Assignment 2

Ouch..

It took me 4 hours today to get a working procedure for Question 2(a).
Yikes! Still a long way to go on this assignment.

 
  ,= ,-_-. =.
 ((_/)o o(\_))
  `-'(. .)`-'
      \_/

http://ilanpillemer.com
Entia non sunt multiplicanda praeter necessitatem

Hmm.. Well 2(b) was pretty similar to 2(a) so got that done relatively quickly.

Now with my solution to 2(c) if some father has only male children, but he has four male children he then appears four times as an option. And then if another father has two male children then he appears 2 times. However if a father has any non male children he does not appear at all.

Now I am not sure if that is the expected answer.. or if a male-gene father has four boys; he is only meant to be offered once as a solution.

Hmmm... I think my answer is sufficient... and I am certainly satisifed with it.
I had to work my way around the universal application of "not" for an uninstantiated variable to get my answer. So I feel like that was probably the main theme of the question.

Any thoughts? Though I think my answer is perfectly correct. And I will proceed onto question 3 tomorrow night.

By the way I am using GNU prolog which is slightly different to AMZI. For example I learnt that predicates are defined as dynamic by default in AMZI when they are consulted from a .pro file; but they are static by default in GNU prolog. That was a head scratched for an hour.

 
  ,= ,-_-. =.
 ((_/)o o(\_))
  `-'(. .)`-'
      \_/

http://ilanpillemer.com
Entia non sunt multiplicanda praeter necessitatem

Hi Ilan

Not sure what the question is but if you are only supposed to list a father if he has only male children then his name should only appear once and not 4 times (if he has 4 sons).


is this the bagof/setof/findall type question?

Well... I implemented my own version of findall for 2(a) and 2(b) (with no googling for hints). That's why it took so long to solve.

It looked like the question was more interested in =.. call assertz etc

It was past midnight when I got to the point of listing the father four times instead of once. I want to implement my own code to do this and have not used the bag built-in procedures; I will use findall,bagof,setof once I know how to write implementations myself. (and this way I get to grok =..,call,assert,dynamic vs static.) The feeling when it works and you really understand completely why is a very beautiful moment worth the mental anguish.

But after your comment I will finish the job tonight. I guess I will build a stack of tested fathers so that I can check it only is collected once. And clear the stack afterwards.

In question 3 I will use setof/bagof/findall.

 
  ,= ,-_-. =.
 ((_/)o o(\_))
  `-'(. .)`-'
      \_/

http://ilanpillemer.com
Entia non sunt multiplicanda praeter necessitatem

ok.. my procedure now only returns each father with a male gene once; no matter how many children he had. I used a stack.

 
  ,= ,-_-. =.
 ((_/)o o(\_))
  `-'(. .)`-'
      \_/

http://ilanpillemer.com
Entia non sunt multiplicanda praeter necessitatem

And finished question 3 this evening...
my setof version of add_element was a two line procedure. hmmm.

 
  ,= ,-_-. =.
 ((_/)o o(\_))
  `-'(. .)`-'
      \_/

http://ilanpillemer.com
Entia non sunt multiplicanda praeter necessitatem

yeah, the setof /bagof etc stuff is sually extremely short compared to the other ways of doing things.

well finished Question 4 (implementing mergesort).. one more question to go..

But I have two other assignments (452,407) due July 14. One is going to be late.. not sure which though. Probably 407.

 
  ,= ,-_-. =.
 ((_/)o o(\_))
  `-'(. .)`-'
      \_/

http://ilanpillemer.com
Entia non sunt multiplicanda praeter necessitatem

and finished question 5.. and then made sample runs for everything... turned into a pdf...

and submitted...

Yay!

Now for COS452 Assignment 2 (after a few days break.. I will start Sunday on this)

 
  ,= ,-_-. =.
 ((_/)o o(\_))
  `-'(. .)`-'
      \_/

http://ilanpillemer.com
Entia non sunt multiplicanda praeter necessitatem

cos407 - dr britz isn't at unisa so you migth get away with submitting her assignmet late