561 returns true on Fermat test - java

I found this Fermat's algorithm for testing primes and I found out that it doesn't always return true for Carmichael numbers (such as 561). I tried to find the problem but I can't find anything wrong with the algorithm. What could be the problem?
import java.util.Scanner;
import java.util.Random;
import java.math.BigInteger;
public class FermatTest {
private final static Random rand = new Random();
private static BigInteger getRandomFermatBase(BigInteger n)
{
while (true)
{
final BigInteger a = new BigInteger (n.bitLength(), rand);
if (a.compareTo(BigInteger.ONE) == 1 && a.compareTo(n) < 0)
{
return a; // 1 <= a < n
}
}
}
public static String checkPrime(BigInteger n, int maxIterations)
{
if (n.equals(BigInteger.ONE))
return "is composite";
for (int i = 0; i < maxIterations; i++)
{
BigInteger a = getRandomFermatBase(n); //generate random a
a = a.modPow(n.subtract(BigInteger.ONE), n); //a^(p-1) mod p
if (!a.equals(BigInteger.ONE)) // not equals 1
return "is composite";
}
return "is probably prime";
}
public static void main(String[] args)
{
long start = System.nanoTime();
BigInteger n = new BigInteger("561");
System.out.println(n + " " + checkPrime(n , 20));
float time = System.nanoTime() - start;
System.out.println("Time: " + (long) time + " nanoseconds");
time = time / (1000000000);
System.out.println("Time: " + time + " seconds");
}
}

You identified the problem yourself. There are some numbers, the Carmichael numbers, that are improperly marked prime by the Fermat test for all possible bases. That's just a fact of number theory. You will need a different test to discriminate numbers as prime or composite.

Related

Fibonacci sequence - How to calculate the sum of the first 100 even-values Fibonacci numbers?

Fibonacci sequence is defined as a sequence of integers starting with 1 and 1, where each subsequent value is the sum of the preceding two I.e.
f(0) = 1
f(1) = 1
f(n) = f(n-1) + f(n-2) where n>=2
My goal is to calculate the sum of the first 100 even-values Fibonacci numbers.
So far I've found this code which works perfectly to calculate the sum of even numbers to 4million , however I'm unable to find edit the code so that it stops at the sum of the 100th value, rather than reaching 4million.
public class Improvement {
public static int Fibonacci(int j) {
/**
*
* Recursive took a long time so continued with iterative
*
* Complexity is n squared.. try to improve to just n
*
*/
int tmp;
int a = 2;
int b = 1;
int total = 0;
do {
if(isEven(a)) total +=a;
tmp = a + b;
b = a;
a = tmp;
} while (a < j);
return total;
}
private static boolean isEven(int a) {
return (a & 1) == 0;
}
public static void main(String[] args) {
// Notice there is no more loop here
System.out.println(Fibonacci(4_000_000));
}
}
Just to show the console from #mr1554 code answer, the first 100 even values are shown and then the sum of all is 4850741640 as can be seen below:
Any help is appreciated, thanks!
You need to use BigInteger because long easily overflows as Fibonacci's scales quite easily. BigInteger is also tricky to check whether is an odd or even number, but you can use BigInteger::testBit returning boolean as explained in this answer.
Here is some complete code:
BigInteger fibonacciSum(int count, boolean isOdd) {
int i = 0;
BigInteger sum = BigInteger.ZERO;
BigInteger current = BigInteger.ONE;
BigInteger next = BigInteger.ONE;
BigInteger temp;
while (i < count) {
temp = current;
current = current.add(next);
next = temp;
if ((current.testBit(0) && isOdd) || ((!current.testBit(0) && !isOdd))) {
sum = sum.add(current);
i++;
}
}
return sum;
}
Or you can have some fun with Stream API:
BigInteger fibonacciSum(int count, boolean isOdd) {
final BigInteger[] firstSecond = new BigInteger[] {BigInteger.ONE, BigInteger.ONE};
return Stream.iterate(
firstSecond,
num -> new BigInteger[] { num[1], num[0].add(num[1]) })
.filter(pair ->
(pair[1].testBit(0) && isOdd) ||
(!pair[1].testBit(0) && !isOdd))
.limit(count)
.map(pair -> pair[1])
.reduce(BigInteger.ZERO, BigInteger::add);
}
In any way, don't forget to test it out:
#Test
void test() {
assertThat(
fibonacciSum(100, false),
is(new BigInteger("290905784918002003245752779317049533129517076702883498623284700")));
}
You said.
My goal is to calculate the sum of the first 100 even-values Fibonacci numbers.
That number gets very large very quickly. You need to:
use BigInteger
use the mod function to determine if even
For this I could have started from (1,1) but it's only one term so ...
BigInteger m = BigInteger.ZERO;
BigInteger n = BigInteger.ONE;
BigInteger sumOfEven= BigInteger.ZERO;
int count = 0;
BigInteger t;
while( count < 100) {
t = n.add(m);
// check if even
if (t.mod(BigInteger.TWO).equals(BigInteger.ZERO)) {
sumOfEven = sumOfEven.add(t);
count++;
}
n = m;
m = t;
}
System.out.println(sumOfEven);
Prints
290905784918002003245752779317049533129517076702883498623284700
If, on the other hand, from your comment.
My aim is to calculate the sum of the first 100 even numbers
Then you can do that like so
sumFirstNeven = (((2N + 2)N)/2 = (N+1)N
so (101)100 = 10100 and the complexity is O(1)
as I figured, you want a program to sum 100 first even values of the Fibonacci series.
here is a sample code, when you run the program it will ask you to determine the number of the even values, you want 100 value e.g, type 100 in consul:
public static void main(String[] args) {
int firstNumber = 0;
int secondNumber = 2;
System.out.print("Enter the number of odd elements of the Fibonacci Series to sum : ");
Scanner scan = new Scanner(System.in);
int elementCount = scan.nextInt(); // number of even values you want
System.out.print(firstNumber + ", ");
System.out.print(secondNumber + ", ");
long sum = 2;
for (int i = 2; i < elementCount; i++) {
int nextNumber = firstNumber + secondNumber;
System.out.print(nextNumber + ", ");
sum += (nextNumber);
firstNumber = secondNumber;
secondNumber = nextNumber;
}
System.out.print("...");
System.out.println("\n" + "the sum of " + elementCount + " values of fibonacci series is: " + sum);
}

