit.yml | 19 ++++++++++++++-----
 1 file changed, 14 insertions(+), 5 deletions(-)

diff --git a/.github/workflows/macos-unit.yml b/.github/workflows/macos-unit.yml
index 493cb9b..02f2ebf 100644
--- a/.github/workflows/macos-unit.yml
+++ b/.github/workflows/macos-unit.yml
@@ -20,20 +20,29 @@ jobs:
       - name: Install libomp without linking
         run: |
           brew reinstall libomp
-          echo "export LDFLAGS=-L/opt/homebrew/opt/libomp/lib" >> $GITHUB_ENV
-          echo "export CPPFLAGS=-I/opt/homebrew/opt/libomp/include" >> $GITHUB_ENV
+          echo "LDFLAGS=-L/opt/homebrew/opt/libomp/lib" >> $GITHUB_ENV
+          echo "CPPFLAGS=-I/opt/homebrew/opt/libomp/include" >> $GITHUB_ENV
 
       - name: Make build directory
         run: mkdir build_dir
 
       - name: CMake configure
-        run: cmake .. -DCMAKE_BUILD_TYPE=Debug -DDTWC_BUILD_TESTING=ON
+        run: |
+          export LDFLAGS=-L/opt/homebrew/opt/libomp/lib
+          export CPPFLAGS=-I/opt/homebrew/opt/libomp/include
+          cmake .. -DCMAKE_BUILD_TYPE=Debug -DDTWC_BUILD_TESTING=ON
         working-directory: build_dir
 
       - name: CMake build
-        run: cmake --build . --parallel 2
+        run: |
+          export LDFLAGS=-L/opt/homebrew/opt/libomp/lib
+          export CPPFLAGS=-I/opt/homebrew/opt/libomp/include
+          cmake --build . --parallel 2
         working-directory: build_dir
 
       - name: CMake test
-        run: ctest -j2 -C Debug --output-on-failure
+        run: |
+          export LDFLAGS=-L/opt/homebrew/opt/libomp/lib
+          export CPPFLAGS=-I/opt/homebrew/opt/libomp/include
+          ctest -j2 -C Debug --output-on-failure
         working-directory: build_dir
\ No newline at end of file

From cdc9d7e664e04e9497783ffc7cbbbf7c8a1c582f Mon Sep 17 00:00:00 2001
From: Volkan Kumtepeli <volkan.kumtepeli@gmail.com>
Date: Sun, 28 Apr 2024 23:39:21 +0100
Subject: [PATCH 10/15] export flags for libomp

---
 .github/workflows/macos-unit.yml | 17 ++++++++---------
 1 file changed, 8 insertions(+), 9 deletions(-)

diff --git a/.github/workflows/macos-unit.yml b/.github/workflows/macos-unit.yml
index 02f2ebf..d4cface 100644
--- a/.github/workflows/macos-unit.yml
+++ b/.github/workflows/macos-unit.yml
@@ -19,30 +19,29 @@ jobs:
       
       - name: Install libomp without linking
         run: |
-          brew reinstall libomp
-          echo "LDFLAGS=-L/opt/homebrew/opt/libomp/lib" >> $GITHUB_ENV
-          echo "CPPFLAGS=-I/opt/homebrew/opt/libomp/include" >> $GITHUB_ENV
+          xcode-select --install 
+          brew install libomp
+          export CC=/usr/bin/clang
+          export CXX=/usr/bin/clang++
+          export CPPFLAGS="$CPPFLAGS -Xpreprocessor -fopenmp"
+          export CFLAGS="$CFLAGS -I/usr/local/opt/libomp/include"
+          export CXXFLAGS="$CXXFLAGS -I/usr/local/opt/libomp/include"
+          export LDFLAGS="$LDFLAGS -Wl,-rpath,/usr/local/opt/libomp/lib -L/usr/local/opt/libomp/lib -lomp"
 
       - name: Make build directory
         run: mkdir build_dir
 
       - name: CMake configure
         run: |
-          export LDFLAGS=-L/opt/homebrew/opt/libomp/lib
-          export CPPFLAGS=-I/opt/homebrew/opt/libomp/include
           cmake .. -DCMAKE_BUILD_TYPE=Debug -DDTWC_BUILD_TESTING=ON
         working-directory: build_dir
 
       - name: CMake build
         run: |
-          export LDFLAGS=-L/opt/homebrew/opt/libomp/lib
-          export CPPFLAGS=-I/opt/homebrew/opt/libomp/include
           cmake --build . --parallel 2
         working-directory: build_dir
 
