I can not come because this time of year is when I make most money and it
also costs quite a lot.
I see J and APL as very close and in the past the APL conferences were
moving around.
I wonder if it would be possible to connect with others like Dyalog and IBM
to create events in Europe?
Many wou
This approach won't work with different width characters(such as in Korean)
in unicode. (Some characters are wide -- taking two-character space, others
narrow -- taking one-character space)
On Wed, Apr 30, 2014 at 6:24 AM, Don Guinn wrote:
> This has bugged me for quite a while too. I tried mak
Suppose I have an Nx2 data matrix, say
x1 y1
x2 y2
x3 y3
and a dyadic verb called, say, df1
Is there an elegant way to get the result matrix
x1 df1 y1
x2 df1 y2
x3 df1 y3
.
without resorting to "from - {" and "take - {." jiggery-pokery
Or perhaps just a means to convert an expresiion fr
Like this?
df1 = :^
]Nx2 =: i.6 2x
0 1
2 3
4 5
6 7
8 9
10 11
df1/ |: Nx2
0 8 1024 279936 134217728 1000
Greetings,
Ben
From: [email protected]
[[email protected]] on behalf of alexgi
sorry, df1 =: ^
From: [email protected]
[[email protected]] on behalf of Ben Gorte - CITG
[[email protected]]
Sent: Wednesday, April 30, 2014 11:56
To: [email protected]
Subject: Re: [Jprogramming]
The thing that I'd think of first would be using ranks.
Using the same example:
Nx2 =: i.6 2x
df1 =: ^
df1/"1 Nx2
0 8 1024 279936 134217728 1000
--
For information about J forums see http://www.jsoftware.com/fo
Time to learn about rank!
In my experience
u"1 N_BY_2_MATRIX
runs faster than
u@:|: N_BY_2_MATRIX
Date: Wed, 30 Apr 2014 10:49:40 +0100
From: alexgian
To: J Programming forum
Subject: [Jprogramming] Applying a dyad to exusting matrix data
Message-ID:
Content-Type: text/plain; charset=
Yes, thanks Ben, that's exactly what I had in mind.
Clever. It never occurred to me to use / to insert the verb between the
"dyadic" data elements.
It might have if I had been using an N x 3 or more array, as I am used to
using / for series as an "accumulator", but in the trivial dyadic case, I
Sorry, but not on my computer:
Nx2 =: i.60 2x
df1=:+
6!:2 'z=:df1/ |:Nx2'
0.46775
6!:2 'z=:df1/"1 Nx2'
1.18481
I was thinking in the first case it does df1 only once (on a lot of data), and
in the second case many times, element by element.
Greetings,
Ben
__
I registered and look forward to see as many as possible. I do not see the cost
too much while I remember that everything else in J is free.
Anssi
-Original Message-
From: [email protected]
[mailto:[email protected]] On Behalf Of Björn Helga
Cost of flight and loss of earnings is the cost I am faced with.
It is actually quite considerable.
-
Björn Helgason
gsm:6985532
skype:gosiminn
On 30.4.2014 11:25, "Anssi Seppälä" wrote:
> I registered and look forward to see as many as possible. I do not see the
> cost too much while I remembe
Yeah, personally (from my minimal understanding of J), I'd have guessed
that feeding the data at rank 1 would be slower, but what do I know?
On 30 April 2014 12:12, Ben Gorte - CITG wrote:
> Sorry, but not on my computer:
>
>Nx2 =: i.60 2x
>df1=:+
>6!:2 'z=:df1/ |:Nx2'
> 0.4677
June is right. I will go further and say that if your chosen font is not
consistently fixed-width, your efforts to get tidy boxes are doomed to
failure. None of the "fixed width" fonts that I've tried (on the Macintosh)
are consistently fixed width -- certainly not over the C-J-K codespaces.
The ma
It is cool, and example looks good:
one shortcomming: (uses J display rather than data)
sbox 3}._32<\ u: a.
this almost lines up: (i think char 173/174 double prints?)
sbox <"1 u: (95 166 188 +"0 1 i.30) { a.