How to calculate Euler's number faster with Java multithreading

So I have a task to calculate Euler's number using multiple threads, using this formula: sum( ((3k)^2 + 1) / ((3k)!) ), for k = 0...infinity.
import java.math.BigDecimal;
import java.math.BigInteger;
import java.io.FileWriter;
import java.io.IOException;
import java.math.RoundingMode;
class ECalculator {
private BigDecimal sum;
private BigDecimal[] series;
private int length;
public ECalculator(int threadCount) {
this.length = threadCount;
this.sum = new BigDecimal(0);
this.series = new BigDecimal[threadCount];
for (int i = 0; i < this.length; i++) {
this.series[i] = BigDecimal.ZERO;
}
}
public synchronized void addToSum(BigDecimal element) {
this.sum = this.sum.add(element);
}
public void addToSeries(int id, BigDecimal element) {
if (id - 1 < length) {
this.series[id - 1] = this.series[id - 1].add(element);
}
}
public synchronized BigDecimal getSum() {
return this.sum;
}
public BigDecimal getSeriesSum() {
BigDecimal result = BigDecimal.ZERO;
for (int i = 0; i < this.length; i++) {
result = result.add(this.series[i]);
}
return result;
}
}
class ERunnable implements Runnable {
private final int id;
private final int threadCount;
private final int threadRemainder;
private final int elements;
private final boolean quietFlag;
private ECalculator eCalc;
public ERunnable(int threadCount, int threadRemainder, int id, int elements, boolean quietFlag, ECalculator eCalc) {
this.id = id;
this.threadCount = threadCount;
this.threadRemainder = threadRemainder;
this.elements = elements;
this.quietFlag = quietFlag;
this.eCalc = eCalc;
}
#Override
public void run() {
if (!quietFlag) {
System.out.println(String.format("Thread-%d started.", this.id));
}
long start = System.currentTimeMillis();
int k = this.threadRemainder;
int iteration = 0;
BigInteger currentFactorial = BigInteger.valueOf(intFactorial(3 * k));
while (iteration < this.elements) {
if (iteration != 0) {
for (int i = 3 * (k - threadCount) + 1; i <= 3 * k; i++) {
currentFactorial = currentFactorial.multiply(BigInteger.valueOf(i));
}
}
this.eCalc.addToSeries(this.id, new BigDecimal(Math.pow(3 * k, 2) + 1).divide(new BigDecimal(currentFactorial), 100, RoundingMode.HALF_UP));
iteration += 1;
k += this.threadCount;
}
long stop = System.currentTimeMillis();
if (!quietFlag) {
System.out.println(String.format("Thread-%d stopped.", this.id));
System.out.println(String.format("Thread %d execution time: %d milliseconds", this.id, stop - start));
}
}
public int intFactorial(int n) {
int result = 1;
for (int i = 1; i <= n; i++) {
result *= i;
}
return result;
}
}
public class TaskRunner {
public static final String DEFAULT_FILE_NAME = "result.txt";
public static void main(String[] args) throws InterruptedException {
int threadCount = 2;
int precision = 10000;
int elementsPerTask = precision / threadCount;
int remainingElements = precision % threadCount;
boolean quietFlag = false;
calculate(threadCount, elementsPerTask, remainingElements, quietFlag, DEFAULT_FILE_NAME);
}
public static void writeResult(String filename, String result) {
try {
FileWriter writer = new FileWriter(filename);
writer.write(result);
writer.close();
} catch (IOException e) {
System.out.println("An error occurred.");
e.printStackTrace();
}
}
public static void calculate(int threadCount, int elementsPerTask, int remainingElements, boolean quietFlag, String outputFile) throws InterruptedException {
long start = System.currentTimeMillis();
Thread[] threads = new Thread[threadCount];
ECalculator eCalc = new ECalculator(threadCount);
for (int i = 0; i < threadCount; i++) {
if (i == 0) {
threads[i] = new Thread(new ERunnable(threadCount, i, i + 1, elementsPerTask + remainingElements, quietFlag, eCalc));
} else {
threads[i] = new Thread(new ERunnable(threadCount, i, i + 1, elementsPerTask, quietFlag, eCalc));
}
threads[i].start();
}
for (int i = 0; i < threadCount; i++) {
threads[i].join();
}
String result = eCalc.getSeriesSum().toString();
if (!quietFlag) {
System.out.println("E = " + result);
}
writeResult(outputFile, result);
long stop = System.currentTimeMillis();
