def test_value_counts(self):
vals = [1, 2, nan, 0, nan, 1, 2, nan, nan, 1, 2, 0, 1, 1]
dense = pd.Series(vals, name='xx')
sparse = pd.SparseSeries(vals, name='xx')
tm.assert_series_equal(sparse.value_counts(), dense.value_counts())
tm.assert_series_equal(sparse.value_counts(dropna=False),
dense.value_counts(dropna=False))
sparse = pd.SparseSeries(vals, name='xx', fill_value=0)
tm.assert_series_equal(sparse.value_counts(), dense.value_counts())
tm.assert_series_equal(sparse.value_counts(dropna=False),
dense.value_counts(dropna=False))
def test_value_counts_int(self):
vals = [1, 2, 0, 1, 2, 1, 2, 0, 1, 1]
dense = pd.Series(vals, name='xx')
# fill_value is np.nan, but should not be included in the result
sparse = pd.SparseSeries(vals, name='xx')
tm.assert_series_equal(sparse.value_counts(), dense.value_counts())
tm.assert_series_equal(sparse.value_counts(dropna=False),
dense.value_counts(dropna=False))
sparse = pd.SparseSeries(vals, name='xx', fill_value=0)
tm.assert_series_equal(sparse.value_counts(), dense.value_counts())
tm.assert_series_equal(sparse.value_counts(dropna=False),
dense.value_counts(dropna=False))
def test_value_counts(self):
vals = [1, 2, nan, 0, nan, 1, 2, nan, nan, 1, 2, 0, 1, 1]
dense = pd.Series(vals, name='xx')
sparse = pd.SparseSeries(vals, name='xx')
tm.assert_series_equal(sparse.value_counts(),
dense.value_counts())
tm.assert_series_equal(sparse.value_counts(dropna=False),
dense.value_counts(dropna=False))
sparse = pd.SparseSeries(vals, name='xx', fill_value=0)
tm.assert_series_equal(sparse.value_counts(),
dense.value_counts())
tm.assert_series_equal(sparse.value_counts(dropna=False),
dense.value_counts(dropna=False))
def test_value_counts_int(self):
vals = [1, 2, 0, 1, 2, 1, 2, 0, 1, 1]
dense = pd.Series(vals, name='xx')
# fill_value is np.nan, but should not be included in the result
sparse = pd.SparseSeries(vals, name='xx')
tm.assert_series_equal(sparse.value_counts(),
dense.value_counts())
tm.assert_series_equal(sparse.value_counts(dropna=False),
dense.value_counts(dropna=False))
sparse = pd.SparseSeries(vals, name='xx', fill_value=0)
tm.assert_series_equal(sparse.value_counts(),
dense.value_counts())
tm.assert_series_equal(sparse.value_counts(dropna=False),
dense.value_counts(dropna=False))
def test_value_counts_dup(self):
vals = [1, 2, nan, 0, nan, 1, 2, nan, nan, 1, 2, 0, 1, 1]
# numeric op may cause sp_values to include the same value as
# fill_value
dense = pd.Series(vals, name='xx') / 0.
sparse = pd.SparseSeries(vals, name='xx') / 0.
tm.assert_series_equal(sparse.value_counts(), dense.value_counts())
tm.assert_series_equal(sparse.value_counts(dropna=False),
dense.value_counts(dropna=False))
vals = [1, 2, 0, 0, 0, 1, 2, 0, 0, 1, 2, 0, 1, 1]
dense = pd.Series(vals, name='xx') * 0.
sparse = pd.SparseSeries(vals, name='xx') * 0.
tm.assert_series_equal(sparse.value_counts(), dense.value_counts())
tm.assert_series_equal(sparse.value_counts(dropna=False),
dense.value_counts(dropna=False))
def test_value_counts_dup(self):
vals = [1, 2, nan, 0, nan, 1, 2, nan, nan, 1, 2, 0, 1, 1]
# numeric op may cause sp_values to include the same value as
# fill_value
dense = pd.Series(vals, name='xx') / 0.
sparse = pd.SparseSeries(vals, name='xx') / 0.
tm.assert_series_equal(sparse.value_counts(),
dense.value_counts())
tm.assert_series_equal(sparse.value_counts(dropna=False),
dense.value_counts(dropna=False))
vals = [1, 2, 0, 0, 0, 1, 2, 0, 0, 1, 2, 0, 1, 1]
dense = pd.Series(vals, name='xx') * 0.
sparse = pd.SparseSeries(vals, name='xx') * 0.
tm.assert_series_equal(sparse.value_counts(),
dense.value_counts())
tm.assert_series_equal(sparse.value_counts(dropna=False),
dense.value_counts(dropna=False))
