Pandas dropna throwing ValueError: "Cannot convert non-finite values (NA or inf) to integer" - pandas

Question

Pandas: 0.25.3
Python: 3.7.4
I have a data frame, and I want to remove the columns which contain only NaN values. That should be easy, because there is a Pandas DataFrame function which does exactly that—dropna. Here's my code:
long_summary = long_summary.dropna(axis='columns', how='all')
But that simple line throws an exception:
ValueError: Cannot convert non-finite values (NA or inf) to integer
I cannot see how calling dropna would lead to this exception. What is going on and how do I fix it?
I'll include the whole exception stack just-in-case that makes the problem clearer:
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-88-b4926abd4d81> in <module>
----> 1 long_summary = long_summary.dropna(axis='columns', how='all')
c:\users\timregan\appdata\local\programs\python\python37\lib\site-packages\pandas\core\frame.py in dropna(self, axis, how, thresh, subset, inplace)
4860 agg_obj = self.take(indices, axis=agg_axis)
4861
-> 4862 count = agg_obj.count(axis=agg_axis)
4863
4864 if thresh is not None:
c:\users\timregan\appdata\local\programs\python\python37\lib\site-packages\pandas\core\frame.py in count(self, axis, level, numeric_only)
7848 result = Series(counts, index=frame._get_agg_axis(axis))
7849
-> 7850 return result.astype("int64")
7851
7852 def _count_level(self, level, axis=0, numeric_only=False):
c:\users\timregan\appdata\local\programs\python\python37\lib\site-packages\pandas\core\generic.py in astype(self, dtype, copy, errors, **kwargs)
5880 # else, only a single dtype is given
5881 new_data = self._data.astype(
-> 5882 dtype=dtype, copy=copy, errors=errors, **kwargs
5883 )
5884 return self._constructor(new_data).__finalize__(self)
c:\users\timregan\appdata\local\programs\python\python37\lib\site-packages\pandas\core\internals\managers.py in astype(self, dtype, **kwargs)
579
580 def astype(self, dtype, **kwargs):
--> 581 return self.apply("astype", dtype=dtype, **kwargs)
582
583 def convert(self, **kwargs):
c:\users\timregan\appdata\local\programs\python\python37\lib\site-packages\pandas\core\internals\managers.py in apply(self, f, axes, filter, do_integrity_check, consolidate, **kwargs)
436 kwargs[k] = obj.reindex(b_items, axis=axis, copy=align_copy)
437
--> 438 applied = getattr(b, f)(**kwargs)
439 result_blocks = _extend_blocks(applied, result_blocks)
440
c:\users\timregan\appdata\local\programs\python\python37\lib\site-packages\pandas\core\internals\blocks.py in astype(self, dtype, copy, errors, values, **kwargs)
557
558 def astype(self, dtype, copy=False, errors="raise", values=None, **kwargs):
--> 559 return self._astype(dtype, copy=copy, errors=errors, values=values, **kwargs)
560
561 def _astype(self, dtype, copy=False, errors="raise", values=None, **kwargs):
c:\users\timregan\appdata\local\programs\python\python37\lib\site-packages\pandas\core\internals\blocks.py in _astype(self, dtype, copy, errors, values, **kwargs)
641 # _astype_nansafe works fine with 1-d only
642 vals1d = values.ravel()
--> 643 values = astype_nansafe(vals1d, dtype, copy=True, **kwargs)
644
645 # TODO(extension)
c:\users\timregan\appdata\local\programs\python\python37\lib\site-packages\pandas\core\dtypes\cast.py in astype_nansafe(arr, dtype, copy, skipna)
698 if not np.isfinite(arr).all():
699 raise ValueError(
--> 700 "Cannot convert non-finite values (NA or inf) to " "integer"
701 )
702
ValueError: Cannot convert non-finite values (NA or inf) to integer
(N.B. the data types of my columns are int64, Int32, and float64)
In the comments Scott asked for data to reproduce this issue. The redacted CSV is available on Dropbox here.
df = pd.read_csv('E:\\Temp\\dropna.csv')
df.dropna(axis='columns', how='all')
But be warned, the CSV is 3.3 GB and the resulting data frame has over 60 million rows. It tried cutting out rows, but it seems to need to be this long to trigger the error.