I need to print the number of coprime pairs (a,b), 0 < a <= b <= n and for n = 10^8 the program should run in less than 10 seconds. I have used this method : http://mathworld.wolfram.com/CarefreeCouple.html
But the program isn't as fast as I expected.
I have heard about an effient way of solving this problem by using something called 'Farey Sequence' but the code was written in PHP and I can only understand C.
So which method can help me solve the problem? thanks for the time.
Your stated interest is in co-prime pairs (a, b). The carefree couple adds an additional restriction that a is square-free. Therefore it is not the same problem, though some of the math is similar. As I understand your problem it is equivalent to summing the Euler totient function from 1 to n, the so-called Totient Summatory Function.
I do not know of any tricks that give one a simple closed form solution to come up with the answer. However, I think modifying a straightforward Sieve of Eratosthenes (SoT) should get you an answer in much less than 10 seconds in most programming languages.
Normally running the SoT simply yields a list of the primes <= n. Our goal will change, however, to computing the complete prime-power factorization of each integer between 1 and n inclusive. To do that we must store more than a single bit of information for each sieve entry, we must store a list. As we sieve a prime p through the array, we add (p, 1) to the list already stored at that entry. Then we sieve by p2 and change the (p,1) entries in each location we hit to (p,2), and so one for each power of p <= n and every p <= n. When it finishes, you
can compute the Totient function quickly for every value 0 <= x <= n and sum them up.
EDIT:
I see that there is already a much better set of answers to the question on math.stackexchange.com here. I'll leave this answer up for awhile until the disposition of the question is settled.
Related
Random question.
I am attempting to create a program which would generate a pseudo-random distribution. I am trying to find the right pseudo-random algorithm for my needs. These are my concerns:
1) I need one input to generate the same output every time it is used.
2) It needs to be random enough that a person who looks at the output from input 1 sees no connection between that and the output from input 2 (etc.), but there is no need for it to be cryptographically secure or truly random.
3)Its output should be a number between 0 and (29^3200)-1, with every possible integer in that range a possible and equally (or close to it) likely output.
4) I would like to be able to guarantee that every possible permutation of sequences of 410 outputs is also a potential output of consecutive inputs. In other words, all the possible groupings of 410 integers between 0 and (29^3200)-1 should be potential outputs of sequential inputs.
5) I would like the function to be invertible, so that I could take an integer, or a series of integers, and say which input or series of inputs would produce that result.
The method I have developed so far is to run the input through a simple halson sequence:
boost::multiprecision::mpz_int denominator = 1;
boost::multiprecision::mpz_int numerator = 0;
while (input>0) {
denominator *=3;
numerator = numerator * 3 + (input%3);
input = input/3;
}
and multiply the result by 29^3200. It meets requirements 1-3, but not 4. And it is invertible only for single integers, not for series (since not all sequences can be produced by it). I am working in C++, using boost multiprecision.
Any advice someone can give me concerning a way to generate a random distribution meeting these requirements, or just a class of algorithms worth researching towards this end, would be greatly appreciated. Thank you in advance for considering my question.
----UPDATE----
Since multiple commenters have focused on the size of the numbers in question, I just wanted to make clear that I recognize the practical problems that working with such sets poses but in asking this question I'm interested only in the theoretical or conceptual approach to the problem - for example, imagine working with a much smaller set of integers like 0 to 99, and the permutations of sets of 10 of output sequences. How would you design an algorithm to meet these five conditions - 1)input is deterministic, 2)appears random (at least to the human eye), 3)every integer in the range is a possible output, 4)not only all values, but also all permutations of value sequences are possible outputs, 5)function is invertible.
