def test_dataframe_corr_with(setup):
rs = np.random.RandomState(0)
raw_df = rs.rand(20, 10)
raw_df = pd.DataFrame(np.where(raw_df > 0.4, raw_df, np.nan),
columns=list('ABCDEFGHIJ'))
raw_df2 = rs.rand(20, 10)
raw_df2 = pd.DataFrame(np.where(raw_df2 > 0.4, raw_df2, np.nan),
columns=list('ACDEGHIJKL'))
raw_s = rs.rand(20)
raw_s = pd.Series(np.where(raw_s > 0.4, raw_s, np.nan))
raw_s2 = rs.rand(10)
raw_s2 = pd.Series(np.where(raw_s2 > 0.4, raw_s2, np.nan),
index=raw_df2.columns)
df = DataFrame(raw_df)
df2 = DataFrame(raw_df2)
result = df.corrwith(df2)
pd.testing.assert_series_equal(result.execute().fetch(),
raw_df.corrwith(raw_df2))
result = df.corrwith(df2, axis=1)
pd.testing.assert_series_equal(result.execute().fetch(),
raw_df.corrwith(raw_df2, axis=1))
result = df.corrwith(df2, method='kendall')
pd.testing.assert_series_equal(result.execute().fetch(),
raw_df.corrwith(raw_df2, method='kendall'))
df = DataFrame(raw_df, chunk_size=4)
df2 = DataFrame(raw_df2, chunk_size=6)
s = Series(raw_s, chunk_size=5)
s2 = Series(raw_s2, chunk_size=5)
with pytest.raises(Exception):
df.corrwith(df2, method='kendall').execute()
result = df.corrwith(df2)
pd.testing.assert_series_equal(result.execute().fetch().sort_index(),
raw_df.corrwith(raw_df2).sort_index())
result = df.corrwith(df2, axis=1)
pd.testing.assert_series_equal(
result.execute().fetch().sort_index(),
raw_df.corrwith(raw_df2, axis=1).sort_index())
result = df.corrwith(s)
pd.testing.assert_series_equal(result.execute().fetch().sort_index(),
raw_df.corrwith(raw_s).sort_index())
result = df.corrwith(s2, axis=1)
pd.testing.assert_series_equal(
result.execute().fetch().sort_index(),
raw_df.corrwith(raw_s2, axis=1).sort_index())
def testDataFrameCorrWith(self):
rs = np.random.RandomState(0)
raw_df = rs.rand(20, 10)
raw_df = pd.DataFrame(np.where(raw_df > 0.4, raw_df, np.nan),
columns=list('ABCDEFGHIJ'))
raw_df2 = rs.rand(20, 10)
raw_df2 = pd.DataFrame(np.where(raw_df2 > 0.4, raw_df2, np.nan),
columns=list('ACDEGHIJKL'))
raw_s = rs.rand(20)
raw_s = pd.Series(np.where(raw_s > 0.4, raw_s, np.nan))
raw_s2 = rs.rand(10)
raw_s2 = pd.Series(np.where(raw_s2 > 0.4, raw_s2, np.nan),
index=raw_df2.columns)
df = DataFrame(raw_df)
df2 = DataFrame(raw_df2)
result = df.corrwith(df2)
pd.testing.assert_series_equal(
self.executor.execute_dataframe(result, concat=True)[0],
raw_df.corrwith(raw_df2))
result = df.corrwith(df2, axis=1)
pd.testing.assert_series_equal(
self.executor.execute_dataframe(result, concat=True)[0],
raw_df.corrwith(raw_df2, axis=1))
result = df.corrwith(df2, method='kendall')
pd.testing.assert_series_equal(
self.executor.execute_dataframe(result, concat=True)[0],
raw_df.corrwith(raw_df2, method='kendall'))
df = DataFrame(raw_df, chunk_size=4)
df2 = DataFrame(raw_df2, chunk_size=6)
s = Series(raw_s, chunk_size=5)
s2 = Series(raw_s2, chunk_size=5)
with self.assertRaises(Exception):
self.executor.execute_dataframe(df.corrwith(df2, method='kendall'),
concat=True)
result = df.corrwith(df2)
pd.testing.assert_series_equal(
self.executor.execute_dataframe(result,
concat=True)[0].sort_index(),
raw_df.corrwith(raw_df2).sort_index())
result = df.corrwith(df2, axis=1)
pd.testing.assert_series_equal(
self.executor.execute_dataframe(result,
concat=True)[0].sort_index(),
raw_df.corrwith(raw_df2, axis=1).sort_index())
result = df.corrwith(s)
pd.testing.assert_series_equal(
self.executor.execute_dataframe(result,
concat=True)[0].sort_index(),
raw_df.corrwith(raw_s).sort_index())
result = df.corrwith(s2, axis=1)
pd.testing.assert_series_equal(
self.executor.execute_dataframe(result,
concat=True)[0].sort_index(),
raw_df.corrwith(raw_s2, axis=1).sort_index())