how to create an Exp(-x^2) function? - java

I am using the "think java" book and I am stuck on exercise 7.6. The goal here is to write a function that can find . It gives you a couple hints:
One way to evaluate is
to use the infinite series expansion:
In other words, we need to add up a series of terms where the ith term
is equal to
Here is the code I came up with, but it is horribly wrong (when compared to Math.exp) for anything other than a power of 1. I don't understand why, as far as I can tell the code is correct with the formula from the book. I'm not sure if this is more of a math question or something related to how big of a number double and int can hold, but I am just trying to understand why this doesn't work.
public static void main(String[] args) {
System.out.println("Find exp(-x^2)");
double x = inDouble("Enter x: ");
System.out.println("myexp(" + -x*x + ") = " + gauss(x, 20));
System.out.println("Math.exp(" + -x*x + ") = " + Math.exp(-x*x));
}
public static double gauss(double x, int n) {
x = -x*x;
System.out.println(x);
double exp = 1;
double prevnum = 1;
int prevdenom = 1;
int i = 1;
while (i < n) {
exp = exp + (prevnum*x)/(prevdenom*i);
prevnum = prevnum*x;
prevdenom = prevdenom*i;
i++;
}
return exp;
} // I can't figure out why this is so inacurate, as far as I can tell the math is accurate to what the book says the formula is
public static double inDouble(String string) {
Scanner in = new Scanner (System.in);
System.out.print(string);
return in.nextDouble();
}

I am about to add to the comment on your question. I do this because I feel I have a slightly better implementation.
Your approach
Your approach is to have the function accept two arguments, where the second argument is the number of iterations. This isn't bad, but as #JamesKPolk pointed out, you might have to do some manual searching for an int (or long) that won't overflow
My approach
My approach would use something called the machine epsilon for a data type. The machine epsilon is the smallest number of that type (in your case, double) that is representable as that number. There exists algorithm for determining what that machine epsilon is, if you are not "allowed" to access machine epsilon in the Double class.
There is math behind this:
The series representation for your function is
Since it is alternating series, the error term is the absolute value of the first term you choose not to include (I leave the proof to you).
What this means is that we can have an error-based implementation that doesn't use iterations! The best part is that you could implement it for floats, and data types that are "more" than doubles! I present thus:
public static double gauss(double x)
{
x = -x*x;
double exp = 0, error = 1, numerator = 1, denominator = 1;
double machineEpsilon = 1.0;
// calculate machineEpsilon
while ((1.0 + 0.5 * machineEpsilon) != 1.0)
machineEpsilon = 0.5 * machineEpsilon;
int n = 0; //
// while the error is large enough to be representable in terms of the current data type
while ((error >= machineEpsilon) || (-error <= -machineEpsilon))
{
exp += error;
// calculate the numerator (it is 1 if we just start, but -x times its past value otherwise)
numerator = ((n == 0) ? 1 : -numerator * x);
// calculate the denominator (denominator gets multiplied by n)
denominator *= (n++);
// calculate error
error = numerator/denominator;
}
return exp;
}
Let me know how this works!

Related

How to compute approximate natural log of 2 n times in java?

The instructions were:
1. Compute the ln of 2, by adding up to n terms in the series.
a) You can approximate the natural logarithm of 2 with a series. The more terms
you use, the closer you get to the natural logarithm of 2. b) ln 2 = 1 - 1/2 + 1/3 - 1/4 + 1/5 - ... 1/n c) You will need to utilize a For Loop to solve this problem d) You will pass only 1 argument, the value of n, for this program. e) You will need to figure out how to change the sign for each consecutive term.
This is not difficult, but it will require some thought.
Forgive me if this doesn't sound very good, I am not the best at English. My code so far is this: even though I'm aware that I've done most wrong. I know it is wrong but I can not find out where to start or how to change the sign. I was at first doing this thinking we needed to find the ln of different numbers but later learned it was with input n.
EDIT: I believe I know what major part I was doing wrong. I fixed this to the best of my ability. I'm now a bit stuck because I'm playing out the logic in my head and I believe it should work. Instead all it prints out is 1.0
1.0
1.0
1.0
1.0
Can anyone help point out what I am doing wrong?
public class aprox_log {
static double findLog (int n)
{
double ln = 1;
for (int i = 1; i <= n; i++)
{
for (int k = 0; i <= n; k++) {
if ((k%2) == 0)
ln = ln - (1/i);
ln = ln + (1/i);
}
}
return 1 - ln;
}
public static void main(String[] args) {
//These lines print the results of test inputs.
System.out.println(findLog(2)); //~0.69314718056
System.out.println(findLog(5)); //~1.60943791243
System.out.println(findLog(10)); //~2.30258509299
System.out.println(findLog(9)); //~2.19722457734
System.out.println(findLog(1)); //0
}
}
NEW CODE:
public class aprox_log {
static double findLn (int n)
{
double ln = 0;
for (int i = 1; i <= n; i++)
{
if (i%2 == 0)
ln = ln - (1/i);
ln = ln + (1/i);
}
return ln;
public static void main(String[] args) {
//These lines print the results of test inputs.
System.out.println(findLn(2)); //0.5
System.out.println(findLn(5)); //0.783333333
System.out.println(findLn(10)); //0.64563492063
System.out.println(findLn(9)); //0.74563492063
System.out.println(findLn(1)); //1
}
}
`
To find the sum of the alternating harmonic series with n terms, just loop over the integers from 1 to n. Then, add the reciprocal of the current number if it is odd and subtract the reciprocal if it is even.
static double findLog(int n){
double res = 0;
for(int i = 1; i <= n; i++) res += 1d / (i % 2 == 0 ? -i: i);
return res;
}
In Java, the operator / calculates a result at the accuracy of the numbers it is given. In particular, if we divide two int, the result will be represented as int, rounding towards 0 if the result can not be represented exactly. That is, 1 / 2 is 0.
If you need floating point accuracy, you should use floating point numbers, for instance by declaring i to be of type double.

