Merge pull request #6231 from jreback/test_factorize

TST: add tests for algos.factorize (GH6212)

Author: jreback

pandas/core/algorithms.py

@@ -129,6 +129,10 @@ def factorize(values, sort=False, order=None, na_sentinel=-1):
 
     Returns
     -------
+    labels : the indexer to the original array
+    uniques : the unique values
+
+    note: an array of Periods will ignore sort as it returns an always sorted PeriodIndex
     """
     from pandas.tseries.period import PeriodIndex
     vals = np.asarray(values)

pandas/tests/test_algos.py

@@ -44,6 +44,78 @@ def test_strings(self):
         expected = Series(np.array([1, 0, np.nan, 0, 1, 2, np.nan]))
         tm.assert_series_equal(result,expected)
 
+class TestFactorize(tm.TestCase):
+    _multiprocess_can_split_ = True
+
+    def test_basic(self):
+
+        labels, uniques = algos.factorize(['a', 'b', 'b', 'a',
+                                           'a', 'c', 'c', 'c'])
+        self.assert_(np.array_equal(labels, np.array([ 0, 1, 1, 0, 0, 2, 2, 2],dtype=np.int64)))
+        self.assert_(np.array_equal(uniques, np.array(['a','b','c'], dtype=object)))
+
+        labels, uniques = algos.factorize(['a', 'b', 'b', 'a',
+                                           'a', 'c', 'c', 'c'], sort=True)
+        self.assert_(np.array_equal(labels, np.array([ 0, 1, 1, 0, 0, 2, 2, 2],dtype=np.int64)))
+        self.assert_(np.array_equal(uniques, np.array(['a','b','c'], dtype=object)))
+
+        labels, uniques = algos.factorize(list(reversed(range(5))))
+        self.assert_(np.array_equal(labels, np.array([0, 1, 2, 3, 4], dtype=np.int64)))
+        self.assert_(np.array_equal(uniques, np.array([ 4, 3, 2, 1, 0],dtype=np.int64)))
+
+        labels, uniques = algos.factorize(list(reversed(range(5))), sort=True)
+        self.assert_(np.array_equal(labels, np.array([ 4, 3, 2, 1, 0],dtype=np.int64)))
+        self.assert_(np.array_equal(uniques, np.array([0, 1, 2, 3, 4], dtype=np.int64)))
+
+        labels, uniques = algos.factorize(list(reversed(np.arange(5.))))
+        self.assert_(np.array_equal(labels, np.array([0., 1., 2., 3., 4.], dtype=np.float64)))
+        self.assert_(np.array_equal(uniques, np.array([ 4, 3, 2, 1, 0],dtype=np.int64)))
+
+        labels, uniques = algos.factorize(list(reversed(np.arange(5.))), sort=True)
+        self.assert_(np.array_equal(labels, np.array([ 4, 3, 2, 1, 0],dtype=np.int64)))
+        self.assert_(np.array_equal(uniques, np.array([0., 1., 2., 3., 4.], dtype=np.float64)))
+
+    def test_mixed(self):
+
+        # doc example reshaping.rst
+        x = Series(['A', 'A', np.nan, 'B', 3.14, np.inf])
+        labels, uniques = algos.factorize(x)
+
+        self.assert_(np.array_equal(labels, np.array([ 0,  0, -1,  1,  2,  3],dtype=np.int64)))
+        self.assert_(np.array_equal(uniques, np.array(['A', 'B', 3.14, np.inf], dtype=object)))
+
+        labels, uniques = algos.factorize(x, sort=True)
+        self.assert_(np.array_equal(labels, np.array([ 2,  2, -1,  3,  0,  1],dtype=np.int64)))
+        self.assert_(np.array_equal(uniques, np.array([3.14, np.inf, 'A', 'B'], dtype=object)))
+
+    def test_datelike(self):
+
+        # M8
+        v1 = pd.Timestamp('20130101 09:00:00.00004')
+        v2 = pd.Timestamp('20130101')
+        x = Series([v1,v1,v1,v2,v2,v1])
+        labels, uniques = algos.factorize(x)
+        self.assert_(np.array_equal(labels, np.array([ 0,0,0,1,1,0],dtype=np.int64)))
+        self.assert_(np.array_equal(uniques, np.array([v1.value,v2.value],dtype='M8[ns]')))
+
+        labels, uniques = algos.factorize(x, sort=True)
+        self.assert_(np.array_equal(labels, np.array([ 1,1,1,0,0,1],dtype=np.int64)))
+        self.assert_(np.array_equal(uniques, np.array([v2.value,v1.value],dtype='M8[ns]')))
+
+        # period
+        v1 = pd.Period('201302',freq='M')
+        v2 = pd.Period('201303',freq='M')
+        x = Series([v1,v1,v1,v2,v2,v1])
+
+        # periods are not 'sorted' as they are converted back into an index
+        labels, uniques = algos.factorize(x)
+        self.assert_(np.array_equal(labels, np.array([ 0,0,0,1,1,0],dtype=np.int64)))
+        self.assert_(np.array_equal(uniques, np.array([v1, v2],dtype=object)))
+
+        labels, uniques = algos.factorize(x,sort=True)
+        self.assert_(np.array_equal(labels, np.array([ 0,0,0,1,1,0],dtype=np.int64)))
+        self.assert_(np.array_equal(uniques, np.array([v1, v2],dtype=object)))
+
 class TestUnique(tm.TestCase):
     _multiprocess_can_split_ = True