Floating point rounding (in java) - java

What is the best way of determining if a given float(or double) has no significant decimal places.
f(234.0) = true
f(34.45) = false
f(3.1322) = false
i.e. equivalent of
EQ(((int)number) * 1.0 , number)
where EQ is a given method to compare floating points and it is OK to assume that the float fits in an integer.

Math.rint(x) == x
Math.rint() returns a double, so it also works for large numbers where the long result of Math.round() overflows.
Note that this also gives true for positive and negative infinity. You can explicitly exclude them by Math.rint(x) == x && !Double.isInfinite(x).

Round the value to the nearest integer, and calculate the absolute difference to the actual value.
If that difference is less than a certain percentage of the actual value you are close "enough".

You could try something like this:
public static boolean f(double d) {
return d % 1 == 0;
}

Related

Infinity in double primitive [duplicate]

Does Java have anything to represent infinity for every numerical data type? How is it implemented such that I can do mathematical operations with it?
E.g.
int myInf = infinity; //However it is done
myInf + 5; //returns infinity
myInf*(-1); //returns negative infinity
I have tried using very large numbers, but I want a proper, easy solution.
double supports Infinity
double inf = Double.POSITIVE_INFINITY;
System.out.println(inf + 5);
System.out.println(inf - inf); // same as Double.NaN
System.out.println(inf * -1); // same as Double.NEGATIVE_INFINITY
prints
Infinity
NaN
-Infinity
note: Infinity - Infinity is Not A Number.
I'm supposing you're using integer math for a reason. If so, you can get a result that's functionally nearly the same as POSITIVE_INFINITY by using the MAX_VALUE field of the Integer class:
Integer myInf = Integer.MAX_VALUE;
(And for NEGATIVE_INFINITY you could use MIN_VALUE.) There will of course be some functional differences, e.g., when comparing myInf to a value that happens to be MAX_VALUE: clearly this number isn't less than myInf. Also, as noted in the comments below, incrementing positive infinity will wrap you back around to negative numbers (and decrementing negative infinity will wrap you back to positive).
There's also a library that actually has fields POSITIVE_INFINITY and NEGATIVE_INFINITY, but they are really just new names for MAX_VALUE and MIN_VALUE.
To use Infinity, you can use Double which supports Infinity: -
System.out.println(Double.POSITIVE_INFINITY);
System.out.println(Double.POSITIVE_INFINITY * -1);
System.out.println(Double.NEGATIVE_INFINITY);
System.out.println(Double.POSITIVE_INFINITY - Double.NEGATIVE_INFINITY);
System.out.println(Double.POSITIVE_INFINITY - Double.POSITIVE_INFINITY);
OUTPUT: -
Infinity
-Infinity
-Infinity
Infinity
NaN
The Double and Float types have the POSITIVE_INFINITY constant.
Integer Infinity :
Integer maxNumber = Integer.MAX_VALUE
Double Infinity
Double maxNumber = Double.MAX_VALUE;
Double positiveInf = Double.POSITIVE_INFINITY;
Double negativeInf = Double.NEGATIVE_INFINITY
Float infinity
Float positiveInf = Float.POSITIVE_INFINITY;
Float negativeInf = Float.NEGATIVE_INFINITY
Float maxNumber = Float.MAX_VALUE;
I'm not sure that Java has infinity for every numerical type but for some numerical data types the answer is positive:
Float.POSITIVE_INFINITY
Float.NEGATIVE_INFINITY
or
Double.POSITIVE_INFINITY
Double.NEGATIVE_INFINITY
Also you may find useful the following article which represents some mathematical operations involving +/- infinity: Java Floating-Point Number Intricacies.
Only Double and Float type support POSITIVE_INFINITY constant.
A generic solution is to introduce a new type. It may be more involved, but it has the advantage of working for any type that doesn't define its own infinity.
If T is a type for which lteq is defined, you can define InfiniteOr<T> with lteq something like this:
class InfiniteOr with type parameter T:
field the_T of type null-or-an-actual-T
isInfinite()
return this.the_T == null
getFinite():
assert(!isInfinite());
return this.the_T
lteq(that)
if that.isInfinite()
return true
if this.isInfinite()
return false
return this.getFinite().lteq(that.getFinite())
I'll leave it to you to translate this to exact Java syntax. I hope the ideas are clear; but let me spell them out anyways.
The idea is to create a new type which has all the same values as some already existing type, plus one special value which—as far as you can tell through public methods—acts exactly the way you want infinity to act, e.g. it's greater than anything else. I'm using null to represent infinity here, since that seems the most straightforward in Java.
If you want to add arithmetic operations, decide what they should do, then implement that. It's probably simplest if you handle the infinite cases first, then reuse the existing operations on finite values of the original type.
There might or might not be a general pattern to whether or not it's beneficial to adopt a convention of handling left-hand-side infinities before right-hand-side infinities or vice versa; I can't tell without trying it out, but for less-than-or-equal (lteq) I think it's simpler to look at right-hand-side infinity first. I note that lteq is not commutative, but add and mul are; maybe that is relevant.
Note: coming up with a good definition of what should happen on infinite values is not always easy. It is for comparison, addition and multiplication, but maybe not subtraction. Also, there is a distinction between infinite cardinal and ordinal numbers which you may want to pay attention to.
For the numeric wrapper types.
e.g Double.POSITIVE_INFINITY
Hope this might help you.
Since the class Number is not final, here is an
idea, that I don't find yet in the other posts.
Namely to subclass the class Number.
This would somehow deliver an object that can be treated
as infinity for Integer, Long, Double, Float,
BigInteger and BigDecimal.
Since there are only two values, we could use the singleton pattern:
public final class Infinity extends Number {
public final static Infinity POSITIVE = new Infinity(false);
public final static Infinity NEGATIVE = new Infinity(true);
private boolean negative;
private Infinity(boolean n) {
negative = n;
}
}
Somehow I think the remaining methods intValue(), longValue()
etc.. should then be overriden to throw an exceptions. So that
the infinity value cannot be used without further precautions.
I'm a beginner in Java...
I found another implementation for the infinity in the Java documentation, for the boolean and double types.
https://docs.oracle.com/javase/specs/jls/se7/html/jls-4.html#jls-4.2.3
Positive zero and negative zero compare equal; thus the result of the
expression 0.0==-0.0 is true and the result of 0.0>-0.0 is false. But
other operations can distinguish positive and negative zero; for
example, 1.0/0.0 has the value positive infinity, while the value of
1.0/-0.0 is negative infinity.
It looks ugly, but it works.
public class Main {
public static void main(String[] args) {
System.out.println(1.0/0.0);
System.out.println(-1.0/0.0);
}
}

