Merge branch 'dev'

cityseer/algos/centrality.py

@@ -523,13 +523,8 @@ def local_centrality(node_data: np.ndarray,
             # only process in one direction
             if to_idx < src_idx:
                 continue
-            # identify true routes from ghosted routes
-            src_ghosted = nodes_ghosted[src_idx]
-            to_ghosted = nodes_ghosted[to_idx]
-            true_route = not src_ghosted and not to_ghosted
             # NODE WORKFLOW
-            # don't count routes originating or ending at ghosted nodes
-            if true_route and betw_nodes:
+            if betw_nodes:
                 # betweenness - only counting truly between vertices, not starting and ending verts
                 inter_idx = int(tree_preds[to_idx])
                 while True:

cityseer/util/mock.py

@@ -215,8 +215,7 @@ def mock_data_dict(G: nx.Graph, length: int = 50, random_seed: int = None) -> di
     for i in range(length):
         data_dict[i] = {
             'x': np.random.uniform(min_x, max_x),
-            'y': np.random.uniform(min_y, max_y),
-            'live': bool(np.random.randint(0, 1))
+            'y': np.random.uniform(min_y, max_y)
         }
 
     return data_dict

docs/intro.md

@@ -83,7 +83,7 @@ The [`Network_Layer.compute_centrality`](/metrics/networks.html#compute-centrali
 from cityseer.metrics import networks
 # create a Network layer from the networkX graph
 N = networks.Network_Layer_From_nX(G, distances=[200, 400, 800, 1600])
-# the full underlying method allows the computation of various centralities simultaneously, e.g.
+# the underlying method allows the computation of various centralities simultaneously, e.g.
 N.compute_centrality(measures=['segment_harmonic', 'segment_betweenness'])
 ```
 

docs/metrics/layers.md

@@ -53,8 +53,8 @@ data_dict_WGS = mock.mock_data_dict(G_wgs, random_seed=25)
 for i, (key, value) in enumerate(data_dict_WGS.items()):
     print(key, value)
     # prints:
-    # 0 {'x': -0.09600470559254023, 'y': 51.592916036617794, 'live': False}
-    # 1 {'x': -0.10621770551738155, 'y': 51.58888719412964, 'live': False}
+    # 0 {'x': -0.09600470559254023, 'y': 51.592916036617794}
+    # 1 {'x': -0.10621770551738155, 'y': 51.58888719412964}
     if i == 1:
         break
 
@@ -64,8 +64,8 @@ data_dict_UTM = layers.dict_wgs_to_utm(data_dict_WGS)
 for i, (key, value) in enumerate(data_dict_UTM.items()):
     print(key, value)
     # prints:
-    # 0 {'x': 701144.5207785056, 'y': 5719758.706109629, 'live': False}
-    # 1 {'x': 700455.0000341447, 'y': 5719282.703221394, 'live': False}
+    # 0 {'x': 701144.5207785056, 'y': 5719758.706109629}
+    # 1 {'x': 700455.0000341447, 'y': 5719282.703221394}
     if i == 1:
         break
 ```

requirements.txt

@@ -1,7 +1,7 @@
 numpy
 numba
 utm
-shapely>=1.7a2
+shapely>=1.7.0
 networkx>=2.4
 tqdm
 matplotlib

setup.py

@@ -8,15 +8,15 @@
 python setup.py sdist bdist_wheel
 pip install --upgrade twine
 TESTING REPO: twine upload --repository-url https://test.pypi.org/legacy/ dist/*
-DOWNLOADING FROM TEST REPO: pip install --extra-index-url https://test.pypi.org/simple/ cityseer==0.10.2.dev0
+DOWNLOADING FROM TEST REPO: pip install --extra-index-url https://test.pypi.org/simple/ cityseer==0.10.3.dev0
 OTHERWISE: twine upload dist/*
 '''
 
 from setuptools import setup
 
 setup(
     name='cityseer',
-    version='0.10.2',
+    version='0.10.3.dev0',
     packages=['cityseer', 'cityseer.algos', 'cityseer.metrics', 'cityseer.util'],
     description='Computational tools for urban analysis',
     url='https://github.com/cityseer/cityseer',

tests/algos/test_centrality.py

@@ -403,16 +403,18 @@ def test_decomposed_local_centrality():
     assert measures_data.shape == (m_range, d_range, len(G))
     assert measures_data_decomposed.shape == (m_range, d_range, len(G_decomposed))
     original_node_idx = np.where(node_data[:, 3] == 0)
-    # node based measures and betweenness segments will not match
-    # node measures can't filter by ghosted because then number of nodes is uneven
-    # segment betweenness become more finely spaced as the graph decomposes
+    # with increasing decomposition:
+    # - node based measures will not match
+    # - node based segment measures will match - these measure to the cut endpoints per thresholds
+    # - betweenness based segment won't match - doesn't measure to cut endpoints
     for m_idx in range(m_range):
+        print(m_idx)
         for d_idx in range(d_range):
             match = np.allclose(measures_data[m_idx][d_idx], measures_data_decomposed[m_idx][d_idx][original_node_idx],
                                 atol=0.1, rtol=0)  # relax precision
             if not match:
                 print('key', measure_keys[m_idx], 'dist:', distances[d_idx], 'match:', match)
-            if m_idx > 4 and m_idx != 10:
+            if m_idx in [5, 6, 7]:
                 assert match
 
 

tests/util/test_mock.py

@@ -51,7 +51,6 @@ def test_mock_data_dict():
         # check that attributes are present
         assert 'x' in v and isinstance(v['y'], (int, float))
         assert 'y' in v and isinstance(v['y'], (int, float))
-        assert 'live' in v and isinstance(v['live'], bool)
         assert v['x'] >= min_x and v['x'] <= max_x
         assert v['y'] >= min_y and v['y'] <= max_y