Fix div by 0 issue.

src/python/deepgnn/graph_engine/snark/preprocess/sampler/metric.py

@@ -9,11 +9,19 @@
 
 def densification(g: nx.digraph) -> float:
     """Return densification constant in NetworkX directed graph."""
-    return math.log(nx.number_of_edges(g), nx.number_of_nodes(g))
+    n = nx.number_of_nodes(g)
+    if n <= 1:
+        return 0.0
+    e = nx.number_of_edges(g)
+    if e == 0:
+        return 0.0
+    return math.log(e, n)
 
 
 def diameter(g: nx.digraph) -> int:
     """Return effective 90% diameter in NetworkX directed graph."""
+    if nx.number_of_nodes(g) == 0:
+        return 0
     largest_strongly_connected_component = g.subgraph(
         max(nx.strongly_connected_components(g), key=len)
     )
@@ -22,16 +30,19 @@ def diameter(g: nx.digraph) -> int:
 
 def largest_connected_component(g: nx.digraph) -> float:
     """Return the scaled size of the largest strongly connected component in NetworkX directed graph."""
+    n = nx.number_of_nodes(g)
+    if n == 0:
+        return 0.0
     largest_strongly_connected_component = g.subgraph(
         max(nx.strongly_connected_components(g), key=len)
     )
-    return nx.number_of_nodes(
-        largest_strongly_connected_component
-    ) / nx.number_of_nodes(g)
+    return nx.number_of_nodes(largest_strongly_connected_component) / n
 
 
 def max_adjacency(g: nx.digraph) -> float:
     """Return the largest eigenvalue of the adjacency matrix in NetworkX directed graph."""
+    if nx.number_of_nodes(g) == 0:
+        return 0.0
     return max(nx.adjacency_spectrum(g))