Am 21.03.2015 um 05:13 schrieb Travis Oliphant <travis@continuum.io>:

I have not tested this, but I suspect that the MATLAB routine is using the erf implementation from the Intel Math Kernel Libraries (MKL).    

There is a function in MKL called vdErf that takes a vector of doubles and is likely tuned to the hardware.    This could be linked to NumPy with similar speed benefits. 

-Travis


The uvml module https://github.com/geggo/uvml exposes the fast MKL/VML erf implementation to numpy. Unfortunately, no binaries available.

Gregor