System.out.println("Calculated in: " + (stop - start) + " milliseconds" );
}
}
I stripped out the prints, etc. in the code that have no effect. My problem is that the more threads I use the slower it gets. Currently the fastest run I have is for 1 thread. I am sure the factorial calculation is causing some issues. I tried using a thread pool but still got the same times.
How can I make it so that running it with more threads, up until some point, will speed up the calculation process?
How would one go about calculating this big factorials?
The precision parameter that is passed is the amount of elements in the sum that are used. Can I set the BigDecimal scale to be somehow dependent on that precision so I don't hard code it?
EDIT
I updated the code block to be in 1 file only and runnable without external libs.
EDIT 2
I found out that the factorial code messes up with the time. If I let the threads ramp up to some high precision without calculating factorials the time goes down with increasing threads. Yet I cannot implement the factorial calculating in any way while keeping the time decreasing.
EDIT 3
Adjusting code to address answers.
private static BigDecimal partialCalculator(int start, int threadCount, int id) {
BigDecimal nBD = BigDecimal.valueOf(start);
BigDecimal result = nBD.multiply(nBD).multiply(BigDecimal.valueOf(9)).add(BigDecimal.valueOf(1));
for (int i = start; i > 0; i -= threadCount) {
BigDecimal iBD = BigDecimal.valueOf(i);
BigDecimal iBD1 = BigDecimal.valueOf(i - 1);
BigDecimal iBD3 = BigDecimal.valueOf(3).multiply(iBD);
BigDecimal prevNumerator = iBD1.multiply(iBD1).multiply(BigDecimal.valueOf(9)).add(BigDecimal.valueOf(1));
// 3 * i * (3 * i - 1) * (3 * i - 2);
BigDecimal divisor = iBD3.multiply(iBD3.subtract(BigDecimal.valueOf(1))).multiply(iBD3.subtract(BigDecimal.valueOf(2)));
result = result.divide(divisor, 10000, RoundingMode.HALF_EVEN)
.add(prevNumerator);
}
return result;
}
public static void main(String[] args) {
int threadCount = 3;
int precision = 6;
ExecutorService executorService = Executors.newFixedThreadPool(threadCount);
ArrayList<Future<BigDecimal> > futures = new ArrayList<Future<BigDecimal> >();
for (int i = 0; i < threadCount; i++) {
int start = precision - i;
System.out.println(start);
final int id = i + 1;
futures.add(executorService.submit(() -> partialCalculator(start, threadCount, id)));
}
BigDecimal result = BigDecimal.ZERO;
try {
for (int i = 0; i < threadCount; i++) {
result = result.add(futures.get(i).get());
}
} catch (Exception e) {
e.printStackTrace();
}
executorService.shutdown();
System.out.println(result);
}
Seems to be working properly for 1 thread but messes up the calculation for multiple.
After a review of the updated code, I've made the following observations:
First of all, the program runs for a fraction of a second. That means that this is a micro benchmark. Several key features in Java make micro benchmarks difficult to implement reliably. See How do I write a correct micro-benchmark in Java? For example, if the program doesn't run enough repetitions, the "just in time" compiler doesn't have time to kick in to compile it to native code, and you end up benchmarking the intepreter. It seems possible that in your case the JIT compiler takes longer to kick in when there are multiple threads,
As an example, to make your program do more work, I changed the BigDecimal precision from 100 to 10,000 and added a loop around the main method. The execution times were measured as follows:
1 thread:
Calculated in: 2803 milliseconds
Calculated in: 1116 milliseconds
Calculated in: 1040 milliseconds
Calculated in: 1066 milliseconds
Calculated in: 1036 milliseconds
2 threads:
Calculated in: 2354 milliseconds
Calculated in: 856 milliseconds
Calculated in: 624 milliseconds
Calculated in: 659 milliseconds
Calculated in: 664 milliseconds
4 threads:
Calculated in: 1961 milliseconds
Calculated in: 797 milliseconds
Calculated in: 623 milliseconds
Calculated in: 536 milliseconds
Calculated in: 497 milliseconds