       - name: CMake test
         run: |
-          export LDFLAGS=-L/opt/homebrew/opt/libomp/lib
-          export CPPFLAGS=-I/opt/homebrew/opt/libomp/include
           ctest -j2 -C Debug --output-on-failure
         working-directory: build_dir
\ No newline at end of file

From 29ec4ac179028dc208a1cbb79f7ea27be79b2cda Mon Sep 17 00:00:00 2001
From: Volkan Kumtepeli <volkan.kumtepeli@gmail.com>
Date: Sun, 28 Apr 2024 23:46:23 +0100
Subject: [PATCH 11/15] removed xcode-select --install

---
 .github/workflows/macos-unit.yml | 1 -
 1 file changed, 1 deletion(-)

diff --git a/.github/workflows/macos-unit.yml b/.github/workflows/macos-unit.yml
index d4cface..67a0cc2 100644
--- a/.github/workflows/macos-unit.yml
+++ b/.github/workflows/macos-unit.yml
@@ -19,7 +19,6 @@ jobs:
       
       - name: Install libomp without linking
         run: |
-          xcode-select --install 
           brew install libomp
           export CC=/usr/bin/clang
           export CXX=/usr/bin/clang++

From 0aa61881b264566b20429815986c97bf617e630d Mon Sep 17 00:00:00 2001
From: Volkan Kumtepeli <volkan.kumtepeli@gmail.com>
Date: Mon, 29 Apr 2024 02:15:46 +0100
Subject: [PATCH 12/15] macos unit test libomp

---
 .github/dependabot.yml           |  9 +++++++++
 .github/workflows/macos-unit.yml | 10 ++++------
 2 files changed, 13 insertions(+), 6 deletions(-)
 create mode 100644 .github/dependabot.yml

diff --git a/.github/dependabot.yml b/.github/dependabot.yml
new file mode 100644
index 0000000..489b692
--- /dev/null
+++ b/.github/dependabot.yml
@@ -0,0 +1,9 @@
+# https://docs.github.com/github/administering-a-repository/configuration-options-for-dependency-updates
+
+version: 2
+updates:
+
+  - package-ecosystem: "github-actions"
+    directory: "/"
+    schedule:
+      interval: "weekly"
\ No newline at end of file
diff --git a/.github/workflows/macos-unit.yml b/.github/workflows/macos-unit.yml
index 67a0cc2..ac9233d 100644
--- a/.github/workflows/macos-unit.yml
+++ b/.github/workflows/macos-unit.yml
@@ -20,12 +20,10 @@ jobs:
       - name: Install libomp without linking
         run: |
           brew install libomp
-          export CC=/usr/bin/clang
-          export CXX=/usr/bin/clang++
-          export CPPFLAGS="$CPPFLAGS -Xpreprocessor -fopenmp"
-          export CFLAGS="$CFLAGS -I/usr/local/opt/libomp/include"
-          export CXXFLAGS="$CXXFLAGS -I/usr/local/opt/libomp/include"
-          export LDFLAGS="$LDFLAGS -Wl,-rpath,/usr/local/opt/libomp/lib -L/usr/local/opt/libomp/lib -lomp"
+          echo "CC=$(brew --prefix llvm)/bin/clang" >> $GITHUB_ENV
+          echo "CXX=$(brew --prefix llvm)/bin/clang++" >> $GITHUB_ENV
+          echo "LDFLAGS=\"-L$(brew --prefix llvm)/lib\"" >> $GITHUB_ENV
+          echo "CPPFLAGS=\"-I$(brew --prefix llvm)/include\"" >> $GITHUB_ENV
 
       - name: Make build directory
         run: mkdir build_dir

From ed0c94c2f8fc247d5d03043f17184a12a6c4002a Mon Sep 17 00:00:00 2001
From: Volkan Kumtepeli <volkan.kumtepeli@gmail.com>
Date: Mon, 29 Apr 2024 02:25:52 +0100
Subject: [PATCH 13/15] Update macos-unit.yml

---
 .github/workflows/macos-unit.yml | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/.github/workflows/macos-unit.yml b/.github/workflows/macos-unit.yml
index ac9233d..3f1e02b 100644
--- a/.github/workflows/macos-unit.yml
+++ b/.github/workflows/macos-unit.yml
@@ -19,11 +19,7 @@ jobs:
       
       - name: Install libomp without linking
         run: |
-          brew install libomp
-          echo "CC=$(brew --prefix llvm)/bin/clang" >> $GITHUB_ENV
-          echo "CXX=$(brew --prefix llvm)/bin/clang++" >> $GITHUB_ENV
-          echo "LDFLAGS=\"-L$(brew --prefix llvm)/lib\"" >> $GITHUB_ENV
-          echo "CPPFLAGS=\"-I$(brew --prefix llvm)/include\"" >> $GITHUB_ENV
+          brew install libomp llvm && brew link --force libomp
 
       - name: Make build directory
         run: mkdir build_dir

From 653710daa34d6b35ba12ff265172cd89c9bc0c88 Mon Sep 17 00:00:00 2001
From: Volkan Kumtepeli <volkan.kumtepeli@gmail.com>
Date: Mon, 29 Apr 2024 02:44:19 +0100
Subject: [PATCH 14/15] UCR dataset all benchmark is enabled.

