BUG: DataFrame.agg - why numpy.size doesn't work?

[aliceliu9988]

• [ x] I have checked that this issue has not already been reported.

• [ x] I have confirmed this bug exists on the latest version of pandas.

• (optional) I have confirmed this bug exists on the master branch of pandas.

---

Note: Please read this guide detailing how to provide the necessary information for us to reproduce your bug.

Code Sample, a copy-pastable example

df = pd.DataFrame([[1, 2, 3],
                   [4, 5, 6],
                   [7, 8, 9]],
                  columns=['A', 'B', 'C'])

df.agg({'A':['mean','std','size']})

import numpy as np
#Somehow this just doesn't work with DF.agg but works with DFGroupby.agg
df.agg({'A':[np.mean,np.std,np.size]})

Problem description

Intuitively, I assumed df.agg({'A':[np.mean,np.std,np.size]}) should work as df.agg({'A':['mean','std','size']}) does, but it doesn't. I wonder why? Looked through docs like the below but still didn't get it:

• https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.agg.html
• https://pandas.pydata.org/pandas-docs/version/0.22/generated/pandas.core.groupby.DataFrameGroupBy.agg.html

Expected Output

A

4.0
3.0
4.0

####

Output of *df.agg({'A':[np.mean,np.std,np.size]})

---

TypeError Traceback (most recent call last)
~\anaconda3\lib\site-packages\pandas\core\base.py in _aggregate_multiple_funcs(self, arg, _axis)
553 try:
--> 554 return concat(results, keys=keys, axis=1, sort=False)
555 except TypeError:

~\anaconda3\lib\site-packages\pandas\core\reshape\concat.py in concat(objs, axis, join, ignore_index, keys, levels, names, verify_integrity, sort, copy)
280 copy=copy,
--> 281 sort=sort,
282 )

~\anaconda3\lib\site-packages\pandas\core\reshape\concat.py in init(self, objs, axis, join, keys, levels, names, ignore_index, verify_integrity, copy, sort)
356 )
--> 357 raise TypeError(msg)
358

TypeError: cannot concatenate object of type '<class 'float'>'; only Series and DataFrame objs are valid

During handling of the above exception, another exception occurred:

ValueError Traceback (most recent call last)
in
1 import numpy as np
----> 2 df.agg({'A':[np.mean,np.std,np.size]})

~\anaconda3\lib\site-packages\pandas\core\frame.py in aggregate(self, func, axis, *args, **kwargs)
6704 result = None
6705 try:
-> 6706 result, how = self._aggregate(func, axis=axis, *args, **kwargs)
6707 except TypeError:
6708 pass

~\anaconda3\lib\site-packages\pandas\core\frame.py in _aggregate(self, arg, axis, *args, **kwargs)
6718 result = result.T if result is not None else result
6719 return result, how
-> 6720 return super()._aggregate(arg, *args, **kwargs)
6721
6722 agg = aggregate

~\anaconda3\lib\site-packages\pandas\core\base.py in _aggregate(self, arg, *args, **kwargs)
426
427 try:
--> 428 result = _agg(arg, _agg_1dim)
429 except SpecificationError:
430

~\anaconda3\lib\site-packages\pandas\core\base.py in _agg(arg, func)
393 result = {}
394 for fname, agg_how in arg.items():
--> 395 result[fname] = func(fname, agg_how)
396 return result
397

~\anaconda3\lib\site-packages\pandas\core\base.py in _agg_1dim(name, how, subset)
377 "nested dictionary is ambiguous in aggregation"
378 )
--> 379 return colg.aggregate(how)
380
381 def _agg_2dim(name, how):

~\anaconda3\lib\site-packages\pandas\core\series.py in aggregate(self, func, axis, *args, **kwargs)
3686 # Validate the axis parameter
3687 self._get_axis_number(axis)
-> 3688 result, how = self._aggregate(func, *args, **kwargs)
3689 if result is None:
3690

~\anaconda3\lib\site-packages\pandas\core\base.py in _aggregate(self, arg, *args, **kwargs)
484 elif is_list_like(arg):
485 # we require a list, but not an 'str'
--> 486 return self._aggregate_multiple_funcs(arg, _axis=_axis), None
487 else:
488 result = None

~\anaconda3\lib\site-packages\pandas\core\base.py in _aggregate_multiple_funcs(self, arg, _axis)
562 result = Series(results, index=keys, name=self.name)
563 if is_nested_object(result):
--> 564 raise ValueError("cannot combine transform and aggregation operations")
565 return result
566

ValueError: cannot combine transform and aggregation operations

[attack68]

Actually this works:

df = pd.DataFrame([[1, 2, 3],
                   [4, 5, 6],
                   [7, 8, 9]],
                  columns=['A', 'B', 'C'])

df.agg({'A':['mean','std']})
df.agg({'A':[np.mean,np.std]})

The only thing relevant to your issue is:

df.agg({'A':['size']})
df.agg({'A':[np.size]})

[attack68]

@aliceliu9988 can yo please add a descriptive title for the issue.

[aliceliu9988]

Hi @attack68,

Thanks for helping pinpoint the problem and reminding me to update the question title.

I tried this: print("df.agg({'A': [np.size]}) is :",df.agg({'A':[np.size]}))

It did go through, but the output is not the row count of column A (3), but this:

df.agg({'A': [np.size]}) is : A
size
0 1
1 1
2 1
3 1

I hope someone knows why.

[rhshadrach]

Internally, np.size is evaluated on a Series. For a UDF, .agg uses .apply which then operates row-by-row. This makes the result a transform, and the others being aggregations caused the error mentioned in the OP.

.agg using .apply being undesirable is mentioned as part of #41112, but I think this issue should stay open.

[aliceliu9988]

Hi Richard @rhshadrach,

Thanks for your explanation. I am a beginner of Python and do appreciate your hints and will look more into the difference between 'transform' and 'aggregation. Ah... I have to say as a beginner I didn't expect the syntax to behave inconsistently like this.

[rhshadrach]

@aliceliu9988

I have to say as a beginner I didn't expect the syntax to behave inconsistently like this.

I realize now I wasn't very clear, but I was trying to say the same thing! Thanks for raising this issue.

[kwhkim]

Wow, this looks serious. I have another example.

>>> df.agg(np.size)
A    3
B    3
C    3
dtype: int64
>>> df.agg({'A':np.size})
   A
0  1
1  1
2  1

so df.agg({'A':}) is more like df.A.agg()?

>>> df.A.agg(np.size)
0    1
1    1
2    1
Name: A, dtype: int64

It gets weirder

>>> df.groupby([1]*len(df)).agg({'A':np.size})  # [1]*len(df) makes the whole rows as group 1
   A
1  3