Thanks both.
There's been quite a lot of subsequent correspondence about the relative
performance
of u/\ and u/\..
What I was actually griping about was the relatively poor performance of
M&|@:u (distributed across an array in some manner) compared to plain u,
as here with add32 cf +
Hmm...
Here's another gerundy example you might want to tag for future reading...
http://rosettacode.org/wiki/Odd_word_problem#J
Honestly, though, most of this stuff is for play rather than serious
work. Then again, the only way you get good at this stuff is if you
play with it. Serious work ten
On Tue, Aug 26, 2014 at 12:05 PM, Raul Miller wrote:
> P.S. example use:
>
> trampoline 3 peano 5
> 8
>
Thank you. This whole thread has pushed me well beyond my comfort
zone. I realize that these examples are simply illustrative of the
trampoline concept and wouldn't be used in the real worl
Raul already described the problem; I often use an adverb (scan) instead of
(/\.) to solve this kind of issues:
u ( scan=. ((~/)\.)(&.|.) )
u~/\.&.|.
Again, the difference between these adverbs can be appreciated by means of
symbolic nouns and verbs:
( YS=. <;._1 ' Y0 Y1 .. YN' )
┌──┬──┬──
Raul wrote:
"
http://weblambdazero.blogspot.com/2010/07/advanced-recursion-in-newlisp.html
looks like it is describing this:
memoize=: M.
"
Yes; however, a trampoline function was defined further down. In any case,
this is not very relevant now because…
Raul wrote:
"
I should also add an imp
I should also add an implementation of the "tco" at the very bottom of that
post:
tco=:(0 :0)(2 :0)
:
context=. cocreate'' NB. context
active__context=: 0
accumulated__context=: ''
v 2 :m context
)
:
accumulated__n=: x,&/pair
end.
active__n=. 0
value__n
end.
)
sum=: (4 :0)
P.S. example use:
trampoline 3 peano 5
8
Thanks,
--
Raul
On Tue, Aug 26, 2014 at 12:04 PM, Raul Miller wrote:
> (I thought I sent this message, this morning, but I see that it has not
> yet been sent. I have been having this happen a lot recently. I have a
> hunch about what is happeni
(I thought I sent this message, this morning, but I see that it has not yet
been sent. I have been having this happen a lot recently. I have a hunch
about what is happening, but for now I'll just note that I am having this
issue, and I apologize if it seems I am ignoring anyone.)
-
e of $:, but I'm not sure there is full TCO (completely in
place sort), even though it doesn't crash the stack on medium data.
- Original Message -----
From: Joe Bogner
To: programm...@jsoftware.com
Cc:
Sent: Tuesday, August 26, 2014 6:10:28 AM
Subject: Re: [Jprogramming] Adler-
On Mon, Aug 25, 2014 at 9:51 PM, Raul Miller wrote:
> http://www.integralist.co.uk/posts/understanding-recursion-in-functional-javascript-programming/#the-solution
> looks like it is describing this:
>
> trampoline=:3 :0
>try.
> while. 1 do.
> y=. y`:6''
> end.
>catch.
ound value is only 0
after call.
11
('blast off!'"_^:(0 = [) ; <:@[) lexT 'f' 11
blast off!
timing is not terrible. timespacex causes 2 calls to function.
erase 'f'
1
('blast off!'"_^:(0 = [) ; <:@[) lexT 'f' 1000
adverbs in general, is that I can type
>> boxscan in source code, but the actual definition will be "compiled away"
>> as if it were called with f. . using the term boxscan can help with
>> readability if the intent of the function is to create a rightmost
>> accum
neat feature regarding tacit adverbs in general, is that I can type
>> boxscan in source code, but the actual definition will be "compiled away"
>> as if it were called with f. . using the term boxscan can help with
>> readability if the intent of the function is to
but the actual definition will be "compiled away"
> as if it were called with f. . using the term boxscan can help with
> readability if the intent of the function is to create a rightmost
> accumulator state.
>
>
>
>
> - Original Message -
> From: Raul
((65521 | {.@:] ({:@:] ,~ [ + +/@:}:@:]) [: +/\ [,~{:@:]) boxscan)
> ad32=: 65521 | [: ({: ,~ <:@:+/) ad32c
>
>ad32 ( ad32c |. 0 0 ; (1, a. i. 'Wik') ; a. i. 'ip') ;~ a. i.'edia'
> 4582 920
>
>
> - Original Message -
> Fr
27;ip') ;~ a. i.'edia'
4582 920
- Original Message -
From: Raul Miller
To: Programming forum
Cc:
Sent: Monday, August 25, 2014 3:15:48 PM
Subject: Re: [Jprogramming] Adler-32
On Mon, Aug 25, 2014 at 2:40 PM, 'Pascal Jasmin' via Programming
wrote:
> boxscan can
On Mon, Aug 25, 2014 at 2:40 PM, 'Pascal Jasmin' via Programming
wrote:
> boxscan can use any list of arguments as the x, and appends any
> shape as the y, as long x u y results in a shape "compatible with
> y" (more accurately compatible with future applications of x u
> result .
boxscan can use
-- Original Message -
From: Raul Miller
To: Programming forum
Cc:
Sent: Monday, August 25, 2014 1:16:58 PM
Subject: Re: [Jprogramming] Adler-32
I have a nitpick with your phrasing here.
