[plt-scheme] help on how to write a frequency-counting function in a more functional way

Hashes might be an even bigger win on datasets constructed to be
sparse, because the vector would allocate many unused slots that hash
tables would not.  And it might be interesting to throw imperative
hashes into the mix.

--Carl

On Sun, Apr 19, 2009 at 1:59 PM, Matthias Felleisen
<matthias at ccs.neu.edu> wrote:
>
> I forgot about functional hash:
>
>  vector: cpu time: 3580 real time: 3626 gc time: 0
>  alist:  cpu time: 7784 real time: 8068 gc time: 1380
>  bst:    cpu time: 15791 real time: 16052 gc time: 368
>  hash:   cpu time: 7439 real time: 7646 gc time: 456
>
> Not bad. -- Matthias