Skip to content

Fixes nan values after pandas update, add documentation example to the unit tests #22

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Jul 11, 2021
Merged

Fixes nan values after pandas update, add documentation example to the unit tests #22

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
37 changes: 36 additions & 1 deletion _unittests/ut_df/test_pandas_groupbynan.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,11 +1,12 @@
# coding: utf-8
"""
@brief test log(time=1s)
"""
import unittest
import pandas
import numpy
from scipy.sparse.linalg import lsqr as sparse_lsqr
from pyquickhelper.pycode import ExtTestCase
from pyquickhelper.pycode import ExtTestCase, ignore_warnings
from pandas_streaming.df import pandas_groupby_nan, numpy_types


Expand Down Expand Up @@ -102,6 +103,40 @@ def test_pandas_groupbynan_regular_nanback(self):
lambda: pandas_groupby_nan(df, ["a", "cc"], nanback=True).sum(),
NotImplementedError)

def test_pandas_groupbynan_doc(self):
data = [dict(a=2, ind="a", n=1),
dict(a=2, ind="a"),
dict(a=3, ind="b"),
dict(a=30)]
df = pandas.DataFrame(data)
gr2 = pandas_groupby_nan(df, ["ind"]).sum()
ind = list(gr2['ind'])
self.assertTrue(numpy.isnan(ind[-1]))
val = list(gr2['a'])
self.assertEqual(val[-1], 30)

@ignore_warnings(UserWarning)
def test_pandas_groupbynan_doc2(self):
data = [dict(a=2, ind="a", n=1),
dict(a=2, ind="a"),
dict(a=3, ind="b"),
dict(a=30)]
df = pandas.DataFrame(data)
gr2 = pandas_groupby_nan(df, ["ind", "a"], nanback=False).sum()
ind = list(gr2['ind'])
self.assertEqual(ind[-1], "²nan")

def test_pandas_groupbynan_doc3(self):
data = [dict(a=2, ind="a", n=1),
dict(a=2, ind="a"),
dict(a=3, ind="b"),
dict(a=30)]
df = pandas.DataFrame(data)
self.assertRaise(lambda: pandas_groupby_nan(df, ["ind", "n"]).sum(),
NotImplementedError)
# ind = list(gr2['ind'])
# self.assertTrue(numpy.isnan(ind[-1]))


if __name__ == "__main__":
unittest.main()
30 changes: 24 additions & 6 deletions pandas_streaming/df/dataframe_helpers.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -289,7 +289,7 @@ def pandas_fillna(df, by, hasna=None, suffix=None):
:param suffix: use a prefix for the NaN value
:return: list of values chosen for each column, new dataframe (new copy)
"""
suffix = suffix if suffix else "²"
suffix = suffix if suffix else "²nan"
df = df.copy()
rep = {}
for c in by:
Expand Down Expand Up @@ -364,7 +364,10 @@ def pandas_groupby_nan(df, by, axis=0, as_index=False, suffix=None, nanback=True

from pandas import DataFrame

data = [dict(a=2, ind="a", n=1), dict(a=2, ind="a"), dict(a=3, ind="b"), dict(a=30)]
data = [dict(a=2, ind="a", n=1),
dict(a=2, ind="a"),
dict(a=3, ind="b"),
dict(a=30)]
df = DataFrame(data)
print(df)
gr = df.groupby(["ind"]).sum()
Expand All @@ -378,7 +381,10 @@ def pandas_groupby_nan(df, by, axis=0, as_index=False, suffix=None, nanback=True
from pandas import DataFrame
from pandas_streaming.df import pandas_groupby_nan

data = [dict(a=2, ind="a", n=1), dict(a=2, ind="a"), dict(a=3, ind="b"), dict(a=30)]
data = [dict(a=2, ind="a", n=1),
dict(a=2, ind="a"),
dict(a=3, ind="b"),
dict(a=30)]
df = DataFrame(data)
gr2 = pandas_groupby_nan(df, ["ind"]).sum()
print(gr2)
Expand Down Expand Up @@ -436,10 +442,22 @@ def pandas_groupby_nan(df, by, axis=0, as_index=False, suffix=None, nanback=True
res.grouper.groupings[0].grouping_vector = arr
if (hasattr(res.grouper.groupings[0], '_cache') and
'result_index' in res.grouper.groupings[0]._cache):
res.grouper.groupings[0]._cache = {}
index = res.grouper.groupings[0]._cache['result_index']
if len(rep) == 1:
key = list(rep.values())[0]
new_index = numpy.array(index)
for i in range(0, len(new_index)): # pylint: disable=C0200
if new_index[i] == key:
new_index[i] = numpy.nan
res.grouper.groupings[0]._cache['result_index'] = (
index.__class__(new_index))
else:
raise NotImplementedError(
"NaN values not implemented for multiindex.")
else:
raise NotImplementedError("Not implemented for type: {0}".format(
type(res.grouper.groupings[0].grouper)))
raise NotImplementedError(
"Not implemented for type: {0}".format(
type(res.grouper.groupings[0].grouper)))
res.grouper._cache['result_index'] = res.grouper.groupings[0]._group_index
else:
if not nanback:
Expand Down