half
Of Michael Goodrich
Sent: Thursday, April 6, 2017 11:41 AM
To: [email protected]
Subject: Re: [Jprogramming] plot help
Beautiful!
On Thu, Apr 6, 2017 at 11:02 AM, Linda A Alvord
wrote:
> Michael, Here is an algebta 2 modificatgion of Cliff Reiter's example
> in his boo
I get the same result; What this is supposed to look like is attached. I
needed some parens to get it to parse properly. i had the wrong idea about
how to package the data for plot because the example I was following was
boxing/unboxing which I mistakenly thought I needed to do and so got
wrappe
y^2'
> f=:y`z
> plot x;f`:(0) x
>
> Maybe this will be a good starting place.
>
> Linda
> -Original Message-
> From: Programming [mailto:[email protected]] On
> Behalf Of Rob Hodgkinson
> Sent: Thursday, April 6, 2017 9:36 AM
> To: prog
ge-
From: Programming [mailto:[email protected]] On Behalf
Of Rob Hodgkinson
Sent: Thursday, April 6, 2017 9:36 AM
To: [email protected]
Subject: Re: [Jprogramming] plot help
Michael wanted to see the 2 lines (that are separated by the ,: ) but just
needed to add
Michael wanted to see the 2 lines (that are separated by the ,: ) but just
needed to add an outer set of parens around the first line, as below (you
should be able to plot this);
Try this plot, which is what Michael was seeking;
plot ( 0 {"1 (50 1$^((-i.50)%10)) ),: 0 {"1 Y
> On 6 Apr 2017,
When I do this:
mean =: +/ % #
R =: (50?1e3)%1e3
Y =: 50 1$^((-i.50)%10) + R-mean R
plot 0 {"1 (50 1$^((-i.50)%10)),: 0 {"1 Y
I get an overlay plot: http://imgur.com/a/szZKz
Maybe I am not seeing the problem?
What happens when you do it?
Thanks,
--
Raul
On Thu, Apr 6, 2017 at 8
Download the addons for stats/base and stats/distrib.
The file random.ijs contains normalrand.
The file normal.ijs contains lot of stuff.
On Thu, Apr 6, 2017 at 8:50 AM, Michael Goodrich wrote:
> Hi Raul,
>
> however I cant seem to find how to draw Gaussian samples in J without
> implementing th
Hi Rob,
I had hoped it was this simple! I think I got wrapped around the axle on
the boxing/unboxing going on the only the only overlay example i could
find. And so
plot (0 {"1 (50 1$^((-i.50)%10))),:(0 {"1 Y)
does the trick!
On Thu, Apr 6, 2017 at 8:51 AM, Rob Hodgkinson wrote:
> If I und
If I understand correctly, you just want to plot 2 x 1-D data vectors as you
say …
eg if data=: i.30, maybe plot data^2 and data as 2 line plots … without
specifying the x values.
data=:i.30
plot (data^2),:data
This plots with an integer x-axis (since not specified 0 … n being the number
Hi Raul,
Thanx for responding. There is nothing special about Y - it is a list of
50 numbers which is a machine learning fit of the supplied function and
could simply be considered a noisy version of it. I .e Y is something like
R =. (50?1e3)%1e3
Y =. 50 1$^((-i.50)%10) + R-mean R
where
Without an example value for Y, it is difficult to say what is going on.
(A significant question is: are we seeing an issue with plot, or is it
really that plot is doing things right but there is some sort of issue
with the data?)
--
Raul
On Thu, Apr 6, 2017 at 8:15 AM, Michael Goodrich
wrote
I am trying to plot two 1-d data vectors in an overlay plot namely:
plot 0 {"1 (50 1$^((-i.50)%10)),: 0 {"1 Y
Y is a data vector s.t.
$ Y
50 1
following the pattern:
'key sin(exp),cos(exp)' plot (;(sin,:cos)@:^) -: >: i:3j100
However is does not work:
The problem seems to be in how to
12 matches
Mail list logo
