CODEWITHSUNDEEP
pythonwarnings
Medulla Oblongata
Medulla Oblongata

Reputation: 3973

Python deprecation warning involving floats and integers

I want to select a small part of a large array and find the average value of that subset. I have tried to specify integers when defining the subset:

import numpy as np

x = np.linspace(-10,10,1e6) # whole dataset

x0 = x[int(len(x)//2-5):int(len(x)//2+5)] # subset
print(x0)
xm = np.mean(x0) # average value of data subset
print(xm)

but my code gives a deprecation warning that says:

DeprecationWarning: object of type <class 'float'> cannot be safely interpreted as an integer.
  x = np.linspace(-10,10,1e6)

Is there a better way of calculating the average of the data subset? What should I do about this warning and will it become a problem in newer versions of Python? I'm using Spyder 3.2.8.

Upvotes: 2

Views: 5002

Answers (1)

301_Moved_Permanently
301_Moved_Permanently

Reputation: 4196

The problem is that np.linspace expect the number of points it should produce as its third argument. Thus, this should be a whole number (an integer). However 1e6 is parsed as a float, hence the need to convert to an integer, hence the warning.

The solution is to write 1e6 as an integer, i.e. 1000000. If you’re using Python 3, you can write 1_000_000 instead to make the number more readable.

Upvotes: 4

Related Questions