The second observation is that there is a significant part of the workload that does not benefit from multiple threads: every thread is computing every factorial. This means the speed-up cannot be linear - as described by Amdahl's law.
So how can we get the result without computing factorials? One way is with Horner's method. As an example, consider the simpler series sum(1/k!) which also conveges to e but a little slower than yours.
Let's say you want to compute sum(1/k!) up to k = 100. With Horner's method you start from the end and extract common factors:
sum(1/k!, k=0..n) = 1/100! + 1/99! + 1/98! + ... + 1/1! + 1/0!
= ((... (((1/100 + 1)/99 + 1)/98 + ...)/2 + 1)/1 + 1
See how you start with 1, divide by 100 and add 1, divide by 99 and add 1, divide by 98 and add 1, and so on? That makes a very simple program:
private static BigDecimal serialHornerMethod() {
BigDecimal accumulator = BigDecimal.ONE;
for (int k = 10000; k > 0; k--) {
BigDecimal divisor = new BigDecimal(k);
accumulator = accumulator.divide(divisor, 10000, RoundingMode.HALF_EVEN)
.add(BigDecimal.ONE);
}
return accumulator;
}
Ok that's a serial method, how do you make it use parallel? Here's an example for two threads: First split the series into even and odd terms:
1/100! + 1/99! + 1/98! + 1/97! + ... + 1/1! + 1/0! =
(1/100! + 1/98! + ... + 1/0!) + (1/99! + 1/97! + ... + 1/1!)
Then apply Horner's method to both the even and odd terms:
1/100! + 1/98! + 1/96! + ... + 1/2! + 1/0! =
((((1/(100*99) + 1)/(98*97) + 1)/(96*95) + ...)/(2*1) + 1
and:
1/99! + 1/97! + 1/95! + ... + 1/3! + 1/1! =
((((1/(99*98) + 1)/(97*96) + 1)/(95*94) + ...)/(3*2) + 1
This is just as easy to implement as the serial method, and you get pretty close to linear speedup going from 1 to 2 threads:
private static BigDecimal partialHornerMethod(int start) {
BigDecimal accumulator = BigDecimal.ONE;
for (int i = start; i > 0; i -= 2) {
int f = i * (i + 1);
BigDecimal divisor = new BigDecimal(f);
accumulator = accumulator.divide(divisor, 10000, RoundingMode.HALF_EVEN)
.add(BigDecimal.ONE);
}
return accumulator;
}
// Usage:
ExecutorService executorService = Executors.newFixedThreadPool(2);
Future<BigDecimal> submit = executorService.submit(() -> partialHornerMethod(10000));
Future<BigDecimal> submit1 = executorService.submit(() -> partialHornerMethod(9999));
BigDecimal result = submit1.get().add(submit.get());
There is a lot of contention between the threads: they all compete to get a lock on the ECalculator object after every little bit of computation, because of this method:
public synchronized void addToSum(BigDecimal element) {
this.sum = this.sum.add(element);
}
In general having threads compete for frequent access to a common resource leads to bad performance, because you're asking for the operating system to intervene and tell the program which thread can continue. I haven't tested your code to confirm that this is the issue because it's not self-contained.
To fix this, have the threads accumulate their results separately, and merge results after the threads have finished. That is, create a sum variable in ERunnable, and then change the methods:
// ERunnable.run:
this.sum = this.sum.add(new BigDecimal(Math.pow(3 * k, 2) + 1).divide(new BigDecimal(factorial(3 * k)), 100, RoundingMode.HALF_UP));
// TaskRunner.calculate:
for (int i = 0; i < threadCount; i++) {
threads[i].join();
eCalc.addToSum(/* recover the sum computed by thread */);
}
By the way would be easier if you used the higher level java.util.concurrent API instead of creating thread objects yourself. You could wrap the computation in a Callable which can return a result.
Q2 How would one go about calculating this big factorials?
Usually you don't. Instead, you reformulate the problem so that it does not involve the direct computation of factorials. One technique is Horner's method.
Q3 The precision parameter that is passed is the amount of elements in the sum that are used. Can I set the BigDecimal scale to be somehow dependent on that precision so I don't hard code it?
Sure why not. You can work out the error bound from the number of elements (it's proportional to the last term in the series) and set the BigDecimal scale to that.

