On Tue, 2007-02-27 at 16:51 +0000, Claus Reinke wrote:
okay, profiling was not available for the Haskell version back then, but using ML profiling to improve a Haskell version sounds highly dangerous to me, even more so if the authors do not even mention any awareness of this danger. in 3.5.1, we see Alleles as a BoolVector, which sounds fine until we see it converted to its list of associations, which is then foldl-ed (!) over with a non-strict function (good that those chromosome lengths appear to be only 60..), for every evaluation. this is the main evaluation function for fitness, so it should be very much inner loop, with population sizes and generations ranging to 7500/700 and 50/150000. . of course, the same function in the ML version just means running a loop over a vector, with a strict accumulator..
It'd be interesting to get the real code for this. Partly to just try optimising it but more so as a real test case for list/array fusion. As far as I see, there's no reason that consuming an assoc list of a bool vector with a foldl' (the ' is probably essential) should be slow. If it's fused properly no list cells should ever be allocated, we should get the loop over the vector. Duncan