I found the following behaviour in Python/NumPy somewhat strange:
In [51]: a = np.arange(10, 20)
In [52]: a = a / 10.0
In [53]: a
Out[53]: array([ 1. , 1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9])
In [54]: a = np.arange(10, 20)
In [55]: a /= 10.0
In [56]: a
Out[56]: array([1, 1, 1, 1, 1, 1, 1, 1, 1, 1])
I felt that a=a/10.0 and a/=10.0 should return the same result. Is this intended and documented somewhere?
The problem with a /= 10.0 is that it modifies the array in place, and it won't change the the dtype of the array, so all the floats are converted to integers. On the other hand a = a / 10.0 created a new array, and the type can be changed if a new array is being created.
From docs:
Note that assignments may result in changes if assigning higher types to lower types (like floats to ints) or even exceptions (assigning complex to floats or ints):
