|
-
December 1st, 2003, 10:41 PM
#16
Re: finding relation ship or algo b/w 2 numbers
Originally posted by solo
its all logic, its all technique .. and its all artifical intelligence ...
And it's also a very, very hard problem algorithmically. Unfortunately, I don't think you're going to find anything close to a decent answer, particularly not one done with VB. Frankly, anyone who could solve this problem should be doing better things with their time than hanging around some message board.
Ain't no party like a VB party, cause a VB party don't stop.
-
December 13th, 2003, 01:38 PM
#17
to pinky
first of really really sorry that i could not reply early ,,, because was busy with exams and other stuffs ...
and so many thanks for all who replied to this query ..
now replying to your reply ...
ya u r right.. there are number of infinite relation b/w these pairs... but i have to get the right relation bcoz i know the sets with the same relations among them .. so this exists ..
u said =>One of the easiest way to try and guess at the algorithm is to use a neural-net.
whats about the nural net ... how i can use that in this scenario ... and how i can find the relation b/w these numbers using this system ..
waiting for answer ...
thanks
take care
-
December 13th, 2003, 01:42 PM
#18
reply to avdav
first of really really sorry that i could not reply early ,,, because was busy with exams and other stuffs ...
and so many thanks for all who replied to this query ..
now replying to your reply ...
to AdDav
ya u got the right thing .. i mean this too which u said as "do you want to determine whether an unknown y[i] has already been calculated for the given x[i]? "
now i did not get from ur this point => "Or perhaps you mean something like inter/extra polation, but it won't guarantee that you'll get the right value for the given table"
and this point => "Or perhaps you can use 'difference calculus' method by guessing that given function is a polynomial of n-th power"
how i can guess a right polynomial for that relation or algo .. i dont know how these number were generated.. and i cant read anothers mind that what he uses in this algorithm but i just know these sets ...
its really a bug problem for me and i have to get the result and solve this query ..
waiting for your help ...
TO KHEUN
u said =>"analysis but I think cubic spline for interpolation is general enough to work out an equation that fix nicely with a given set of points"
so in your view what will be the right equation for these sets or which equation can nearly fit these points ..
actually i m a programer and i have not been with extra mathematical and statistical background .. so i m feeling some difficulties in evaluating these ..
hope u work it out ..
waiting for your help ...
-
December 13th, 2003, 01:43 PM
#19
to oreabans
first of really really sorry that i could not reply early ,,, because was busy with exams and other stuffs ...
and so many thanks for all who replied to this query ..
now replying to your reply ...
ya u r right .. its a big deal ...
ya there many be a number of relation b/w two number .. but if a relatoin b/w these numbers is use to extract 2nd pair then 3rd pair then 4th and so on and v know all these pair .. then using all these pair v can surely find the right relation ... and this will be the relation which can be use to extract more pairs giing first number to that relation ..
and i m looking for that technique ... and hope u will find it for me or try to evaluate this query ...
waiting for your answer
to hometown
thanks dear for your answer
oh is this the little information ...
well actually they are alphanumeric or hexadecimal pairs with the same relation or technique uses to generate them ..
as i give number 1234 ... computer returns me 8765 after performing some calculation or formula .. then using the same calculation or formula or equation i enter 34566 and computer generate 23485 for me
and here i m going to find out that relation or formula or equation with the help of these number which already been calculated by that formula ...
hope it works ...
waiting for your answer
thanks
take care
thanks
take care
-
December 13th, 2003, 01:45 PM
#20
to SolarFlare
first of really really sorry that i could not reply early ,,, because was busy with exams and other stuffs ...
and so many thanks for all who replied to this query ..
now replying to your reply ...
well u said there is no definate algo for any set of pairs ... i guess there is bcoz v have different pairs which follows the same algo
as i give number 1234 ... computer returns me 8765 after performing some calculation or formula .. then using the same calculation or formula or equation i enter 34566 and computer generate 23485 for me
and here i m going to find out that relation or formula or equation with the help of these number which already been calculated by that formula ...
hope it works .. and waiting for your queries and answer
regards
thanks
take care
-
December 13th, 2003, 07:15 PM
#21
Can you try and restate what you want to do.
What do you know at the beginning. What do you want to
know at the end.
Wakeup in the morning and kick the day in the teeth!! Or something like that.
"i don't want to write leak free code or most efficient code, like others traditional (so called expert) coders do."
-
December 14th, 2003, 03:56 AM
#22
Look.. plain and simple: It can not be done. The best you
can do is find a formula which will agree with the finite function
you have so far. And even this is in general not possible.
Wakeup in the morning and kick the day in the teeth!! Or something like that.
"i don't want to write leak free code or most efficient code, like others traditional (so called expert) coders do."
-
December 14th, 2003, 05:18 AM
#23
Originally posted by solo
____snip____
What exactly are you asking ?
I actually dont understand a thing about it...
Can you tell me from the begining of how it s started ? Because your problem sounds interesting !!
-
December 14th, 2003, 11:13 AM
#24
Originally posted by souldog
Look.. plain and simple: It can not be done. The best you
can do is find a formula which will agree with the finite function
you have so far. And even this is in general not possible.
It is always possible to find a formula which will agree, but from the looks of things, it's not the one anyone would want. There is always a polynomial in P(n-1), where there are n terms, which will pass through all the points.
SolarFlare
Those who cling to life die and those who defy death live. -Sun Tzu
cout << endl;
return 0;
}
-
December 14th, 2003, 11:25 AM
#25
assuming the points represent a finite function, but
yes that is clearly the case here so you are right
Last edited by souldog; December 14th, 2003 at 11:28 AM.
Wakeup in the morning and kick the day in the teeth!! Or something like that.
"i don't want to write leak free code or most efficient code, like others traditional (so called expert) coders do."
-
December 14th, 2003, 01:23 PM
#26
in the beginning i know some pairs
like 1234 generated from 9876
2nd pair as 84746 generated from 73734
both numbres are generated using the same method, function, algoritham , formula, or quation whatever v say
i-e in the beginning v have 9876 and 73734
and in the end v have the generated numbers 1234 and 84746
i guess it helps ..
solo
-
December 14th, 2003, 01:30 PM
#27
hi hometown
actually i want to find the formula, equatoin , algo or some thing like this using which i can find the requied digits after providing some numbrs to it
i give 1234 to that equation or formula or whatever and it gives 9876 then using same eqation or formula or whatever i give 83847 it gives 83747 ..
so now i have these 2 set of numbrs .. and i want to find out that formula or equation or whatever
thas what i m asking ...
i m sure it is possible .. but how .. v have to find out .. its mind reading .......
solo...
-
December 14th, 2003, 01:34 PM
#28
You just don't get it.
Do you want A equation: Possible but probably useless
(I assume you want to generate other pairs that will agree with
the algorithm. )
or
Do you want THE equation (Not possible)
Wakeup in the morning and kick the day in the teeth!! Or something like that.
"i don't want to write leak free code or most efficient code, like others traditional (so called expert) coders do."
-
December 14th, 2003, 02:12 PM
#29
[Merged 6 cross-postings]
SolarFlare
Those who cling to life die and those who defy death live. -Sun Tzu
cout << endl;
return 0;
}
-
December 15th, 2003, 03:48 PM
#30
ya i want to get A equation which and need the another paris which follows that algoritham
Posting Permissions
- You may not post new threads
- You may not post replies
- You may not post attachments
- You may not edit your posts
-
Forum Rules
|
Click Here to Expand Forum to Full Width
|