Mathematical Operations Without Using Java Math Class

I've been making my own class library in Java and I've run into a small annoyance. The library is centered around math. I started with the intention of not using the Java Math Class. Unfortunately, my lack of skill paired with my inability to find a resource online that tackles this problem has resulted in me falling back onto the Java Math Class. Is there a way I can do logarithms without using Math.log?
We could try to use a power series as per: https://math.stackexchange.com/a/61283
For example:
#Test
public void printMathCalculation() {
double libValue = Math.log(100);
double myValue = getLn(100);
System.out.println("Actual: \t" + libValue);
System.out.println("Approximate: \t" + myValue);
}
//Take double to an integer power
private double pow(double num, int power) {
double result = 1;
for(int i = 0; i < power; i++) {
result *= num;
}
return result;
}
//Get natural log
private double getLn(double num) {
int accuracy = 1000;
double sum = 0;
for(int n = 0; n < accuracy; n++) {
double num1 = (1.0/(2*n+1));
double num2 = (num-1)/(num+1);
sum += num1*pow(num2,2*n+1);
}
return 2 * sum;
}
The results I get are:
Actual: 4.605170185988092
Approximate: 4.605170185988078
You should use Math.log
Try this, using simple maths to calculate
public static int toLog2N(int num){
return num>1 ? 1 + toLog2N(num/2) : 0;
}
Its just an example, you should use Math class as it provides different methods like log(double a), log10(double a), log1p(double a). It will make your code more readable and easier to understand.
hope it helps!

Calculating a BigDecimal results in inaccurate answer

So I just wanted to screw around to see if I can make it so I can calculate E, but instead have it so I can have dynamic degrees of precision. While I did technically accomplish it, no matter what int i put in for the variable PRECISION, the last few numbers are always different from what the actual value of E is suppose to be. I'm not entirely sure why, but help would be appreciated.
import java.math.BigDecimal; //To use for calculating E
public class ComputeE {
public static double calcDenominator(int n)
{
double denominator = 1.0; //Start the BigInt with 1
for(int i = 1; i < n; i++) // Run n-1 amount of times
{
denominator = denominator * i; // Multiply BigInteger by the BigInteger obtained with the int value i
}
return denominator;
}
public static void main(String[] args) {
BigDecimal e = new BigDecimal(0.0);
int PRECISION = 15;
int iterations = 0;
for(int i = 0; i < PRECISION; i++)
{
iterations++;
BigDecimal numerator = new BigDecimal(1.0); // to divide, we need two BigDecimals, the numerator is 1
BigDecimal factorial = new BigDecimal(calcDenominator(i)); // the denominator is i! which we get from calling the factorial method
factorial = numerator.divide(factorial, PRECISION, BigDecimal.ROUND_UNNECESSARY); // compute 1/i!, note divide is overloaded, this version is used to
// ensure a limit to the iterations when division is limitless like 1/3
e = e.add(factorial); // add the latest 1/i! to e
}
System.out.println("Computed value of e : " + e);
System.out.println("Expected value of e : " + Math.E);
}
}
Rounding is necessary here. Use something like HALF_EVEN. Even better, use the enum value RoundingMode.HALF_EVEN, because the integer constants for rounding mode are deprecated.
In calcDenominator, change your for loop condition to i <= n, or else you'll add 1 one too many times in main and you'll get a value that's 1 too high.
You can use BigDecimal.ONE to initialize numerator. This doesn't affect the result, but why create an unnecessary object? Same comment on the initialization of e, except with BigDecimal.ZERO.
You are using the first PRECISION terms of an infinite series (Maclaurin Series) that approximates e, an irrational number. There is an error term when you cut off the for loop, and that is expected mathematically. With the above changes, and bumping PRECISION to 50, I get the following, which looks sufficiently precise.
Computed value of e : 2.71828182845904523536028747135266249775496954201584
Expected value of e : 2.718281828459045
It is precise, despite using the double constructor for BigDecimal because the significant digits for a double start with the first non-zero bit, so even if you're calculating 1/n! for large n, the significant digits are good enough for adding to the existing approximation for e.

