Thanks. Linda
-Original Message-
From: programming-boun...@forums.jsoftware.com
[mailto:programming-boun...@forums.jsoftware.com] On Behalf Of Henry Rich
Sent: Friday, November 28, 2014 10:34 PM
To: programm...@jsoftware.com
Subject: Re: [Jprogramming] Better way to locate coomplex or rea
That does it! Linda
-Original Message-
From: programming-boun...@forums.jsoftware.com
[mailto:programming-boun...@forums.jsoftware.com] On Behalf Of tom arneson
Sent: Friday, November 28, 2014 10:29 PM
To: programm...@jsoftware.com
Subject: Re: [Jprogramming] Get unique results each time
No matter where I insert either version I always get the same final list.
7?7
1 4 5 0 6 2 3
10?10
5 1 3 2 4 0 7 8 6 9
NB.m3=: 9!:1@<.@(+/ .^&2@(6!:0@])) NB.Randomizing random number seed
16!:45
16!:45
7?7
1 5 6 4 0 2 3
10?10
4 3 1 6 7 9 5 2 0 8
10?10
9 0 1 6 4 2 3 8 5 7
Li
Here's another way. From addons/stats/base/random.ijs
NB. =
NB.*randomize v sets a random value into random link
randomize=: 3 : 0
try.
require 'guid'
tmp=. _2 (3!:4) , guids 1
catch. NB. general/misc/guid.ijs not availabl
]A=:(i:2)j./i:2
_2j_2 _2j_1 _2 _2j1 _2j2
_1j_2 _1j_1 _1 _1j1 _1j2
0j_2 0j_1 0 0j1 0j2
1j_2 1j_1 1 1j1 1j2
2j_2 2j_1 2 2j1 2j2
(0 ~: 11&o.) A
1 1 0 1 1
1 1 0 1 1
1 1 0 1 1
1 1 0 1 1
1 1 0 1 1
Henry Rich
On 11/28/2014 10:27 PM, Linda Alvord wrote:
A is an array of real and co
Linda,
Run m3 and it sets the seed of the random number generator (RNG) to new
values based on the time. Do this before you start to generate random
numbers. With a new seed you'll get a different sequence of RNs.
-Original Message-
From: programming-boun...@forums.jsoftware.com
[mailto:
A is an array of real and complex numbers. Is there an easier way to locate
all the complex numbers. Signum locates each numbers intersection with a
line from it to the origin and the circle.
]A=:(i:2)j./i:2
_2j_2 _2j_1 _2 _2j1 _2j2
_1j_2 _1j_1 _1 _1j1 _1j2
0j_2 0j_1 0 0j1 0j2
1j_2 1j_
This agrees with {Pascal's answer. I thought you could take some number of
seconds in the time and find the last digits or something and then roll that
number and then begin your code. Maybe I isn't possible.
Linda
-Original Message-
From: programming-boun...@forums.jsoftware.com
[mailt
This gave a new set of numbers, but they are repeated the next time I run
the code.
Linda
-Original Message-
From: programming-boun...@forums.jsoftware.com
[mailto:programming-boun...@forums.jsoftware.com] On Behalf Of 'Pascal
Jasmin' via Programming
Sent: Friday, November 28, 2014 9:15 P
Try this, from: http://www.jsoftware.com/jwiki/JPhrases/RandomNumbers
m3=: 9!:1@<.@(+/ .^&2@(6!:0@])) NB.Randomizing random number seed
-Original Message-
From: programming-boun...@forums.jsoftware.com
[mailto:programming-boun...@forums.jsoftware.com] On Behalf Of Linda Alvord
Sent: F
9!:45 will set the seed. J on startup has the same seed.
For the default generator, 9!:45 will take a long array as the seed (basically
a polynomial representation of a 19000 bit number)
If you use the first random number generator (gbflip), the seed is a 31bit
number. I've posted ways to get
7?7
1 4 5 0 6 2 3
10?10
5 1 3 2 4 0 7 8 6 9
Now run the code again after closing J and you get the same results. How
can you get different results each time. I vaguely rember using some date or
other number with roll to insure different results.
Linda
--
x=: 'case_id', LF, 'GCTAGTCG', LF, 'ACGTC', LF
;:' '((x=LF)#i.$x)}x
┌───┬┬─┐
│case_id│GCTAGTCG│ACGTC│
└───┴┴─┘
Linda
-Original Message-
From: programming-boun...@forums.jsoftware.com
[mailto:programming-boun...@forums.jsoftware.com]
Updated. Thanks.
On 28.11.2014, at 22:33, Vijay Lulla wrote:
> OOPS! There was a typo in my path. It should be 'libsqlite3.dylib' and not
> 'libsqlie3.dylib'. Sorry.
>
> On Fri, Nov 28, 2014 at 1:41 AM, bill lam wrote:
>
>> Thanks Vijay, ddsqlite updated to check for /opt/local/lib/libsqlie3
http://www.mathmarauder.com/archives/178
That function F, written in tacit J, might look like this:
F=: ([: +/@:- [: */"1@:- [ ^/ (^"1 #:@i.@^~&2@#)@]) q:@]
Or, in explicit J, might look like this:
F=:dyad define
t=. q:y
+/-*/"1-x^/t^"1#:i.2^#t
)
As the explicit form is more compact tha
OOPS! There was a typo in my path. It should be 'libsqlite3.dylib' and not
'libsqlie3.dylib'. Sorry.
On Fri, Nov 28, 2014 at 1:41 AM, bill lam wrote:
> Thanks Vijay, ddsqlite updated to check for /opt/local/lib/libsqlie3.dylib
>
> Please report if this patch will cause any regression.
>
>
> Чт,
16 matches
Mail list logo