Using OPTICS clustering instead of DBSCAN

src/data-hub/community-detection-microservice/app/processing/clustering/clusterer.py

 import json
 import numpy as np
 import matplotlib.pyplot as plt
-from sklearn.cluster import DBSCAN
+from sklearn.cluster import DBSCAN, OPTICS
 from typing import List, Dict, Any, TypeVar
 from deprecated import deprecated
 
@@ -59,7 +59,7 @@ class Clusterer:
         if features is None or len(features) == 0:
             return features # trash in trash out
     
-        dbsc = DBSCAN(eps = self.epsilon, min_samples = self.min_points)
+        dbsc = OPTICS() # DBSCAN(eps = self.epsilon, min_samples = self.min_points)
         dbsc = dbsc.fit(features)
         labels = dbsc.labels_
 

src/data-hub/community-detection-microservice/app/run_clustering.py

@@ -4,6 +4,7 @@ modules_path = '../../../modules/'
 if os.path.exists(modules_path):
     sys.path.insert(1, modules_path)
 
+import json
 from db.entities import *
 from typing import List, Dict, Tuple
 from db.repository import Repository, AgiRepository
@@ -21,11 +22,13 @@ def run_generic_clustering():
     all_layers:List[Layer] = repo.get_layers()
 
     for layer in all_layers:
+        if layer.properties is None or len(layer.properties) == 0:
+            continue
+
         print(f"Clustering {layer.layer_name}")
         clusters = run_clustering_for_layer(layer)
-        
         cluster_set = ClusterSet(layer.layer_name, clusters)
-        repo.add_clusterset(cluster_set)
+        store_clusterset(cluster_set)
 
 
 def run_clustering_for_layer(layer: Layer) -> List[Cluster]:
@@ -38,6 +41,11 @@ def run_clustering_for_layer(layer: Layer) -> List[Cluster]:
 
     return [Cluster(key, value) for key, value in res.items()]
 
+def store_clusterset(cluster_set: ClusterSet):
+    repo.add_clusterset(cluster_set)
+    # with open(f'clusterset_{cluster_set.layer_name}.txt', 'w') as file:
+    #     file.write(json.dumps(cluster_set.to_serializable_dict()))
+
 
 def run_location_clustering():
     user_clusterer = Clusterer()