JAVA running too much time

This is my code,it has been one hour but it hasn't returned a value yet,is there anything wrong?
import java.math.BigInteger;
public class PROJECTV1 {
public static void main(String[] args) {
BigInteger bResult = bigFunctionExample_2();
System.out.println(" => result_got:" + bResult);
System.out.println(); //newline
}// end_main
public static BigInteger bigFunctionExample_2() {
BigInteger bSum = BigInteger.ZERO;
BigInteger bTmp;
String sSum;
// BigInteger bResult =0;
for (int i = 1; ; i++) {
bTmp = BigInteger.valueOf(i);
bTmp = bTmp.pow(2); // i^2
bSum = bSum.add(bTmp); // sum = i^2+ (i-1)^2 + ....
sSum = bSum.toString();
if (sSum.length() > 30) {
System.out.println("i=" + i + " bSum =" + bSum);
break;
}
}//end_for
return bSum; // result
}
// end_bigFunctionExample_2
}
For that loop to break, it must reach 10^30 ~= 2^100. Sum of the squares of the first n natural numbers is approximately equal to n^3. So your loop will break approximately when i becomes 10^10 ~= 2^33. I guess int i is 32-bit so you MIGHT be are overflowing that integer, I didn't do the exact math but it's very possible.
If you go for a 64-bit variable (long?), which has an upper limit of approximately 10^19, you might have a chance it will be OK.
Edit: Here's the exact math from WolframAlpha.

Does Java's Random Object create random numbers through equal opportunity?

