fix for #132

pgamit/__init__.py

@@ -2,6 +2,7 @@
 __all__ = [
     'cluster',
     'network',
+    'plots',
     'pyRinexName',
     'Utils',
     'pyJobServer',

pgamit/network.py

@@ -38,7 +38,7 @@
 from pgamit.pyGamitSession import GamitSession
 from pgamit.pyStation import StationCollection
 from pgamit.cluster import over_cluster, select_central_point, BisectingQMeans
-from pgamit.NetPlots import plot_global_network
+from pgamit.plots import plot_global_network
 
 BACKBONE_NET = 45
 NET_LIMIT = 40
@@ -214,19 +214,19 @@ def make_clusters(self, points, stations, net_limit=NET_LIMIT):
             # append to a regular list for integer indexing at line ~400
             cluster_ties.append(my_cluster_ties)
 
-        # put everything in a dictionary
-        clusters = {'centroids': points[central_points],
-                    'labels': cluster_labels,
-                    'stations': station_labels}
-
         # define output path for plot
         solution_base = self.GamitConfig.gamitopt['solutions_dir'].rstrip('/')
         end_path = '/%s/%s/%s' % (self.date.yyyy(), self.date.ddd(), self.name)
         path = solution_base + end_path + '_cluster.png'
 
         # generate plot of the network segmentation
-        plot_global_network(central_points, OC, qmean.labels_, points,
-                            output_path=path, lat_lon=False)
+        central_points = plot_global_network(central_points, OC, qmean.labels_,
+                                             points, output_path=path)
+
+        # put everything in a dictionary
+        clusters = {'centroids': points[central_points],
+                    'labels': cluster_labels,
+                    'stations': station_labels}
 
         return clusters, cluster_ties
 

pgamit/NetPlots.py → pgamit/plots.py

@@ -8,6 +8,7 @@
 import numpy as np
 import matplotlib.pyplot as plt
 from mpl_toolkits.basemap import Basemap
+from sklearn.neighbors import NearestNeighbors
 
 from pgamit.Utils import ecef2lla
 
@@ -86,10 +87,19 @@ def plot_global_network(central_points, OC, labels, points,
         # Flag centroid point
         remove = np.where(points == central_points[label])[0]
         points = points.tolist()
-        # remove centroid point so it's not repeated
-        points.pop(remove[0])
-        # add central point to beginning so it's the central connection point
-        points.insert(0, central_points[label])
+        try:
+            # remove centroid point so it's not repeated
+            points.pop(remove[0])
+            # add same point to beginning of list
+            points.insert(0, central_points[label])
+        except IndexError:
+            nbrs = NearestNeighbors(n_neighbors=1, algorithm='ball_tree',
+                                    metric='haversine').fit(LL[points])
+            idx = nbrs.kneighbors(LL[central_points[label]].reshape(1, -1),
+                                  return_distance=False)
+            # add central point to beginning as the central connection point
+            points.insert(0, points.pop(idx.squeeze()))
+            central_points[label] = points[0]
         nx.add_star(nodes[label], points)
         for position, proj in zip(positions, projs):
             mxy = np.zeros_like(LL[points])
@@ -112,3 +122,5 @@ def plot_global_network(central_points, OC, labels, points,
     fig.supxlabel("Figure runtime:  " + ("%.2fs" % (t1 - t0)).lstrip("0"))
     plt.savefig(output_path)
     plt.close()
+
+    return central_points