Turning a iteration into a recursion

Good day!
I just wrote a code as an iteration which is supposed to sum up even numbers in between 0 and y.
I've been sitting now on my desk for about two hours thinking on how to write the same code in a recursion - without any progress so far.
The only explanations I find on the internet explain it for one simple repeat of one specific change - unlike my code which includes two. ("result = result + x;" and "x = x + 2;" )
Could someone please explain to me how I turn this kind of iteration into a recursion? Thanks in advance!
public class Sum {
static int method(int y) {
int result = 0;
for (int x = 2; x<=y;)
{
result = result + x;
x = x + 2;
}
return result;
}
public static void main(String[ ] args) {
int result = method(35);
System.out.println("Sum of even numbers between 0-35: " +result);
}
}
The total of the numbers is the total of this number plus the total of the number minus 2. Written in code:
int method(int y) {
if (y <= 0) {
return 0;
} else {
return y + method(y - 2);
}
}
Needless to say that recursion in this form is not necessary, and will create a StackoverflowException when y is a really large number. Some languages allow you to write a recursive function and indicate it is a tail recursion so that the compiler actually transforms it to an iteration.

power function to find the power where exponent is in decimal and less than 1

I was trying to create a program that finds the power of a real number . The problem is that exponent is in decimal and less than 1 but not negative.
suppose we have to find the power of
50.76
what i really tried was i wrote 0.76 as 76/100 and it would be 576/100
and after that i wrote
here is the code if you want to see what i did
public class Struct23 {
public static void main(String[] args) {
double x = 45;
int c=0;
StringBuffer y =new StringBuffer("0.23");
//checking whether the number is valid or not
for(int i =0;i<y.length();i++){
String subs = y.substring(i,i+1);
if(subs.equals(".")){
c=c+1;
}
}
if(c>1){
System.out.println("the input is wrong");
}
else{
String nep= y.delete(0, 2).toString();
double store = Double.parseDouble(nep);
int length = nep.length();
double rootnum = Math.pow(10, length);
double skit = power(x,store,rootnum);
System.out.println(skit);
}
}
static double power(double x,double store,double rootnum){
//to find the nth root of number
double number = Math.pow(x, 1/rootnum);
double power = Math.pow(number, store);
return power;
}
}
the answer would come but the main problem is that i cannot use pow function to do that
i can't also use exp() and log() functions.
i can only use
+
-
*
/
help me suggest your ideas .
thanks in advance
def newtons_sqrt(initial_guess, x, threshold=0.0001):
guess = initial_guess
new_guess = (guess+float(x)/guess)/2
while abs(guess-new_guess) > threshold :
guess=new_guess
new_guess = (guess+float(x)/guess)/2
return new_guess
def power(base, exp,threshold=0.00001):
if(exp >= 1): # first go fast!
temp = power(base, exp / 2);
return temp * temp
else: # now deal with the fractional part
low = 0
high = 1.0
sqr = newtons_sqrt(base/2,base)
acc = sqr
mid = high / 2
while(abs(mid - exp) > threshold):
sqr = newtons_sqrt(sqr/2.0,sqr)
if (mid <= exp):
low = mid
acc *= sqr
else:
high = mid
acc *= (1/sqr)
mid = (low + high) / 2;
return acc
print newtons_sqrt(1,8)
print 8**0.5
print power(5,0.76)
print 5**0.76
I reapropriated most of this answer from https://stackoverflow.com/a/7710097/541038
you could also expound on newtons_sqrt to give newtons_nth_root ... but then you have to figure out that 0.76 == 76/100 (which im sure isnt too hard really)
you can convert your number to complex form of it and then use de Moivre' formula to compute the nth root of your number using your legal oprations.

Categories

Resources