---
 CHANGELOG.md           | 7 ++++---
 benchmark/UCR_dtwc.cpp | 7 ++++---
 2 files changed, 8 insertions(+), 6 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 02951a3..321b4bf 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -26,9 +26,10 @@ This changelog contains a non-exhaustive list of new features and notable bug-fi
 
 ## Developer updates: 
 * The software is now being tested via Catch2 library. 
+* Dependabot is added. 
 
 <br/><br/>
-# DTWC v0.0.3
+# DTWC v0.3.0
 
 ## New features
 * UCR_test_2018 data integration for benchmarking. 
@@ -43,7 +44,7 @@ This changelog contains a non-exhaustive list of new features and notable bug-fi
 * `std::filesystem::path operator+` was unnecessary and removed. 
 
 <br/><br/>
-# DTWC v0.1.0
+# DTWC v0.2.0
 
 A user interface is created for other people's use. 
 
@@ -67,7 +68,7 @@ A user interface is created for other people's use.
   * Required C++ standard is upgraded from C++17 to C++20. 
 
 <br/><br/>
-# DTWC v0.0.1
+# DTWC v0.1.0
 
 This is the initial release of DTWC. 
 
diff --git a/benchmark/UCR_dtwc.cpp b/benchmark/UCR_dtwc.cpp
index 429f0df..3aa0ced 100644
--- a/benchmark/UCR_dtwc.cpp
+++ b/benchmark/UCR_dtwc.cpp
@@ -97,6 +97,9 @@ inline void UCR_2018()
     //(settings::root_folder / "data/benchmark/UCRArchive_2018/AllGestureWiimoteZ/AllGestureWiimoteZ_TEST.tsv")
 
   };
+
+  dataofInterest = UCR_list; // Comment this out to do an individual testing.
+
   size_t solved = 0;
   for (auto &file_path : dataofInterest) {
     dl.path(file_path);
@@ -108,10 +111,8 @@ inline void UCR_2018()
 
     std::cout << "Now, number " << solved << " " << file_path << " is being solved.\n";
     solved++;
-    // if (solved < 25) // We already calculated this part
-    //   continue;
 
-    if (prob.data.size() > 1000) // DOnt calculate large data it is not good. For example Crop.
+    if (prob.data.size() > 1000) // Don't calculate large data it is not good. For example Crop.
       continue;
 
     prob.set_numberOfClusters(Nc); // Nc = number of clusters.

From e14e072b086eaeeac050f2a8c579da0db4a11bd7 Mon Sep 17 00:00:00 2001
From: Volkan Kumtepeli <volkan.kumtepeli@gmail.com>
Date: Mon, 29 Apr 2024 02:46:35 +0100
Subject: [PATCH 15/15] version changes.

---
 CMakeLists.txt | 2 +-
 docs/Doxyfile  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/CMakeLists.txt b/CMakeLists.txt
index 023b32d..0e9c3fe 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -6,7 +6,7 @@ cmake_minimum_required(VERSION 3.21)
 # so that it works well with tools like CPM or other
 # manual dependency management
 
-project(DTWC++ VERSION "0.1.0"
+project(DTWC++ VERSION "1.0.0"
   DESCRIPTION "A C++ library for fast Dynamic Time Wrapping Clustering"
   HOMEPAGE_URL https://battery-intelligence-lab.github.io/dtw-cpp/
   LANGUAGES CXX C)
diff --git a/docs/Doxyfile b/docs/Doxyfile
index 3e7d5e7..2472419 100644
--- a/docs/Doxyfile
+++ b/docs/Doxyfile
@@ -48,7 +48,7 @@ PROJECT_NAME           = DTWC++
 # could be handy for archiving the generated documentation or if some version
 # control system is used.
 
-PROJECT_NUMBER         = 0.1.0
+PROJECT_NUMBER         = 1.0.0-rc0
 
 # Using the PROJECT_BRIEF tag one can provide an optional one line description
 # for a project that appears at the top of each page and should give viewer a