NDFrame.interpolate(): variable 'ax' not assigned when axis='index'

[ellequelle]

The NDFrame.interpolate function fails when passing a string as axis. Example:

>>> import numpy as np
>>> import pandas as pd
>>> df = pd.DataFrame(np.zeros((3,2)), columns=['a','b'])
>>> df.iloc[1] = np.nan
>>> df
     a    b
0  0.0  0.0
1  NaN  NaN
2  0.0  0.0
>>> df.interpolate(axis=0)
     a    b
0  0.0  0.0
1  0.0  0.0
2  0.0  0.0
>>> df.interpolate(axis='index')
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/Users/lehanson/anaconda3/envs/plots/lib/python3.7/site-packages/pandas/core/generic.py", line 7006, in interpolate
    ax = _maybe_transposed_self._get_axis_number(ax)
UnboundLocalError: local variable 'ax' referenced before assignment

From the documentation and from the function itself, it looks like df.interpolate(axis='index') was intended to work, but that maybe someone accidentally deleted a line in generic.py? The function seems to work properly if I add ax = axis in the else block here:

pandas/pandas/core/generic.py

Lines 6998 to 7006 in 171c716

	if axis == 0:
	ax = self._info_axis_name
	_maybe_transposed_self = self
	elif axis == 1:
	_maybe_transposed_self = self.T
	ax = 1
	else:
	_maybe_transposed_self = self
	ax = _maybe_transposed_self._get_axis_number(ax)

I am using pandas version 0.25.1

[TomAugspurger]

Thanks for the report, confirmed on master.

We should have an axis = self._get_axis_number(axis) at the start of NDFrame.interpolate (and a whatsnew note & test to ensure we don't regress).

[ellequelle]

@TomAugspurger do you mean something like this? Also: there doesn't seem to be a whatsnew file since 0.25.2, does this go into v0.25.3.rst?

axis = self._get_axis_number(axis)
if axis == 0:
    ax = self._info_axis_name
    _maybe_transposed_self = self
elif axis == 1:
    _maybe_transposed_self = self.T
    ax = 1
else:
    ax = axis
    _maybe_transposed_self = self
ax = _maybe_transposed_self._get_axis_number(ax)

Edit: sorry @gabriellm1 I didn't see your comment!

[TomAugspurger]

whatsnew will go in v1.0.0.rst

…

On Mon, Oct 21, 2019 at 3:59 PM ellequelle ***@***.***> wrote: @TomAugspurger <https://github.com/TomAugspurger> do you mean something like this? Also: there doesn't seem to be a whatsnew file since 0.25.2, does this go into v0.25.3.rst? axis = self._get_axis_number(axis) if axis == 0: ax = self._info_axis_name _maybe_transposed_self = self elif axis == 1: _maybe_transposed_self = self.T ax = 1 else: ax = axis _maybe_transposed_self = self ax = _maybe_transposed_self._get_axis_number(ax) — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#29132?email_source=notifications&email_token=AAKAOIUOC55VGO2P4XPEPQ3QPYJ2TA5CNFSM4JC7IVS2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEB3YWGA#issuecomment-544705304>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAKAOISKWQ3QQ7MB53HPSNLQPYJ2TANCNFSM4JC7IVSQ> .

[coderkalyan]

Hi, I'm new to pandas development (but a fan). Can I try this as part of #hacktoberfest, and where do I start?