---second update---
with many thanks to #Severin Pappadeux I was able to invert an lcg. I thought I'd add a little bit about what I did to hopefully make it easier for anyone seeing this in the future. First of all, these are excellent sources on inverting modular functions:
https://www.khanacademy.org/computing/computer-science/cryptography/modarithmetic/a/modular-inverses
https://www.khanacademy.org/computer-programming/discrete-reciprocal-mod-m/6253215254052864
If you take the equation next=ax+c%m, using the following code with your values for a and m will print out the euclidean equations you need to find ainverse, as well as the value of ainverse:
int qarray[12];
qarray[0]=0;
qarray[1]=1;
int i =2;
int reset = m;
while (m % a >0) {
int remainder=m%a;
int quotient=m/a;
std::cout << m << " = " << quotient << "*" << a << " + " << remainder << "\n";
qarray[i] =qarray[i-2]-(qarray[i-1]*quotient);
m=a;
a=remainder;
i++;
}
if (qarray[i-1]<0) {qarray[i-1]+=reset;}
std::cout << qarray[i-1] << "\n";
The other thing it took me a while to figure out is that if you get a negative result, you should add m to it. You should add a similar term to your new equation:
prev = (ainverse(next-c))%m;
if (prev<0) {prev+=m;}
I hope that helps anyone who ventures down this road in the future.
Ok, I'm not sure if there is a general answer, so I would concentrate on random number generator having, say, 64bit internal state/seed, producing 64bit output and having 2^64-1 period. In particular, I would look at linear congruential generator (aka LCG) in the form of
next = (a * prev + c) mod m
where a and m are primes to each other
So:
1) Check
2) Check
3) Check (well, for 64bit space of course)
4) Check (again, except 0 I believe, but each and every permutation of 64bits is output of LCG starting with some seed)
5) Check. LCG is known to be reversible, i.e. one could get
prev = (next - c) * a_inv mod m
where a_inv could be computed from a, m using Euclid's algorithm
Well, if it looks ok to you, you could try to implement LCG in your 15546bits space
UPDATE
And quick search shows reversible LCG discussion/code here
Reversible pseudo-random sequence generator
In your update, "appears random (to the human eye)" is the phrasing you use. The definition of "appears random" is not a well agreed upon topic. There are varying degrees of tests for "randomness."
However, if you're just looking to make it appear random to the human eye, you can just use ring multiplication.
Start with the idea of generating N! values between 0 and M (N>=410, M>=29^3200)
Group this together into one big number. we're going to generate a single number ranging from 0 to *M^N!. If we can show that the pseudorandom number generator generates every value from 0 to M^N!, we guarantee your permutation rule.
Now we need to make it "appear random." To the human eye, Linear Congruent Generators are enough. Pick a LCG with a period greater than or equal to 410!*M^N satisfying the rules to ensure a complete period. Easiest way to ensure fairness is to pick a LCG in the form x' = (ax+c) mod M^N!
That'll do the trick. Now, the hard part is proving that what you did was worth your time. Consider that the period of just a 29^3200 long sequence is outside the realm of physical reality. You'll never actually use it all. Ever. Consider that a superconductor made of Josephine junctions (10^-12kg processing 10^11bits/s) weighing the mass of the entire universe 3*10^52kg) can process roughly 10^75bits/s. A number that can count to 29^3200 is roughly 15545 bits long, so that supercomputer can process roughly 6.5x10^71 numbers/s. This means it will take roughly 10^4600s to merely count that high, or somewhere around 10^4592 years. Somewhere around 10^12 years from now, the stars are expected to wink out, permanently, so it could be a while.
There are M**N sequences of N numbers between 0 and M-1.
You can imagine writing all of them one after the other in a (pseudorandom) sequence and placing your read pointer randomly in the resulting loop of N*(M**N) numbers between 0 and M-1...
def output(input):
total_length = N*(M**N)
index = input % total_length
permutation_index = shuffle(index / N, M**N)
element = input % N
return (permutation_index / (N**element)) % M
Of course for every permutation of N elements between 0 and M-1 there is a sequence of N consecutive inputs that produces it (just un-shuffle the permutation index). I'd also say (just using symmetry reasoning) that given any starting input the output of next N elements is equally probable (each number and each sequence of N numbers is equally represented in the total period).
I must implement a problem that computes combinations and multisets of m elements from a set of n elements.
The formulas for them are the following ones:
The problem is that with factorial it is easy to overflow, so basically what solutions can be for this problem?
Since it is a subproblem of a problem in TopCoder, I have following constraints:
1) A program must be written in C++.
2) I cannot load external libraries.
