You may think that it is a waste of memory to use an int64 data type if the range of your values is so limited.
In fact, conscious of data-intensive situations, you can calculate how much memory space your Array_1 object is taking:
In: import numpy as np Array_1.nbytesOut: 24
Please note that on 32-bit platforms (or when using a 32-bit Python version on a 64-bit platform), the result is 12.
In order to save memory, you can specify the type that best suits your array beforehand:
In: Array_1 = np.array(list_of_ints, dtype= 'int8')
Now, your simple array occupies just a fourth of the previous memory space. It may seem an obvious and overly simplistic example, but when dealing with millions of rows and columns, defining ...
