2.1.0 (#24)

* Improved matching (#20)

* Added new atom matching function.

* Added performance tests.

* Update.

* Fixed formatting.

* Updated tests.

* Added performance scripts.

* Updated performance script.

* Updated performance test.

* Replaced test.

* Updated graph search.

* Fixed linting.

* Fixed linting.

* Fixed gitignore.

* Cleanup.

* Updated numpy dependency restriction.

* Started simplifying API.

* Distance calc optimization (#21)

* Trying to disable chemical similarity check.

* updated workflow.

* Fixed formatting.

* Fixed class instantiation.

* Simplified LinkedUnitCollection.

* Fixed formatting.

* Disabled test.

* Simplified LinkedUnitCollection.

* Trying to undo.

* Update.

* Fixed missing default.

* Updated actions.

* Cleaned up old implementations.

* Updated benchmarks.

* Fixed formatting.

* Added branch to tests.

* Optimization (#23)

* Removed unnecesary calculation of finite displacements.

* Removed storing of displacements in matching.

* Removed unnecessary distance calculation with numpy.

* Added guard against invalid cell shape.

* Added guard agains invalid cell

* Fixed formatting.

* Added exception check on cluster cleanup.

* Updated docstring.

* Improved tests.

* Enabled other tests.

* Cleaned up tests.

* Completely removed chemical similarity check.

.github/workflows/test.yml

@@ -6,7 +6,7 @@ name: test
 
 on:
   push:
-    branches: [ "main", "v2.0.0"]
+    branches: [ "main", "2.1.0", "optimization"]
   pull_request:
     branches: [ "main"]
 
@@ -21,7 +21,7 @@ jobs:
     steps:
     - uses: actions/checkout@v4
     - name: Set up Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v4
+      uses: actions/setup-python@v5
       with:
         python-version: ${{ matrix.python-version }}
     - name: Install dependencies

.gitignore

@@ -191,4 +191,11 @@ cython_debug/
 #.idea/
 
 # VSCode
-.vscode/*
+.vscode/*
+
+# Coverage report
+coverage
+tests/.coverage
+
+# performance results
+tests/performance/results

matid/classification/classifications.py

@@ -50,27 +50,9 @@ def basis_indices(self):
 
     @property
     def outliers(self):
-        return self.region.get_outliers()
-
-    @property
-    def interstitials(self):
-        return self.region.get_interstitials()
-
-    @property
-    def adsorbates(self):
-        return self.region.get_adsorbates()
-
-    @property
-    def substitutions(self):
-        return self.region.get_substitutions()
-
-    @property
-    def vacancies(self):
-        return self.region.get_vacancies()
-
-    @property
-    def unknowns(self):
-        return self.region.get_unknowns()
+        all = set(list(range(len(self.atoms))))
+        region = self.region.get_basis_indices()
+        return list(all - region)
 
     @property
     def prototype_cell(self):

matid/classification/classifier.py

@@ -267,26 +267,18 @@ class that represents a classification.
                 n_region_conn = np.sum(region_conn)
                 region_is_periodic = n_region_conn == 2
 
-                # This might be unnecessary because the connectivity of the
-                # unit cell is already checked.
-                clusters = best_region.get_clusters()
-                basis_indices = set(list(best_region.get_basis_indices()))
-                split = True
-                for cluster in clusters:
-                    if basis_indices.issubset(cluster):
-                        split = False
-
                 # Check that the found region covers enough of the entire
                 # system. If not, then the region alone cannot be used to
                 # classify the entire structure. This happens e.g. when one
                 # 2D sheet is found from a 2D heterostructure, or a local
                 # pattern is found inside a structure.
+                basis_indices = set(list(best_region.get_basis_indices()))
                 n_atoms = len(system)
                 n_basis_atoms = len(basis_indices)
                 coverage = n_basis_atoms / n_atoms
                 covered = coverage >= self.min_coverage
 
-                if not split and covered and region_is_periodic:
+                if covered and region_is_periodic:
                     if best_region.is_2d:
                         classification = Material2D(input_system, best_region)
                     else:
@@ -329,14 +321,12 @@ def cross_validate_region(
                         cell_size_tol=self.cell_size_tol,
                         max_2d_cell_height=self.max_2d_cell_height,
                         max_2d_single_cell_size=self.max_2d_single_cell_size,
-                        chem_similarity_threshold=self.chem_similarity_threshold,
                     )
                     region = periodicfinder.get_region(
                         system,
                         index,
                         size,
                         tol,
-                        self.abs_delaunay_threshold,
                         self.bond_threshold,
                         distances=distances,
                     )

matid/clustering/sbc.py

@@ -117,9 +117,7 @@ def get_clusters(
         distances = matid.geometry.get_distances(system_copy, radii)
 
         # Iteratively search for new clusters until whole system is covered
-        periodic_finder = PeriodicFinder(
-            angle_tol=angle_tol, chem_similarity_threshold=0
-        )
+        periodic_finder = PeriodicFinder(angle_tol=angle_tol)
         indices = set(list(range(len(system_copy))))
         clusters = []
         while len(indices) != 0:
@@ -325,12 +323,18 @@ def _clean_clusters(self, clusters, bond_threshold):
         Returns:
             List of Clusters where any outlier atoms have been removed.
         """
+        clusters_cleaned = []
         for cluster in clusters:
-            dbscan_clusters = matid.geometry.get_clusters(
-                cluster._get_distance_matrix_radii_mic(),
-                bond_threshold,
-                min_samples=1,
-            )
+            # If the cluster cleaning fails, the cluster is not reported
+            try:
+                dbscan_clusters = matid.geometry.get_clusters(
+                    cluster._get_distance_matrix_radii_mic(),
+                    bond_threshold,
+                    min_samples=1,
+                )
+            except Exception:
+                continue
             largest_indices = max(dbscan_clusters, key=lambda x: len(x))
             cluster.indices = np.array(cluster.indices)[largest_indices].tolist()
-        return clusters
+            clusters_cleaned.append(cluster)
+        return clusters_cleaned

matid/core/distances.py

@@ -7,14 +7,12 @@ def __init__(
         self,
         disp_tensor_mic,
         disp_factors,
-        disp_tensor_finite,
         dist_matrix_mic,
         dist_matrix_radii_mic,
         cell_list=None,
     ):
         self.disp_tensor_mic = disp_tensor_mic
         self.disp_factors = disp_factors
-        self.disp_tensor_finite = disp_tensor_finite
         self.dist_matrix_mic = dist_matrix_mic
         self.dist_matrix_radii_mic = dist_matrix_radii_mic
         self.cell_list = cell_list