Add code to skip random neighborhood in NCP

kfoynt · Nov 6, 2018 · 865c97e · 865c97e
1 parent d05e389
commit 865c97e
Show file tree

Hide file tree

Showing 2 changed files with 17 additions and 5 deletions.
diff --git a/localgraphclustering/ncp.py b/localgraphclustering/ncp.py
@@ -431,6 +431,7 @@ def approxPageRank(self,
                        localmin_ratio: float = 0.5,
                        neighborhood_ratio: float = 0.1,
                        neighborhoods: bool = True,
+                       random_neighborhoods: bool = True,
                        timeout: float = 1000,
                        spectral_args: Dict = {},
                        deep: bool = False,
@@ -481,10 +482,14 @@ def approxPageRank(self,
             del kwargs['ratio']
 
         deeptimeout = timeout # save the timeout in case we also run the deep params
+        nruns = 1
+        if random_neighborhoods:
+            nruns += 1
+        if localmins:
+            nruns += 1
 
         log = SimpleLogForLongComputations(120, "approxPageRank:%s"%(methodname))
 
-
         if neighborhoods:
             self.add_random_neighborhood_samples(
                 method=_evaluate_set,
@@ -501,9 +506,11 @@ def approxPageRank(self,
                     methodname="%s_localmin:rho=%.0e"%(methodname, rho*10),
                     neighborhoods=True,
                     ratio=localmin_ratio,
-                    timeout=timeout/(3*len(rholist)),**kwargs)
+                    timeout=timeout/(nruns*len(rholist)),**kwargs)
                 log.log("localmin rho=%.1e"%(rho))
-            timeout -= timeout/3 # reduce the time left...
+            timeout -= timeout/nruns # reduce the time left...
+            nruns -= 1
+
 
         for rho in rholist:
             if myratio is not None:
@@ -512,10 +519,10 @@ def approxPageRank(self,
                 method=functools.partial(
                     spectral_clustering,**spectral_args,alpha=alpha,rho=rho,method=method),
                 methodname="%s:rho=%.0e"%(methodname, rho),
-                timeout=timeout/(2*len(rholist)), **kwargs)
+                timeout=timeout/(nruns*len(rholist)), **kwargs)
             log.log("random_node rho=%.1e"%(rho))
 
-        timeout -= timeout/2 # reduce the time left...
+        timeout -= timeout/nruns # reduce the time left...
 
         for rho in rholist:
             if myratio is not None:
@@ -549,6 +556,7 @@ def approxPageRank(self,
             self.approxPageRank(gamma=deepgamma,rholist=deeprhos,
                 localmins=localmins, localmin_ratio=localmin_ratio,
                 neighborhoods=False, neighborhood_ratio=neighborhood_ratio,
+                random_neighborhoods=random_neighborhoods,
                 timeout = deeptimeout, spectral_args=spectral_args, deep=False,
                 methodname_prefix=deepmethodname_prefix,
                 **kwargs)

diff --git a/localgraphclustering/tests/test_ncp.py b/localgraphclustering/tests/test_ncp.py
@@ -109,6 +109,10 @@ def test_ncp_sets():
 def test_apr_deep():
     G = load_example_graph()
     df = lgc.NCPData(G).approxPageRank(ratio=1, gamma=0.1, rholist=[1e-2, 1e-3], deep=True)
+
+def test_apr_only_node_samples():
+    G = load_example_graph()
+    df = lgc.NCPData(G).approxPageRank(ratio=1, gamma=0.1, rholist=[1e-2, 1e-3], random_neighborhoods=False, localmins=False)    
 
 @pytest.mark.long_tests
 def test_ncp_crd_big():