Re: How to apply a list of functions

*To*: mathgroup at smc.vnet.net*Subject*: [mg112887] Re: How to apply a list of functions*From*: Leonid Shifrin <lshifr at gmail.com>*Date*: Tue, 5 Oct 2010 05:32:40 -0400 (EDT)

Vince, How about this: In[51]:= Inner[Compose,Hold[f1,f2,f3],Hold[{{1,2},{3,4}},{{5,6},{7,8}},{{9,10},{11,12}}],List] Out[51]= {f1[{{1,2},{3,4}}],f2[{{5,6},{7,8}}],f3[{{9,10},{11,12}}]} Basically, wrap your arguments in Hold rather than List, and you don't have this problem. Regards, Leonid On Mon, Oct 4, 2010 at 9:37 PM, Vincent N. Virgilio <virgilio at ieee.org>wrote: > Leonid, > > I notice that Inner/Compose is a little fragile if x is, say, a list of > matrices, and the goal is to apply each f to each whole matrix. > MapThread/Compose handles that transparently. > > Thanks again, > > Vince > > On Mon, Oct 4, 2010 at 1:07 PM, Leonid Shifrin <lshifr at gmail.com> wrote: > >> Hi Vince, >> >> First, you could use Compose in your code - perhaps it is a little faster: >> >> In[2]:= MapThread[Compose,{{f1,f2,f3},{x1,x2,x3}}] >> >> Out[2]= {f1[x1],f2[x2],f3[x3]} >> >> Using Inner can be a bit faster still: >> >> In[4]:= Inner[Compose,{f1,f2,f3},{x1,x2,x3},List] >> >> Out[4]= {f1[x1],f2[x2],f3[x3]} >> >> In most cases, you probably won't see the difference in performance, >> unless your functions >> do very little and you have lots of them. >> >> Regards, >> Leonid >> >> >> >> >> On Mon, Oct 4, 2010 at 2:06 PM, Vince Virgilio <blueschi at gmail.com>wrote: >> >>> On Oct 3, 3:39 am, Sam Takoy <sam.ta... at yahoo.com> wrote: >>> > Hi, >>> > >>> > As a follow up to my question about apply {Sin, Cos} to x, I came up >>> > with the tasteless >>> > >>> > Map[Apply[#, {x}] &, {Sin, Cos}] >>> > >>> > but I expect that the pros in this ng will be able to improve upon it. >>> > >>> > Thanks, >>> > >>> > Sam >>> >>> Tangentially, >>> >>> I've always wondered if there was a better way to apply a list of >>> functions to a list of arguments. Here's how I do it, where f and x >>> are the respective lists. >>> >>> MapThread[#1@#2 &, {f, x}] >>> >>> Seems performant. >>> >>> Anyone? >>> >>> Vince >>> >>> >> >