I'm trying to write a short program that plays a dice game (normal 6 sided dice) for me. The first roll's number is added to the score. After the first roll, if I roll a 6 then the game stops and the score is recorded (without adding the 6). If a 6 is rolled on the first roll, that's fine and it's added like any other number 1 through 5. I'm trying to run a bunch of iterations of this game such that I have a long list of scores pre-bust (a bust being a rolled 6). I rearrange those scores to be in order from smallest to largest and then find the median of the list which is the score at which it is optimal to stop.
For some reason I keep getting 13 when I run the program but I know for a fact that the answer should be 15. Would using Random in Java would have some sort of effect on the median? I don't exactly know how Random generates the numbers and whether it creates them with equal opportunity. Also, is there anything that just pops out that shouldn't work?
import java.util.*;
public class DiceRoller {
private static Random r = new Random();
private static final int games = 10001;
private static int[] totalScores = new int[games];
private static int index = 0;
public static void main(String[] args) {
int score = 0; boolean firstRoll = true;
while (index < games) {
int roll = roll();
if (firstRoll) {
score += roll;
firstRoll = false;
} else {
if (roll == 6) {
totalScores[index] = score;
index++;
score = 0; firstRoll = true;
} else {
score += roll;
}
}
}
System.out.println("The median is " + median() + ".");
}
public static int roll() {
return r.nextInt(6) + 1;
}
public static int median() {
Arrays.sort(totalScores);
int temp = totalScores[games / 2];
return temp;
}
}
You get 13 because that's the correct result. A little mathematics: if S is the random variable representing the score of any one of these games, then you can consider the Probability generating function f(z) of S. From the description of the game, this probability generating function satisfies the equation:
f(z) = (z + z^2 + z^3 + z^4 + z^5 + z^6) / 36 + f(z)(z + z^2 + z^3 + z^4 + z^5) / 6
This takes a bit of thought, or familiarity with this sort of construction: the left-hand term on the right-hand side takes account of the probabilities of getting 1 through 6 in a simple 2-roll game; the right-hand term involving f(z) takes account of games involving 3 or more rolls, expressing them in terms of the final pre-6 roll (which must be in the range 1 through 5) and the preceding rolls, whose probabilities we can express recursively using f again.
Anyway, after getting this far, one can rearrange to describe f as a rational function of z, and then expand as a power series, which begins:
f(z) = 1/36*z + 7/216*z^2 + 49/1296*z^3 + 343/7776*z^4 + 2401/46656*z^5 + 16807/279936*z^6 + 63217/1679616*z^7 + 388087/10077696*z^8 + 2335585/60466176*z^9 + 13681927/362797056*z^10 + 77103313/2176782336*z^11 + 409031959/13060694016*z^12 + 2371648321/78364164096*z^13 + 13583773735/470184984576*z^14 + ...
(I used Pari/GP to get this.)
The coefficient of z^k then describes the probability of the value of the game being k; thus there's a 1 in 36 chance of the score being 1, a 7 in 216 chance of getting 2, and so on. The sum of the first 12 coefficients is 0.472828864487196328..., while the sum of the first 13 coefficients is 0.5030933144224321950968.... So the median is indeed 13.
To provide an independent check, I wrote a quick Python program:
from __future__ import division
import random
def roll():
return random.randint(1, 6)
def play():
score = roll()
while True:
throw = roll()
if throw == 6:
break
score += throw
return score
all_scores = sorted(play() for _ in xrange(1000001))
print "median is: ",all_scores[len(all_scores) // 2]
print "fraction of scores <= 12: ",all_scores.index(13) / len(all_scores)
print "fraction of scores <= 13: ",all_scores.index(14) / len(all_scores)
Sure enough, here are the results:
iwasawa:~ mdickinson$ python dice_game.py
median is: 13
fraction of scores <= 12: 0.472811527188
fraction of scores <= 13: 0.502863497137
So to answer your question, the results you're seeing are not evidence of any sort of weakness in Java's random number generation.
Random is not perfectly random and has some deficiencies. However for this use case you are very unlikely to notice the difference. You can assume every value 1 to 6 is equally likely.
For comparison here is another solution which counts the number of occurrences of a total rather than recording every value. As you can see this performs well even if you have 1000x more games. This works best when you have a small number of outcomes and a high number duplicates. It is naturally sorted.
import java.util.Random;
public class DiceRoller {
private static final int MAX_VALUE = 300; // assume at most this total
private static final int GAMES = 10000001;
public static void main(String... args) {
int[] count = new int[MAX_VALUE];
Random rand = new Random();
for (int i = 0; i < GAMES; i++)
count[totalScore(rand)]++;
System.out.println("The median is " + median(count, GAMES) + ".");
}
private static int median(int[] count, int games) {
int findTotal = games/2;
for (int i = 0; i < count.length; i++) {
findTotal -= count[i];
if (findTotal <= 0) return i;
}
throw new AssertionError();
}
private static int totalScore(Random rand) {
int total = rand.nextInt(6) + 1;
for(int n;(n = rand.nextInt(6) + 1) != 6;)
total += n;
return total;
}
}
Here is some code that shows you the distribution of the results. It doesn't really answer the question, but maybe it helps you in your research.
package so7297660;
import java.util.Random;
public class DiceRoller {
private static final int N = 10000000;
private static final Random r = new Random();
private static final int[] result = new int[100];
public static int roll() {
return r.nextInt(6) + 1;
}
private static int singleGame() {
int score = roll();
while (true) {
int roll = roll();
if (roll == 6) {
return score;
} else {
score += roll;
}
}
}
private static int median() {
int n = 0;
for (int i = 0; i < result.length; i++) {
if (n + result[i] >= N / 2) {
return i;
}
n += result[i];
}
throw new IllegalStateException();
}
public static void main(String[] args) {
for (int i = 0; i < N; i++) {
int score = singleGame();
int index = Math.min(score, result.length - 1);
result[index]++;
}
for (int i = 0; i < result.length; i++) {
System.out.println(i + "\t" + result[i]);
}
System.out.println("median\t" + median());
}
}