- Original Message -
From: Ian Clark
To: Programming forum
Cc:
Sent:
> one shortcomming: (uses J display rather than data)
Don't understand you.
sbox returns a result (of precision: unicode) which can be assigned.
z=: cutopen '€1.23 €45.67 €890.12'
zz=: sbox z
zz
+-+--+---+
|€1.23|€45.67|€890.12|
+-+--+---+
As for boxes not quite
I'm trying to combine/stitch two matrices into a boxed representation. I
tried various ideas for about 45 minutes and this is the best I came up
with.
I need the data to be boxed for another algorithm I'm supplying it to
The output should be a 5 item list of boxed 3x2 arrays with each column
side
this?
<"1 (i. 5 3) ,&<"0 i. 5 3
- Original Message -
From: Joe Bogner
To: [email protected]
Cc:
Sent: Wednesday, April 30, 2014 11:00:55 AM
Subject: [Jprogramming] stitching two matrices
I'm trying to combine/stitch two matrices into a boxed representation. I
tried various ide
arr -: P1 <@(,."1) P2
1
On Wed, Apr 30, 2014 at 8:00 AM, Joe Bogner wrote:
> I'm trying to combine/stitch two matrices into a boxed representation. I
> tried various ideas for about 45 minutes and this is the best I came up
> with.
>
> I need the data to be boxed for another algorithm I'm s
Dyadic ". is more efficient for converting strings to numbers. e.g.
p2 =: _999 ". ];._2 (0 : 0)
32130.00 0.01369 1.00568
32130.00 0.51354 1.0
32130.00 0.31124 1.0
32130.00 0.00591 1.0
32130.00 0.15562 1.0
)
On Wed, Apr 30, 2014 at 8:00 AM, Joe Bogner wrote:
> I'm trying to c
Roger, thank you. That is perfect. Also, I appreciate the tip on Dyadic ".
Your clue gave me another idea, which isn't as simple as yours
arr-: <"2 (P1 ,."1 P2)
1
I typically forget about atop and just move the verb outside and use
parentheses. This will help it stick.
On Wed, Apr 30, 201
Simpler: P1 <@,."1 P2
On Wed, Apr 30, 2014 at 8:22 AM, Joe Bogner wrote:
> Roger, thank you. That is perfect. Also, I appreciate the tip on Dyadic ".
>
> Your clue gave me another idea, which isn't as simple as yours
>
> arr-: <"2 (P1 ,."1 P2)
>
> 1
>
>
> I typically forget about atop and jus
I can't understand why
func1 =. 3&+ @: (2&-)
func2 =. (3&+) (2&-)
give different results as mondaic verbs.
func1 5 gives 0, which is what I would expect.
func2 5 gives 12, which I can't understand.
I would like to know what the difference is between func1 and func2. It is my
understanding that fo
func2 is a hook. The semantics of a hook are more complex than simple
composition.
On Apr 30, 2014 11:42 AM, "Jon Hough" wrote:
> I can't understand why
> func1 =. 3&+ @: (2&-)
> func2 =. (3&+) (2&-)
>
> give different results as mondaic verbs.
> func1 5 gives 0, which is what I would expect.
> f
Sorry @Jon, your understanding is faulty. You can't just omit the (@:) in
(u@:v).
J should come with a prominent health warning that u(v y) is not the same
as (u v)y. Yet when I do a low pass over the documentation as a newcomer
might see it, I don't think that fact is made sufficiently clear.
IM
Hi Jon,
Like so many things in J, an added character usually does make a difference.
That is the cost of using less to say more. :)
In this case, I think that you have a pretty good handle on func1, so I will
focus on what func2 is doing.
func2 is a hook which means that func2 5 is acting like
Raul Miller wrote:
> I was able to run plot i.21 on both jqt and jhs from j64-801
I tried it with jhs, j32-802. It does not have plot package preinstalled and
that machine does not go to the internet, so I cannot use JAL also.