Instead of saying "... use / recursively... (tail call optimization)"
I think you should say
510 <\ 1, Randdata'
> 0.412953 5.36889e8
>timespacex '(65521 | [: ( {: , <:@:+/) (65521 | {.@:] ({:@:] ,~ [ +
> +/@:}:@:]) [: +/\ [,~{:@:]) boxscan) |. 0 0 ; _151 <\ 1, Randdata'
> 0.414528 3.77498e8
>
> 4096 to 32764 chunks are the fastest, a
0.414528 3.77498e8
4096 to 32764 chunks are the fastest, and use the same memory. 4096 would work
best on 32 bit J for character data
- Original Message -
From: bill lam
To: 'Pascal Jasmin' via Programming
Cc:
Sent: Sunday, August 24, 2014 11:04:01 PM
Subject: Re: [Jpro
; is that it provides the basis for s2
>
> <: +/ 1 88 193 300 405 517 618 718 823 920
> 4582
>
>
>
> - Original Message -
> From: 'Pascal Jasmin' via Programming
> To: "programm...@jsoftware.com"
> Cc:
> Sent: Sunday, August 24, 2014
;programm...@jsoftware.com"
Cc:
Sent: Sunday, August 24, 2014 11:48:23 AM
Subject: Re: [Jprogramming] Adler-32
sorry that version doesn't work on a stream. will post one later.
- Original Message -
From: bill lam
To: 'Pascal Jasmin' via Programming
Cc:
Sent: Sunday
sorry that version doesn't work on a stream. will post one later.
- Original Message -
From: bill lam
To: 'Pascal Jasmin' via Programming
Cc:
Sent: Sunday, August 24, 2014 11:15:57 AM
Subject: Re: [Jprogramming] Adler-32
Did you mean _4]\ ?
Anyway, from the wikiped
1, a. i. 'Wikipedia'
> 920 4582
>
>
>
>
> - Original Message -----
> From: bill lam
> To: Programming forum
> Cc:
> Sent: Sunday, August 24, 2014 7:35:26 AM
> Subject: Re: [Jprogramming] Adler-32
>
> I realized that my loop tr
47 2.68445e8
65521&|@:([: ({: , [: <: +/) +/\)"1 ] _524288 ]/ 1 ,~ Randdata
46312 20147
- Original Message -
From: 'Pascal Jasmin' via Programming
To: "programm...@jsoftware.com"
Cc:
Sent: Sunday, August 24, 2014 10:01:41 AM
Subject: Re: [Jprogrammin
On Sun, Aug 24, 2014 at 6:43 AM, mike_liz@tiscali.co.uk
wrote:
> I continue to be puzzled by the poor performance of something
> like add32/\ compared to +/\ .
The answer, here, can be explained in terms of an issue which I remember
being taught in gradeschool - "the associative law". (Of co
reduction at the end.
([: ({: , [: <: +/) +/\)"1 ] _4 ]/ 1, a. i. 'Wikipedia'
920 4582
- Original Message -
From: bill lam
To: Programming forum
Cc:
Sent: Sunday, August 24, 2014 7:35:26 AM
Subject: Re: [Jprogramming] Adler-32
I realized that my loop t
I realized that my loop translation solution was equally faulty in its
final step. It should be
({: (23 b.) 16&(33 b.)@{.) s2,s1
On Aug 24, 2014 6:40 PM, "bill lam" wrote:
> J32 should give a 32-bit integer so that the last train
> should be
> [: ({: (23 b.) 16&(33 b.)@{.)
> instead of
> 65536
What overflow problem I thought when I came up with
add32 =: 65521&|@:
+ NB. add modulo 65521
a32mdslow =: ((1 add32{:)(+ 16 (33 b.)])
(#add32(add32/)))@:(add32/\)@(a.&i.)
and then I tested it on modest
size input and found it was ORDERS
of magnitude slower than Bill's
loopy version!
So
J32 should give a 32-bit integer so that the last train
should be
[: ({: (23 b.) 16&(33 b.)@{.)
instead of
65536#.
This shoud be good enough. Thanks.
Вс, 24 авг 2014, Jan-Pieter Jacobs написал(а):
> You could incorporate the modulo operation in the summation directly:
>
> As per Raul's solut
You could incorporate the modulo operation in the summation directly:
As per Raul's solution, with the modulo addition:
a32_2=: 65536#. _1 0+ [:((65521|+)/ , {.) [: (65521|+)/\. 1,~ a.i. |.
NB. Test on 10M of random data
Randdata =: a.{~ ?1000$256
timespacex 'r1=:a32_2 Randdata'
7.26671 2.684
Since s2 is quadratic, it overflows quickly in J32. However
ieee floating points have 52 bits mantissa, so it can still
give correct result for medium size data albeit overflowed.
9!:11[16
(3!:0;]) a32 1e6{.a.
Вс, 24 авг 2014, Raul Miller написал(а):
> Hmm... actually there might be a floating
Hmm... actually there might be a floating point overflow problem for
gigabyte length arguments. I do not have the patience, however, to
find what (for example) adler32 1e9#{:a. is.
Thanks,
--
Raul
On Sun, Aug 24, 2014 at 4:37 AM, Raul Miller wrote:
> I believe this is equivalent, and a bit f
I believe this is equivalent, and a bit faster:
a32=: 65536#. 65521| _1 0+ [:(+/ , {.) 65521| [:+/\. 1,~ a.i. |.
a32 'Wikipedia'
300286872
a32 'The quick brown fox jumps over the lazy dog'
1541148634
Thanks,
--
Raul
On Sun, Aug 24, 2014 at 4:05 AM, bill lam wrote:
> Does anyone have a
Does anyone have a J-ish solution to adler32 checksum
and avoid overflow to floating points ?
Adler-32 is composed of two sums accumulated per byte: s1 is
the sum of all bytes, s2 is the sum of all s1 values. Both sums
are done modulo 65521. s1 is initialized to 1, s2 to zero. The
Adler-32 check
35 matches
Mail list logo