simplifying fractions in Java

My task is to develop a rational class. If 500 and 1000 are my inputs, then (½) must be my output.
I have written a program on my own to find it.
Is there another best way to find the solution, or my program is already the best one?
public class Rational {
public static void main(String[] args){
int n1 = Integer.parseInt(args[0]);
int n2 = Integer.parseInt(args[1]);
int temp1 = n1;
int temp2 = n2;
while (n1 != n2){
if(n1 > n2)
n1 = n1 - n2;
else
n2 = n2 - n1;
}
int n3 = temp1 / n1 ;
int n4 = temp2 / n1 ;
System.out.print("\n Output :\n");
System.out.print(n3 + "/" + n4 + "\n\n" );
System.exit(0);
}
}
Interesting question. Here's some executable code that does it with minimal code:
/** #return the greatest common denominator */
public static long gcd(long a, long b) {
return b == 0 ? a : gcd(b, a % b);
}
public static String asFraction(long a, long b) {
long gcd = gcd(a, b);
return (a / gcd) + "/" + (b / gcd);
}
// Some tests
public static void main(String[] args) {
System.out.println(asFraction(500, 1000)); // "1/2"
System.out.println(asFraction(17, 3)); // "17/3"
System.out.println(asFraction(462, 1071)); // "22/51"
}
Bonus methods:
/** #return the lowest common multiple */
public static long lcm(long a, long b) {
return a * b / gcd(a, b);
}
/** #return the greatest common denominator */
public static long gcd(List<? extends Number> numbers) {
return numbers.stream().map(Number::longValue).reduce((a, b) -> gcd(a, b)).orElseThrow(NoSuchElementException::new);
}
/** #return the lowest common multiple */
public static long lcm(List<? extends Number> numbers) {
return numbers.stream().map(Number::longValue).reduce((a, b) -> lcm(a, b)).orElseThrow(NoSuchElementException::new);
}
You need the GCD. Either use BigInteger like Nathan mentioned or if you can't, use your own.
public int GCD(int a, int b){
if (b==0) return a;
return GCD(b,a%b);
}
Then you can divide each number by the GCD, like you have done above.
This will give you an improper fraction. If you need a mixed fraction then you can get the new numbers. Example if you had 1500 and 500 for inputs you would end up with 3/2 as your answer. Maybe you want 1 1/2. So you just divide 3/2 and get 1 and then get the remainder of 3/2 which is also 1. The denominator will stay the same.
whole = x/y;
numerator x%y;
denominator = y;
In case you don't believe me that this works, you can check out
http://en.wikipedia.org/wiki/Euclidean_algorithm
I just happen to like the recursive function because it's clean and simple.
Your algorithm is close, but not exactly correct. Also, you should probably create a new function if you want to find the gcd. Just makes it a little cleaner and easier to read. You can also test that function as well.
For reference, what you implemented is the original subtractive Euclidean Algorithm to calculate the greatest common divisor of two numbers.
A lot faster version is using the remainder from integer division, e.g. % instead of - in your loop:
while (n1 != 0 && n2 != 0){
if(n1 > n2)
n1 = n1 % n2;
else
n2 = n2 % n1;
}
... and then make sure you will use the one which is not zero.
A more streamlined version would be this:
while(n1 != 0) {
int old_n1 = n1;
n1 = n2 % n1;
n2 = old_n1;
}
and then use n1. Matt's answer shows a recursive version of the same algorithm.
You should make this class something other than a container for static methods. Here is a skeleton
import java.math.BigInteger;
public class BigRational
{
private BigInteger num;
private BigInteger denom;
public BigRational(BigInteger _num, BigInteger _denom)
{
//put the negative on top
// reduce BigRational using the BigInteger gcd method
}
public BigRational()
{
this(BigInteger.ZERO, BigInteger.ONE);
}
public BigRational add(BigRational that)
{
// return this + that;
}
.
.
.
//etc
}
}
I have a similar BigRational class I use. The GcdFunction is makes use of BigInteger's gcd function:
public class GcdFunction implements BinaryFunction {
#Override
public BigRational apply(final BigRational left, final BigRational right) {
if (!(left.isInteger() && right.isInteger())) {
throw new EvaluationException("GCD can only be applied to integers");
}
return new BigRational(left.getNumerator().gcd((right.getNumerator())));
}
}
BigRational contains a BigInteger numerator and denominator. isInteger() returns true if the simplified ratio's denominator is equal to 1.
Noticed that all answers here do not mention the iterative implementation of the Euclidean algorithm.
public static long gcdLongIterative(long a, long b) {
long tmp;
while (0 != b) {
tmp = b;
b = a % b;
a = tmp;
}
return a;
}
I implemented the validation test like #Bohemian and both recursive and iterative implementations work the same, however the iterative approach is faster. The benchmarks show small improvement, but it's improvement and overall it feels better to not use the stack so much and depend fully on the Java VM to optimize its implementation depend. Even if the benchmarks would be the same I would still feel better with the iterative as that would be more portable while the recursive was only optimized by my host Java, but might not be so well optimized on other's VMs.
