BUG: fix non-existent variable in NDFrame.interpolate #29142
Conversation
NDFrame.interpolate fails if axis is specified by name and not integer
pandas/core/generic.py
Outdated
| 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 |
There was a problem hiding this comment.
It is impossible to get here; L734-736 can just be deleted
|
Can you add a test for the bug this fixes |
Added test_interpolate_axis_argument() to TestNDFrame
Absolutely! I added a test in the test_generic.TestNDFrame, is that reasonable? |
gfyoung
added
the
Missing-data
pandas/tests/generic/test_generic.py
Outdated
| @@ -948,3 +948,7 @@ def test_deprecated_get_dtype_counts(self): | |||
| df = DataFrame([1]) | |||
| with tm.assert_produces_warning(FutureWarning): | |||
| df.get_dtype_counts() | |||
|
|
|||
| @pytest.mark.parametrize("axis", [0, 1, "index", "columns", "rows"]) | |||
| def test_interpolate_axis_argument(self, axis): | |||
There was a problem hiding this comment.
Reference issue number above this line as a comment.
doc/source/whatsnew/v1.0.0.rst
Outdated
| @@ -249,6 +249,7 @@ Bug fixes | |||
| ~~~~~~~~~ | |||
|
|
|||
| - Bug in :meth:`DataFrame.to_html` when using ``formatters=<list>`` and ``max_cols`` together. (:issue:`25955`) | |||
| - Bug in :func:`pandas.core.generic.NDFrame.interpolate` where specifying axis by name references variable before it is assigned (:issue:`29132`) | |||
There was a problem hiding this comment.
These are user facing, so we shouldn't reference NDFrame. Instead, refer to just :meth:`DataFrame.interpolate`, even if it affects Series as well.
pandas/tests/frame/test_missing.py
Outdated
| @@ -872,10 +873,22 @@ def test_interp_rowwise(self): | |||
| result = df.interpolate(axis=1, method="values") | |||
| assert_frame_equal(result, expected) | |||
|
|
|||
| # GH 29142: test axis names | |||
There was a problem hiding this comment.
rather than adding tests like this, can you create a new test function and paramterize on the axis
There was a problem hiding this comment.
I've added a new parametrized function in the latest commit
There was a problem hiding this comment.
pytest is returning with an uncaught warning in the same file from a call to df.median(). Should I add tm.assert_produces_warning()?
pandas/tests/frame/test_missing.py::TestDataFrameMissingData::test_fillna_categorical_nan
python3.7/site-packages/numpy/lib/nanfunctions.py:1115: RuntimeWarning: All-NaN slice encountered
overwrite_input=overwrite_input)
Also moved whatsnew comments
RuntimeWarning occurs due to a call to df.median (np.nanmedian)
|
thanks @ellequelle |
👍 |
black pandasgit diff upstream/master -u -- "*.py" | flake8 --diff