Purr was my kinder gentler purrcent. You are right about percent. Multiply
by 100. Write assert 1=100{+/purr d as the final line.
I tend to use casual names so that these exercises will be understood to be
experiences for examining the development of code. They are not designes as
commercial grade code. I do appreciate being advised when I have been
inaccurate, so thanks for the reminder about percent.
Linda
----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Raul Miller
Sent: Wednesday, December 07, 2011 9:41 AM
To: Programming forum
Subject: Re: [Jprogramming] Many Turkey Rolls - No embargo
purr=: [: (%"(1) 1 _ <."1 +/) fdrm
That said, note that I think of the quantity in the second column as
"fraction of total" rather than "percent". For them to be percents, I
would want to multiply them by 100:
perc=: [: (%"(1) 1 _ <."(1) 0.01 * +/) fdrm
But I understand that "percent" is a more convenient phrase than "fraction
of total".
--
Raul
On Wed, Dec 7, 2011 at 9:26 AM, Linda Alvord <[email protected]>wrote:
> Roul, Your expression is a good one. To define a verb you could start
here:
>
> (~.,.#/.~)/:~+/1+?2 1e6$6
> 2 28046
> 3 55258
> 4 83532
> 5 111206
> 6 138742
> 7 166167
> 8 139145
> 9 111078
> 10 83270
> 11 55635
> 12 27921
>
> Now define a function. It could easily be monadic:
>
> fdrm=: 13 :'(~.,.#/.~)/:~+/1+?y'
> fdrm
> [: (~. ,. #/.~) [: /:~ [: +/ 1 + ?
>
> ]d=:fdrm 2 1e6$6
> 2 27744
> 3 55994
> 4 83221
> 5 111031
> 6 138908
> 7 166726
> 8 138794
> 9 111511
> 10 82566
> 11 55995
> 12 27510
>
> There is only one more design issue and we'll be finished with the turkey
> roll.
>
> Finally define a function to produce the percentage of each of the rolls
> occurred. Percent was traumatic for some students, so we'll give the
> function a warm fuzzy name:
>
> purr=:
> purr
> (your tacit definition should land here)
>
> Now see what happens to your data:
>
> purr d
> 2 0.027797
> 3 0.05549
> 4 0.082886
> 5 0.111182
> 6 0.138412
> 7 0.166141
> 8 0.140112
> 9 0.11171
> 10 0.083511
> 11 0.055302
> 12 0.027457
>
> assert 1=1{+/purr d
>
> Put in your function and data (purr d) and no response indicates the sum
of
> the percentages in the second column had a sum of 1 or 100%
>
> Now see if you can write purr.
>
> Linda
>
>
>
>
>
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of Raul Miller
> Sent: Monday, December 05, 2011 10:03 AM
> To: Programming forum
> Subject: Re: [Jprogramming] Many Turkey Rolls - No embargo
>
> I am having trouble understanding this one also.
>
> I think I understand 1:
> #/.~/:~+/?2 1e6$6
> 27773 55724 83930 111049 138508 166122 138982 111263 83609 55632 27408
>
> Or, perhaps:
>
> (~.,.#/.~)/:~+/1+?2 1e6$6
> 2 27704
> 3 55158
> 4 83565
> 5 111238
> 6 138938
> 7 167140
> 8 138898
> 9 110833
> 10 83313
> 11 55783
> 12 27430
>
> Item 2, however, seems mysterious. The platonic solids would have face
> counts of 4, 6, 8, 12, 20, but if we had 100 of each in a bucket and toss
> the bucket 200 times and total the numbers we will almost always have over
> 100 unique values in those resulting sums. This would result in a rather
> odd frequency distribution so I'm not at all sure that I understand the
> problem correctly.
>
> Furthermore, the additional text seemed to have little relation to these
> two items, so I am very confused.
>
> --
> Raul
>
>
>
> On Sun, Dec 4, 2011 at 6:12 PM, Linda Alvord <[email protected]
> >wrote:
>
> > Below you will find my version with some changes. However, I've pointed
> > out
> > some other versions that were superior to mine. When I tried to mix
them
> I
> > got all mixed up. To do these two exercises, you may use any of the
> > statements below. Modify their use to:
> >
> > 1. Toss a pair of dice a million times and produce a frequency
> > distribution of the sums. This a simulation and there will be no graphic
> > images of the results.
> >
> > 2. Image a large bucket with 100 dice of each of the five Platonic
> > Solids. Each die is numbered from 1 with consecutive counting numbers.
> > Simulate the result and summarize your results in a frequency
> distribution
> > of the dice are all tosses from the bucket 200 times.
> >
> >
> >
> > Here are the expression you may choose from along with J symbols. Make
> > your
> > expressions as simple as possible. (I think Ric, Kip and Henry should
> just
> > watch and maybe referee if needed.)
> >
> >
> >
> >
> >
> > d1=:' o '
> >
> > d2=:'o o'
> >
> > d3=:'o o o'
> >
> > d4=:'o o o o'
> >
> > d5=:'o o o o o'
> >
> > d6=:'o oo oo o'
> >
> > d=:6 9$d1,d2,d3,d4,d5,d6
> >
> > dice=:(<"2)3 3$"1 d
> >
> > dice
> >
> > s=: 13 :'c=:1+?2 10$y'
> >
> > toss=: 13 :'(<"2)3 3$"1(<:s y){d'
> >
> > c
> >
> > toss 6
> >
> > c
> >
> > toss=: 13 :'(<"2)3 3$"1(<:s y){d'
> >
> > t=: 13 :'+/"2 y'
> >
> > fd=: [: /:~ ~. ,. [: +/"1 =
> >
> > dice
> >
> > toss 6
> >
> > c
> >
> > t c
> >
> > fd t c
> >
> > toss 6
> >
> > c
> >
> > t c
> >
> > fd t c
> >
> > toss 4
> >
> > c
> >
> > t c
> >
> > fd t c
> >
> > assert 0 0 3 3 3 3 = 4!:0 ;:'dice c s t toss fd'
> >
> >
> >
> > NB. dice is a graphic image of the faces of a die
> >
> > NB. c captures the data from an array of random rolls of the dice
> >
> > NB. s is the shape and s y allows for different dice
> >
> > NB. t is a list of totals for each of the tosses
> >
> > NB. fd is a frequency distribution of data in a list
> >
> >
> >
> >
> >
> >
> > ----------------------------------------------------------------------
> > For information about J forums see http://www.jsoftware.com/forums.htm
> >
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
>
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
>
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