You don't really need to calculate n! directly, which may easily get overflowed. Since
C(n,m) = C(n-1, m-1) + C(n-1,m)
C(n,0) = C(n,n) =1
You can build a table with size (n+1,m+1) and use dynamic programming to build up the table in bottom up manner.
Algorithm pseudocode may like the following:
for i ← 0 to n do // fill out the table row wise
for j = 0 to min(i, m) do
if j==0 or j==i then C[i, j] ← 1
else C[i, j] ← C[i-1, j-1] + C[i-1, j]
return C[n, m]
If you declare c(n,m) to be long double and n is not that big, this way should work. Otherwise, you need to define your own BigInteger class to avoid overflow and define how the + operator works for BigIntegers, which are usually represented as array of chars or string.
Factorials are quite big numbers (they don't fit into a 64 bits word). So you need to use bignums (arbitrary precision arithmetic) to compute them in full. Consider using GMPlib for that purpose (or code in a language and implementation, e.g. Common Lisp with SBCL, which gives them natively)
See also this and that answers to a question very similar to yours.
Instead of using a recursive approach for calculating factorials which might lead to stack overflow, use iterative approach! This could save you overflow even for larger numbers.
I am trying to solve the SPOJ problem PGCD, which asks how many primes appear in a table of greatest common divisors.
The first idea that came to my mind is to generate the primes first by sieving.
Then, for each prime p, see how many pairs (a, b), where a and b are less than the given bounds, satisfy GCD(a,b)=p.
For example, how many pairs less than (20, 20) satisfy GCD(a,b)=7?
Of course, as mentioned, a and b are bounded.
So is it possible to reverse GCD? Or is this solution completely invalid?
Obviously the GCD function is not reversible/invertible because, for example,
GCD(10,15) == 5
GCD(5, 15) == 5
So if you are given 5 and try to guess the inputs, it is impossible.
I might be missing something here because I don't understand what you're saying about the bounding, but I think that's your responsibility to explain the problem better. Exactly what information do you have and what information are you trying to calculate? Example inputs and ouputs would be really useful. Also, proofreading and spell-checking.
If we have an array of all the numbers up to N (N < 10), what is the best way to find all the numbers that are missing.
Example:
N = 5
1 5 3 2 3
Output: 1 5 4 2 3
In the ex, the number 4 was the missing one and there were 2 3s, so we replaced the first one with 4 and now the array is complete - all the numbers up to 5 are there.
Is there any simple algorithm that can do this ?
Since N is really small, you can use F[i] = k if number i appears k times.
int F[10]; // make sure to initialize it to 0
for ( int i = 0; i < N; ++i )
++F[ numbers[i] ];
Now, to replace the duplicates, traverse your number array and if the current number appears more than once, decrement its count and replace it with a number that appears 0 times and increment that number's count. You can keep this O(N) if you keep a list of numbers that don't appear at all. I'll let you figure out what exactly needs to be done, as this sounds like homework.
Assume all numbers within the range 1 ≤ x ≤ N.
Keep 2 arrays of size N. output, used (as an associative array). Initialize them all to 0.
Scan from the right, fill in values to output unless it is used.
Check for unused values, and put them into the empty (zero) slots of output in order.
O(N) time complexity, O(N) space complexity.
You can use a set data structure - one for all the numbers up to N, one for the numbers you actually saw, and use a set difference.
One way to do this would be to look at each element of the array in sequence, and see whether that element has been seen before in elements that you've already checked. If so, then change that number to one you haven't seen before, and proceed.
Allow me to introduce you to my friend Schlemiel the Painter. Discovery of a more efficient method is left as a challenge for the reader.
This kind of looks like homework, please let us know if it isn't. I'll give you a small hint, and then I'll improve my answer if you confirm this isn't homework.
My tip for now is this: If you were to do this by hand, how would you do it? Would you write out an extra list of numbers of some time, would you read through the list (how many times?)? etc.
For simple problems, sometimes modelling your algorithm after an intuitive by-hand approach can work well.
