BanditPAM v6.0.2 -- Fixing merged PRs

.github/workflows/linux_build_wheels.yml

@@ -78,7 +78,7 @@ jobs:
         with:
           user: __token__
           password: ${{ secrets.TEST_PYPI_APIKEY }}
-          repository_url: https://test.pypi.org/legacy/
+          repository-url: https://test.pypi.org/legacy/
 
       - name: Upload to PyPI on published release
         if: ${{ github.event_name == 'release' && github.event.action == 'published' }}

.github/workflows/mac_build_wheels.yml

@@ -59,7 +59,7 @@ jobs:
         with:
           user: __token__
           password: ${{ secrets.TEST_PYPI_APIKEY }}
-          repository_url: https://test.pypi.org/legacy/
+          repository-url: https://test.pypi.org/legacy/
 
       - name: Upload to PyPI on published release
         if: ${{ github.event_name == 'release' && github.event.action == 'published' }}

R_package/banditpam/src/kmedoids_algorithm.hpp

@@ -12,6 +12,25 @@
 #include <string>
 
 namespace km {
+
+/**
+ * @brief Enum for different loss function types
+ */
+enum class LossType {
+  MANHATTAN,
+  COS,
+  COSINE,
+  INF,
+  EUCLIDEAN,
+  LP_NORM,
+  UNKNOWN
+};
+
+/**
+ * @brief Enum for different distance categories
+ */
+enum class AlgorithmStep { MISC, BUILD, SWAP };
+
 /**
  * @brief KMedoids class. Creates a KMedoids object that can be used to find the medoids
  * for a particular set of input data.
@@ -462,6 +481,30 @@ class KMedoids {
    */
   void checkAlgorithm(const std::string& algorithm) const;
 
+    /**
+   * @brief Converts a string loss function name to LossType enum
+   *
+   * @param loss The loss function string
+   * @returns The corresponding LossType enum value
+   */
+  LossType getLossType(const std::string &loss) const {
+    if (loss == "manhattan") {
+      return LossType::MANHATTAN;
+    } else if (loss == "cos") {
+      return LossType::COS;
+    } else if (loss == "cosine") {
+      return LossType::COSINE;
+    } else if (loss == "inf") {
+      return LossType::INF;
+    } else if (loss == "euclidean") {
+      return LossType::EUCLIDEAN;
+    } else if (std::regex_match(loss, std::regex("l\\d*"))) {
+      return LossType::LP_NORM;
+    } else {
+      return LossType::UNKNOWN;
+    }
+  }
+
   /// Number of medoids to use -- the "k" in k-medoids
   size_t nMedoids;
 

headers/algorithms/kmedoids_algorithm.hpp

@@ -31,11 +31,7 @@ enum class LossType {
 /**
  * @brief Enum for different distance categories
  */
-enum class AlgorithmStep {
-  MISC,
-  BUILD,
-  SWAP
-};
+enum class AlgorithmStep { MISC, BUILD, SWAP };
 
 /**
  * @brief KMedoids class. Creates a KMedoids object that can be used to find the
@@ -468,17 +464,18 @@ class KMedoids {
    * @returns The Manhattan distance between points i and j
    */
   float manhattan(const arma::fmat &data, const size_t i, const size_t j) const;
-    
+
   /**
    * @brief Assigns ranks to each element in the input vector.
    * Smallest element receives rank 1, the next 2, and so on.
    * Ties are assigned the average of the ranks they would encompass.
    *
    * @param vec A vector containing the elements to be ranked.
    *
-   * @returns A vector of the same size as `vec`, with each element replaced by its rank.
+   * @returns A vector of the same size as `vec`, with each element replaced by
+   * its rank.
    */
-  arma::fvec rank(const arma::fvec& vec) const;
+  arma::fvec rank(const arma::fvec &vec) const;
 
   /**
    * @brief Computes the Pearson correlation between the
@@ -491,7 +488,8 @@ class KMedoids {
    * @returns The Pearson correlation between points i and j
    */
   float pearson(const arma::fmat &data, const size_t i, const size_t j) const;
-  float clippedCos(const arma::fmat &data, const size_t i, const size_t j) const;
+  float clippedCos(const arma::fmat &data, const size_t i,
+                   const size_t j) const;
 
   /**
    * @brief Computes the Spearman correlation between the
@@ -517,7 +515,7 @@ class KMedoids {
 
   /**
    * @brief Converts a string loss function name to LossType enum
-   * 
+   *
    * @param loss The loss function string
    * @returns The corresponding LossType enum value
    */

pyproject.toml

@@ -1,5 +1,5 @@
 [project]
-license = {text = "MIT"}
+license = "MIT"
 version = "6.0.2"
 name = "banditpam"
 requires-python = ">= 3.10"

readme_ex1.py

@@ -5,7 +5,7 @@
 # Generate data from a Gaussian Mixture Model with the given means:
 np.random.seed(0)
 n_per_cluster = 40
-means = np.array([[0,0], [-5,5], [5,5]])
+means = np.array([[0, 0], [-5, 5], [5, 5]])
 X = np.vstack([np.random.randn(n_per_cluster, 2) + mu for mu in means])
 
