list sampler

dedupeio · Jul 6, 2017 · ffac79d · ffac79d
1 parent ccfe451
commit ffac79d
Show file tree

Hide file tree

Showing 2 changed files with 5 additions and 6 deletions.
diff --git a/dedupe/sampling.py b/dedupe/sampling.py
@@ -18,9 +18,9 @@ def blockedSample(sampler, sample_size, predicates, *args) :
     while remaining_sample and predicates :
         random.shuffle(predicates)
 
-        new_sample = sampler(remaining_sample, 
-                             predicates,
-                             *args)
+        new_sample = list(sampler(remaining_sample,
+                                  predicates,
+                                  *args))
 
         filtered_sample = (subsample for subsample 
                            in new_sample if subsample)
@@ -39,7 +39,6 @@ def blockedSample(sampler, sample_size, predicates, *args) :
                           % (sample_size, len(blocked_sample)))
             break
 
-
         predicates = [pred for pred, pred_sample 
                       in zip(predicates, new_sample)
                       if pred_sample or pred_sample is None]

diff --git a/setup.py b/setup.py
@@ -17,7 +17,7 @@
                   'doublemetaphone',
                   'highered>=0.2.0',
                   'canonicalize',
-                  'simplecosine>=1.1',
+                  'simplecosine>=1.2',
                   'haversine>=0.4.1',
                   'BTrees>=4.1.4',
                   'simplejson',
@@ -27,7 +27,7 @@
 setup(
     name='dedupe',
     url='https://github.com/dedupeio/dedupe',
-    version='1.6.15',
+    version='1.6.16',
     author='Forest Gregg',
     author_email='fgregg@datamade.us',
     description='A python library for accurate and scaleable data deduplication and entity-resolution',