Benchmark results (code is on the bottom of the answer):
(100 000 000 iterations)
gcd recursive: 3113ms
gcd iterative: 3079ms
gcd BigInteger: 13672ms
Signs:
One difference I noticed (besides the performance) is that the signs are handled differently, hand implemented Euclidean algorithm gcdLong and my gcdLongIterative behave the same, but both are different from BigInteger which tends to 'keep' the signs as they are. It seems that in essence the gcd and gcdLongIterative can return a negative number, while BigInteger will return positive only.
gcdLong and gcdLongIterative implementations:
-4/-2 => 2/1
-10/200 => 1/-20
10/-200 => 1/-20
BigInteger implementation tends to 'keep' the signs:
-4/-2 => -2/-1
-10/200 => -1/20
10/-200 => 1/-20
All results when used for fractions are valid, but it's worth considering post-process normalization if you expect the numbers in a specific 'style'.
For example, if the BigInteger behavior is preferred, then just returning absolute value should be enough, like here:
public static long gcdLongIterative(long a, long b) {
long tmp;
while (0 != b) {
tmp = b;
b = a % b;
a = tmp;
}
return Math.abs(a);
}
Performance:
Inspired by #Xabster benchmark (from Java: Get Greatest Common Divisor, which method is better?) I extended it to test all 3 implementations, in some cases both recursive and iterative were performing the same, however the iterative is slightly faster in most of the cases.
The benchmark code:
import java.math.BigInteger;
import java.util.Arrays;
import java.util.LinkedList;
import java.util.List;
import java.util.Random;
public class Test {
private static final int BENCHMARK_ITERATIONS = 100000000;
public static long gcdLong(long a, long b) {
return b == 0 ? a : gcdLong(b, a % b);
}
public static long gcdLongIterative(long a, long b) {
long tmp;
while (0 != b) {
tmp = b;
b = a % b;
a = tmp;
}
return a;
}
public static long gcdLongBigInteger(long a, long b) {
return BigInteger.valueOf(a).gcd(BigInteger.valueOf((b))).longValue();
}
public static String asFractionGcdLong(long a, long b) {
long gcd = gcdLong(a, b);
return (a / gcd) + "/" + (b / gcd);
}
public static String asFractionGcdLongIterative(long a, long b) {
long gcd = gcdLongIterative(a, b);
return (a / gcd) + "/" + (b / gcd);
}
public static String asFractionGcdLongBI(long a, long b) {
long gcd = gcdLongBigInteger(a, b);
return (a / gcd) + "/" + (b / gcd);
}
public static void test(String actual, String expected) {
boolean match = expected.equals(actual);
if (match) {
System.out.println("Actual and expected match=" + expected);
} else {
System.out.println("NO match expected=" + expected + " actual=" + actual);
}
}
public static class Values {
public long a;
public long b;
public String expected;
public Values(long a, long b, String expected) {
this.a = a;
this.b = b;
this.expected = expected;
}
}
public static void validityTest() {
List<Values> vals = new LinkedList<Values>(Arrays.asList(
new Values(500, 1000, "1/2"),
new Values(17, 3, "17/3"),
new Values(462, 1071, "22/51"),
new Values(-4, -2, "2/1"),
new Values(-10, 200, "1/-20"),
new Values(10, -200, "1/-20")
));
System.out.println("------ Recursive implementation -------");
vals.forEach(v -> test(asFractionGcdLong(v.a, v.b), v.expected));
System.out.println();
System.out.println("------ Iterative implementation -------");
vals.forEach(v -> test(asFractionGcdLongIterative(v.a, v.b), v.expected));
System.out.println();
System.out.println("------ BigInteger implementation -------");
vals.forEach(v -> test(asFractionGcdLongBI(v.a, v.b), v.expected));
System.out.println();
}
public static void benchMark() {
Random r = new Random();
long[] nums = new long[BENCHMARK_ITERATIONS];
for (int i = 0 ; i < nums.length ; i++) nums[i] = r.nextLong();
System.out.println("Waming up for benchmark...");
for (int i = 0 ; i < nums.length-1; i++) gcdLong(i, i + 1);
for (int i = 0 ; i < nums.length-1; i++) gcdLongIterative(i, i + 1);
for (int i = 0 ; i < nums.length-1; i++) gcdLongBigInteger(i, i + 1);
System.out.println("Started benchmark...");
long s = System.currentTimeMillis();
for (int i = 0 ; i < nums.length-1; i++) gcdLong(i, i + 1);
System.out.println("recursive: " + (System.currentTimeMillis() - s) + "ms");
s = System.currentTimeMillis();
for (int i = 0 ; i < nums.length-1; i++) gcdLongIterative(i, i + 1);
System.out.println("iterative: " + (System.currentTimeMillis() - s) + "ms");
s = System.currentTimeMillis();
for (int i = 0 ; i < nums.length-1; i++) gcdLongBigInteger(i, i + 1);
System.out.println("BigInteger: " + (System.currentTimeMillis() - s) + "ms");
}
public static void main(String[] args) {
validityTest();
benchMark();
}
}

Categories

Resources