Re: [julia-users] broadcast performance is slow
Stefan, thanks. On Friday, 18 March 2016 15:11:22 UTC+1, Stefan Karpinski wrote: > > This was one of the motivating issues for the function redesign which is > now on Julia master. If you try the latest development version of Julia, > these should both be much faster. > > On Fri, Mar 18, 2016 at 8:50 AM, Igor Cerovsky > wrote: > >> Hello, >> >> By writing a code for unit normal scaling I've found a big differences >> related to where a function used in broadcast is defined, *globally* vs* >> locally*. Consider functions below: >> >> function scun2!(A) >> shift = mean( A, 1) >> stretch = std(A, 1) >> >> f(a, b, c) = (a - b) / c # defined locally >> broadcast!(f, A, A, shift, stretch) >> >> shift, stretch >> end >> >> f_scun(a, b, c) = (a - b) / c # defined globally >> function scun3!(A) >> shift = mean( A) >> stretch = std(A, 1) >> >> broadcast!(f_scun, A, A, shift, stretch) >> >> shift, stretch >> end >> >> Resulting performance is: >> >> R2 = copy(T) >> >> @time sh2, sc2 = scun2!(R2); >> >> 0.035527 seconds (19.51 k allocations: 967.273 KB) >> >> >> R3 = copy(T) >> >> @time sh3, sc3 = scun3!(R3); >> >> 0.009705 seconds (54 allocations: 17.547 KB) >> >> >> How can be explained, that if f_scun is defined outside the function the >> performance is 3.6 times better (number of allocations is also large)? I'm >> using Julia 0.4.3 >> >> Thank you, >> Igor >> >> >> >> >> >
Re: [julia-users] broadcast performance is slow
This was one of the motivating issues for the function redesign which is now on Julia master. If you try the latest development version of Julia, these should both be much faster. On Fri, Mar 18, 2016 at 8:50 AM, Igor Cerovsky wrote: > Hello, > > By writing a code for unit normal scaling I've found a big differences > related to where a function used in broadcast is defined, *globally* vs* > locally*. Consider functions below: > > function scun2!(A) > shift = mean( A, 1) > stretch = std(A, 1) > > f(a, b, c) = (a - b) / c # defined locally > broadcast!(f, A, A, shift, stretch) > > shift, stretch > end > > f_scun(a, b, c) = (a - b) / c # defined globally > function scun3!(A) > shift = mean( A) > stretch = std(A, 1) > > broadcast!(f_scun, A, A, shift, stretch) > > shift, stretch > end > > Resulting performance is: > > R2 = copy(T) > > @time sh2, sc2 = scun2!(R2); > > 0.035527 seconds (19.51 k allocations: 967.273 KB) > > > R3 = copy(T) > > @time sh3, sc3 = scun3!(R3); > > 0.009705 seconds (54 allocations: 17.547 KB) > > > How can be explained, that if f_scun is defined outside the function the > performance is 3.6 times better (number of allocations is also large)? I'm > using Julia 0.4.3 > > Thank you, > Igor > > > > >
[julia-users] broadcast performance is slow
Hello, By writing a code for unit normal scaling I've found a big differences related to where a function used in broadcast is defined, *globally* vs* locally*. Consider functions below: function scun2!(A) shift = mean( A, 1) stretch = std(A, 1) f(a, b, c) = (a - b) / c # defined locally broadcast!(f, A, A, shift, stretch) shift, stretch end f_scun(a, b, c) = (a - b) / c # defined globally function scun3!(A) shift = mean( A) stretch = std(A, 1) broadcast!(f_scun, A, A, shift, stretch) shift, stretch end Resulting performance is: R2 = copy(T) @time sh2, sc2 = scun2!(R2); 0.035527 seconds (19.51 k allocations: 967.273 KB) R3 = copy(T) @time sh3, sc3 = scun3!(R3); 0.009705 seconds (54 allocations: 17.547 KB) How can be explained, that if f_scun is defined outside the function the performance is 3.6 times better (number of allocations is also large)? I'm using Julia 0.4.3 Thank you, Igor