 # Fit the data with BanditPAM:
@@ -18,9 +18,8 @@
 # Visualize the data and the medoids:
 for p_idx, point in enumerate(X):
     if p_idx in map(int, kmed.medoids):
-        plt.scatter(X[p_idx, 0], X[p_idx, 1], color='red', s = 40)
+        plt.scatter(X[p_idx, 0], X[p_idx, 1], color='red', s=40)
     else:
-        plt.scatter(X[p_idx, 0], X[p_idx, 1], color='blue', s = 10)
+        plt.scatter(X[p_idx, 0], X[p_idx, 1], color='blue', s=10)
 
 plt.show()
-

requirements.txt

@@ -1,5 +1,5 @@
 pandas>=0.24.1
-pybind11>=2.5.0
+pybind11>=3.0.0
 numpy>=1.16.2
 matplotlib>=3.2.1
 myst-parser>=0.16.0

setup.py

@@ -447,7 +447,8 @@ def build_extensions(self):  # noqa: C901
                 for package in ["libomp"]:
                     package_prefix = get_package_prefix(package)
                     comp_opts.append("-I{}/include".format(package_prefix))
-                    comp_opts.append("-L{}/lib".format(package_prefix))  # Remove?
+                    # Remove addition to comp_opts?
+                    comp_opts.append("-L{}/lib".format(package_prefix))
                     link_opts.append("-L{}/lib".format(package_prefix))
         elif sys.platform == "linux":
             link_opts += ["-lm", "-lpthread"]
@@ -456,10 +457,10 @@ def build_extensions(self):  # noqa: C901
 
         compiler_name = compiler_check()
         # if sys.platform == "darwin":
-            # if compiler_name == "gcc":
-            #      link_opts.append("-lomp")
-            # else:
-            #     link_opts.append("-lomp")
+        #     if compiler_name == "gcc":
+        #         link_opts.append("-lomp")
+        #     else:
+        #         link_opts.append("-lomp")
         if sys.platform == "linux" or sys.platform == "linux2":
             if compiler_name == "gcc":
                 link_opts += ["-lgomp", "-lm", "-lpthread"]
@@ -488,8 +489,8 @@ def build_extensions(self):  # noqa: C901
             ext.extra_link_args = link_opts
             ext.extra_link_args += [
                 "-v",
-            ] # "-arch", "x86_64"]
-            ext.extra_link_args =  link_opts
+            ]  # "-arch", "x86_64"]
+            ext.extra_link_args = link_opts
 
         build_ext.build_extensions(self)
 
@@ -502,11 +503,9 @@ def main():
             "headers",
             os.path.join("headers", "algorithms"),
             os.path.join("headers", "python_bindings"),
-            os.path.join("../", "carma", "include"),
-            # os.path.join("headers", "carma", "include"),
-            # os.path.join("headers", "carma", "include", "carma_bits"),
+            os.path.join("headers", "carma", "include"),
+            os.path.join("headers", "carma", "include", "carma_bits"),
             os.path.join("/", "usr", "local", "include"),
-
         ]
     elif sys.platform == "darwin":  # OSX
         include_dirs = [
@@ -562,8 +561,8 @@ def main():
 
     compiler_name = compiler_check()
     if sys.platform == "darwin" and os.environ.get(GHA, False):
-        # Do NOT link omp here because it'll add an -lomp flag to dynamically link it
-        #  (and we want to statically link it)
+        # Do NOT link omp here because it'll add an -lomp flag to
+        #  dynamically link it (and we want to statically link it)
         libraries = ["armadillo"]
     elif sys.platform == "win32":
         libraries = ["libopenblas"]
@@ -612,7 +611,6 @@ def main():
                 os.path.join(
                     "src", "python_bindings", "swap_times_python.cpp"
                 ),
-
             ],
             include_dirs=include_dirs,
             library_dirs=library_dirs,
@@ -651,7 +649,7 @@ def main():
         url="https://github.com/motiwari/BanditPAM",
         long_description=long_description,
         ext_modules=ext_modules,
-        setup_requires=["pybind11>=2.5.0", "numpy>=1.18"],
+        setup_requires=["pybind11>=3.0.0", "numpy>=1.18"],
         data_files=my_data_files,
         include_package_data=True,
         cmdclass={"build_ext": BuildExt},