Here's a link I read just today that may be helpful.
http://research.swtch.com/2008/03/using-uninitialized-memory-for-fun-and.html
I understand this is a classic programming problem and therefore I want to be clear I'm not looking for code as a solution, but would appreciate a push in the right direction. I'm learning C++ and as part of the learning process I'm attempting some programming problems. I'm attempting to write a program which deals with numbers up to factorial of 1billion. Obviously these are going to be enormous numbers and way too big to be dealing with using normal arithmetic operations. Any indication as to what direction I should go in trying to solve this type of problem would be appreciated.
I'd rather try to solve this without using additional libraries if possible
Thanks
PS - the problem is here http://www.codechef.com/problems/FCTRL
Here's the method I used to solve the problem, this was achieved by reading the comments below:
Solution -- The number 5 is a prime factor of any number ending in zero. Therefore, dividing the factorial number by 5, recursively, and adding the quotients, you get the number of trailing zeros in the factorial result
E.G. - Number of trailing zeros in 126! = 31
126/5 = 25 remainder 1
25/5 = 5 remainder 0
5/5 = 1 remainder 0
25 + 5 + 1 = 31
This works for any value, just keep dividing until the quotient is less
than 5
Skimmed this question, not sure if I really got it right but here's a deductive guess:
First question - how do you get a zero on the end of the number? By multiplying by 10.
How do you multiply by 10? either by multiplying by either a 10 or by 2 x 5...
So, for X! how many 10s and 2x5s do you have...?
(luckily 2 & 5 are prime numbers)
edit: Here's another hint - I don't think you need to do any multiplication. Let me know if you need another hint.
Hint: you may not need to calculate N! in order to find the number of zeros at the end of N!
To solve this question, as Chris Johnson said you have to look at number of 0's.
The factors of 10 will be 1,2,5,10 itself. So, you can go through each of the numbers of N! and write them in terms of 2^x * 5^y * 10^z. Discard other factors of the numbers.
Now the answer will be greaterof(x,y)+z.
One interesting thing I learn from this question is, its always better to store factorial of a number in terms of prime factors for easy comparisons.
To actually x^y, there is an easy method used in RSA algorithm, which don't remember. I will try to update the post if I find one.
This isn't a good answer to your question as you've modified it a bit from what I originally read. But I will leave it here anyway to demonstrate the impracticality of actually trying to do the calculations by main brute force.
One billion factorial is going to be out of reach of any bignum library. Such numbers will require more space to represent than almost anybody has in RAM. You are going to have to start paging the numbers in from storage as you work on them. There are ways to do this. The guy who recently calculated π out to 2700 billion places used such a library
Do not use the naive method. If you need to calculate the factorial, use a fast algorithm: http://www.luschny.de/math/factorial/FastFactorialFunctions.htm
I think that you should come up with a way to solve the problem in pseudo code before you begin to think about C++ or any other language for that matter. The nature of the question as some have pointed out is more of an algorithm problem than a C++ problem. Those who suggest searching for some obscure library are pointing you in the direction of a slippery slope, because learning to program is learning how to think, right? Find a good algorithm analysis text and it will serve you well. In our department we teach from the CLRS text.
You need a "big number" package - either one you use or one you write yourself.
I'd recommend doing some research into "large number algorithms". You'll want to implement the C++ equivalent of Java's BigDecimal.
Another way to look at it is using the gamma function. You don't need to multiply all those values to get the right answer.
To start you off, you should store the number in some sort of array like a std::vector (a digit for each position in the array) and you need to find a certain algorithm that will calculate a factorial (maybe in some sort of specialized class). ;)
//SIMPLE FUNCTION TO COMPUTE THE FACTORIAL OF A NUMBER
//THIS ONLY WORKS UPTO N = 65
//CAN YOU SUGGEST HOW WE CAN IMPROVE IT TO COMPUTE FACTORIAL OF 400 PLEASE?
#include <iostream>
#include <cmath>
using namespace std;
int factorial(int x); //function to compute factorial described below
int main()
{
int N; //= 150; //you can also get this as user input using cin.
cout<<"Enter intenger\n";
cin>>N;
factorial(N);
return 0;
}//end of main
int factorial(int x) //function to compute the factorial
{
int i, n;
long long unsigned results = 1;
for (i = 1; i<=x; i++)
{
results = results * i;
}
cout<<"Factorial of "<<x<<" is "<<results<<endl;
return results;
}