Numpy dot too clever about symmetric multiplications

Question

Anybody know about documentation for this behaviour?

import numpy as np
A  = np.random.uniform(0,1,(10,5))
w  = np.ones(5)
Aw = A*w
Sym1 = Aw.dot(Aw.T)
Sym2          


        

Answer 1

I suspect this is to do with promotion of intermediate floating point registers to 80 bit precision. Somewhat confirming this hypothesis is that if we use fewer floats we consistently get 0 in our results, ala

A  = np.random.uniform(0,1,(4,2))
w  = np.ones(2)
Aw = A*w
Sym1 = Aw.dot(Aw.T)
Sym2 = (A*w).dot((A*w).T)
diff = Sym1 - Sym2
# diff is all 0's (ymmv)