Odd error message, concerning numpy using an alias versus numpy package name

[apiszcz]

This produces a poor warning

>>> import pandas as pd
>>> import numpy as np
>>> np.ptp(pd.Series([1, 2, 3]))
/lib/python3.7/site-packages/numpy/core/fromnumeric.py:2495: FutureWarning: Method .ptp is deprecated and will be removed in a future version. Use numpy.ptp instead.

because np.ptp calls Series.ptp, which is deprecated. We should instruct the user
to convert the Series to an array before calling ptp.

np.ptp(pd.Series([1, 2, 3]).array)

The warning is defined at

pandas/pandas/core/generic.py

Lines 10197 to 10199 in 20e4c18

	warnings.warn(
	"Method .ptp is deprecated and will be removed "
	"in a future version. Use numpy.ptp instead.",

---

In this case I am using numpy ptp, however the error messages states I am not

>>> dfagg = df.groupby(['idf']).agg({ 'f1' : [np.min, np.max, np.ptp] })
C:\test\lib\numpy\core\fromnumeric.py:2495: FutureWarning: Method .ptp is deprecated and will be removed in a future version. Use numpy.ptp instead.

NO warning
>>> dfagg = df.groupby(['idf']).agg({ 'f1' : [np.min, np.max, numpy.ptp] })

[DavidMStraub]

I'm also running into this warning. It's actually even simpler to reproduce:

import pandas as pd
import numpy as np
np.ptp(pd.Series([1, 2, 3]))

This will raise the warning. The reason, I think, is ironic: pandas warns against using the Series' method inspite of the function, but numpy, when using the function, tries if there is a method and uses that preferably:
https://github.com/numpy/numpy/blob/master/numpy/core/fromnumeric.py#L2541-L2548

[TomAugspurger]

We faced a similar issue with argmin / argmax: https://github.com/pandas-dev/pandas/pull/16955/files

The recommendation is to have people do np.ptp(pd.Series([1, 2, 3]).to_numpy()) in the meantime. Does anyone want to update the deprecation message?

[hasnain2808]

I would like to take up this issue but as I am new would need some support

[hasnain2808]

take

[jbrockmendel]

closed by #30458.