def test_num_2(self):
""" Tests on a numeric column while keeping two values """
assert pandas_utils.keepTopN(example_dataframe['column_low'],
2).equals(
pandas.Series([
numpy.nan, 1, 1, 2, 2, numpy.nan,
numpy.nan, numpy.nan, numpy.nan
]))
def test_str_2_with_def_num(self):
""" Tests on a string column while keeping two values and using a numeric default """
assert pandas_utils.keepTopN(example_dataframe['column_catstr'],
2,
default=1).equals(
pandas.Series([
1, 'alpha', 'a', 'a', 'a', 1,
'alpha', 1, 'alpha'
]))
def test_str_2(self):
""" Tests on a string column while keeping two values """
assert pandas_utils.keepTopN(example_dataframe['column_catstr'],
2).equals(
pandas.Series([
numpy.nan, 'alpha', 'a', 'a', 'a',
numpy.nan, 'alpha', numpy.nan,
'alpha'
]))
def test_str_all(self):
""" Tests on a string column while keeping all values """
assert pandas_utils.keepTopN(example_dataframe['column_catstr'],
99).equals(
pandas.Series([
'one', 'alpha', 'a', 'a', 'a',
'george', 'alpha', 'seventeen',
'alpha'
]))
def test_num_2_with_def_str(self):
""" Tests on a numeric column while keeping two values and using a string default """
assert pandas_utils.keepTopN(example_dataframe['column_low'],
2,
default='string').equals(
pandas.Series([
'string', 1, 1, 2, 2, 'string',
'string', 'string', 'string'
]))
def test_str_none(self):
""" Tests on a string column while keeping no values """
assert pandas_utils.keepTopN(example_dataframe['column_catstr'],
0).equals(
pandas.Series([numpy.nan] * 9,
dtype='object'))
def test_num_none(self):
""" Tests on a numeric column while keeping no values """
assert pandas_utils.keepTopN(example_dataframe['column_low'],
0).equals(pandas.Series([numpy.nan] * 9))
def test_num_2_with_def_num(self):
""" Tests on a numeric column while keeping two values and using a numeric default """
assert pandas_utils.keepTopN(
example_dataframe['column_low'], 2,
default=-1).equals(pandas.Series([-1, 1, 1, 2, 2, -1, -1, -1, -1]))
def test_num_all(self):
""" Tests on a numeric column while keeping all values """
assert pandas_utils.keepTopN(
example_dataframe['column_low'],
99).equals(pandas.Series([0, 1, 1, 2, 2, 3, 4, 5, 6]))