How do I install plot for jhs manually?
The information from http://jsoftware.com/help/dictionary/special.htm should
probably be incorporated into http://www.jsoftware.com/jwiki/NuVoc?
Here's a list of the entries which are relevant in the context of ("1):
$, dyad also ($,)"r ; also {, {., }., e., ; avoids ravel; see the J 4.06
release
Also worth keeping in mind might be:
http://www.jsoftware.com/pipermail/general/2006-January/026271.html
"With the hindsight from 17 years of usage, I would define 2-verb trains
differently ... "
but note also
http://www.jsoftware.com/pipermail/general/2006-January/026274.html
"There is no int
The basic idea should be to get the graphics_plot_* archive for your system
from http://www.jsoftware.com/jal/j802/ and unpack it in your j addons
directory (try jpath '~addons' to see what this directory is).
Then go to the manifest.ijs which it contained and repeat the process for
the addons men
If only the rough spots of the other programming languages I use were so
beautiful as Hook.
--
For information about J forums see http://www.jsoftware.com/forums.htm
With the hindsight from 17 years of usage, I would
define 2-verb trains differently, viz.
(f g) y <-> f g y
x (f g) y <-> f (x g y)
a bit surprising to me:
I agree with monad, but existing dyad hook is useful, as per yesterday's
discussion. That proposal would make dyad equivalent
On Wed, Apr 30, 2014 at 1:15 PM, Andrew Nikitin wrote:
> Raul Miller wrote:
> > I was able to run plot i.21 on both jqt and jhs from j64-801
> I tried it with jhs, j32-802. It does not have plot package preinstalled
> and that machine does not go to the internet, so I cannot use JAL also.
> H
Perhaps we could emphasize that
df1/"1 Nx2
works for any dyadic df1 that can handle two scalars, whereas
df1/ |:Nx2'
only makes sense if df1 is able to deal with entire arrays at once (and then it
may be faster).
Is this right?
greetings,
Ben
Fro
([: f g) and f@:g are already equivalent in functionality, at least in
contexts which are independent of how they are phrased.
It might have been nice to have a third way of expressing the same concept.
This would make teaching people about J easier for those not interested in
the language and wou
the slight difference between ([: f g) and f@:g is that it is arguably cleaner
to apply an adverb to just g in the first version:
([: f g"1) vs. f@:(g"1)
Overall, the point that hooks are distracting from forks is a major hurdle,
that I do not help with my proposals other than to recommend igno
Maybe not elegant, but interesting.
]A=:i.4 2
]A=:i.4 2
0 1
2 3
4 5
6 7
f=: 13 :'(}."1 y)g }:"1 y'
f
}."1 g }:"1
g=:%
g f A
0
0.67
0.8
0.857143
g=: o.
g f A
0
_6.86451
85.7338
3.14155
Linda
-Original Message-
From:
([: f g"1) is equivalent to f@:(g"1)
([: f g)"1 is equivalent to f@:g"1 which could be expressed as (f@:g)"1
Conjunctions and adverbs stack while checking for a leftmost instance (to
determine where to start), this effectively reverses the right-to-left
parsing order of verbs, in the context of co
On 4/30/2014 2:21 PM, Raul Miller wrote:
The information from http://jsoftware.com/help/dictionary/special.htm should
probably be incorporated into http://www.jsoftware.com/jwiki/NuVoc?
Coming soon. Stay tuned.
Henry Rich
--
Good ideas Raul,
I like the purpose driven approach of using J in a practical environment. I
also think that the current labs provide a much more powerful and interactive
way to allow learners to explore new concepts. As I think of areas to explore
in the process of communicating J, I am leanin
I have been noticing your efforts with jhs and highlighting and coloring.
I think it would be good if you could write up a page on the
primitives, tools, references, whatever else you find most useful
here. It will also be interesting to see how your work evolves.
Thanks,
--
Raul
On Thu, May
40 matches
Mail list logo
