'and' (boolean) vs '&' (bitwise) - Why difference in behavior with lists vs numpy arrays?

Question

What explains the difference in behavior of boolean and bitwise operations on lists vs NumPy arrays?

I\'m confused about the appropriate use of

Answer 1

1. In Python an expression of X and Y returns Y, given that bool(X) == True or any of X or Y evaluate to False, e.g.:

   True and 20 
   >>> 20
   
   False and 20
   >>> False
   
   20 and []
   >>> []
   

2. Bitwise operator is simply not defined for lists. But it is defined for integers - operating over the binary representation of the numbers. Consider 16 (01000) and 31 (11111):

   16 & 31
   >>> 16
   

3. NumPy is not a psychic, it does not know, whether you mean that e.g. [False, False] should be equal to True in a logical expression. In this it overrides a standard Python behaviour, which is: "Any empty collection with len(collection) == 0 is False".

4. Probably an expected behaviour of NumPy's arrays's & operator.