src/algorithms/banditpam.cpp

@@ -87,8 +87,8 @@ arma::frowvec BanditPAM::buildSigma(
 #pragma omp parallel for if (this->parallelize)
   for (size_t i = 0; i < N; i++) {
     for (size_t j = 0; j < batchSize; j++) {
-      float cost =
-        KMedoids::cachedLoss(data, distMat, i, referencePoints(j), AlgorithmStep::MISC);
+      float cost = KMedoids::cachedLoss(data, distMat, i, referencePoints(j),
+                                        AlgorithmStep::MISC);
       if (useAbsolute) {
         sample(j) = cost;
       } else {
@@ -136,9 +136,8 @@ arma::frowvec BanditPAM::buildTarget(
   for (size_t i = 0; i < target->n_rows; i++) {
     float total = 0;
     for (size_t j = 0; j < referencePoints.n_rows; j++) {
-      float cost =
-        KMedoids::cachedLoss(data, distMat, (*target)(i), referencePoints(j),
-                             AlgorithmStep::BUILD);
+      float cost = KMedoids::cachedLoss(
+        data, distMat, (*target)(i), referencePoints(j), AlgorithmStep::BUILD);
       if (useAbsolute) {
         total += cost;
       } else {
@@ -286,8 +285,8 @@ arma::fmat BanditPAM::swapSigma(
 
     // calculate change in loss for some subset of the data
     for (size_t j = 0; j < batchSize; j++) {
-      float cost =
-        KMedoids::cachedLoss(data, distMat, n, referencePoints(j), AlgorithmStep::MISC);
+      float cost = KMedoids::cachedLoss(data, distMat, n, referencePoints(j),
+                                        AlgorithmStep::MISC);
 
       if (k == (*assignments)(referencePoints(j))) {
         if (cost < (*secondBestDistances)(referencePoints(j))) {
@@ -365,9 +364,8 @@ arma::fmat BanditPAM::swapTarget(
   for (size_t i = 0; i < T; i++) {
     // TODO(@motiwari): pragma omp parallel for?
     for (size_t j = 0; j < tmpBatchSize; j++) {
-      float cost =
-        KMedoids::cachedLoss(data, distMat, (*targets)(i), referencePoints(j),
-                             AlgorithmStep::SWAP);
+      float cost = KMedoids::cachedLoss(
+        data, distMat, (*targets)(i), referencePoints(j), AlgorithmStep::SWAP);
       size_t k = (*assignments)(referencePoints(j));
       if (cost < (*bestDistances)(referencePoints(j))) {
         // We might be able to change this to

src/algorithms/banditpam_orig.cpp

@@ -89,8 +89,8 @@ arma::frowvec BanditPAM_orig::buildSigma(
   for (size_t i = 0; i < N; i++) {
     for (size_t j = 0; j < batchSize; j++) {
       // 0 for MISC
-      float cost =
-        KMedoids::cachedLoss(data, distMat, i, referencePoints(j), km::AlgorithmStep::MISC);
+      float cost = KMedoids::cachedLoss(data, distMat, i, referencePoints(j),
+                                        km::AlgorithmStep::MISC);
       if (useAbsolute) {
         sample(j) = cost;
       } else {
@@ -138,9 +138,8 @@ arma::frowvec BanditPAM_orig::buildTarget(
   for (size_t i = 0; i < target->n_rows; i++) {
     float total = 0;
     for (size_t j = 0; j < referencePoints.n_rows; j++) {
-      float cost =
-        KMedoids::cachedLoss(data, distMat, (*target)(i), referencePoints(j),
-                             AlgorithmStep::BUILD);
+      float cost = KMedoids::cachedLoss(
+        data, distMat, (*target)(i), referencePoints(j), AlgorithmStep::BUILD);
       if (useAbsolute) {
         total += cost;
       } else {
@@ -288,8 +287,8 @@ arma::fmat BanditPAM_orig::swapSigma(
 
     // calculate change in loss for some subset of the data
     for (size_t j = 0; j < batchSize; j++) {
-      float cost =
-        KMedoids::cachedLoss(data, distMat, n, referencePoints(j), AlgorithmStep::MISC);
+      float cost = KMedoids::cachedLoss(data, distMat, n, referencePoints(j),
+                                        AlgorithmStep::MISC);
 
       if (k == (*assignments)(referencePoints(j))) {
         if (cost < (*secondBestDistances)(referencePoints(j))) {
@@ -353,8 +352,8 @@ arma::fvec BanditPAM_orig::swapTarget(
     size_t k = (*targets)(i) % medoidIndices->n_cols;
     // calculate total loss for some subset of the data
     for (size_t j = 0; j < tmpBatchSize; j++) {
-      float cost =
-        KMedoids::cachedLoss(data, distMat, n, referencePoints(j), AlgorithmStep::SWAP);
+      float cost = KMedoids::cachedLoss(data, distMat, n, referencePoints(j),
+                                        AlgorithmStep::SWAP);
       if (k == (*assignments)(referencePoints(j))) {
         if (cost < (*secondBestDistances)(referencePoints(j))) {
           total += cost;