Difference between number/10 and number*0.1 in java

I've been working on an interview question for 1.5 hours and could not find the bug in my Java program.
And then I found what the problem was, which I don't understand (don't pay attention to the values, there were others, it's about the types):
int size=100;
Integer a=12;
if(a >= size/10)...
//didn't work
is different than
if(a >= size*0.1)...
//worked
I understand that there is a conversion, but still, how is it possible that with a=12, if(a>=size/10) returns false?
Why is that?
/10 is integer division. While *0.1 first converts the first operand to a double and performs a floating point multiplication.
If you use the /10, and the operand is 14, it will result in 1 indeed, 14/10=1.4 but integer division rounds this down. Thus 29/10=2.
If you use *0.1, the Java compiler will first convert the value of size to a double, thus 14.0 and then muliplies it with 0.1 resulting in 1.4.
On the other hand it's not all beaty that comes out of floating points. float and double can't represent every integer, and round off after computation.
For the given values for size however, it will result in the effect because 100 is a multiple of 10 and a float or double is capable of representing any integer value in the range from zero to hundred.
Finally /10 is not always an integer division: if the first operand is a floating point (e.g. 14.0d/10), the compiler will convert this to a floating point division.
Short version:
int/int is an integer division that rounds down to the nearest (lower) integer.
int*double is a double multiplication that - with rounding off errors - results in the floating point value, nearest to the correct result (with decimal digits).
I just tested here:
public class a {
public static void main(String[] args) {
int size = 100;
int a = 12;
System.out.println((a >= size / 10) ? "OK" : "Failed?");
}
}
And it worked. I don't think this is your real problem. Probably it's in another part of your code.

Java: Adding Double.MIN_NORMAL to a double doesn't change the number

I thought that MIN_NORMAL was a value that you could add to a "normal" double and the number will change. E.g. add Double.MIN_NORMAL to 0.1d and you get a value different from 0.1d, however my understanding is wrong:
public static void test(double val) {
if (val == (val - Double.MIN_NORMAL*1e50d))
System.out.printf("val == (val - Double.MIN_NORMAL*1e50d) for val=%.20f\n", val);
else
System.out.printf("val != (val - Double.MIN_NORMAL*1e50d) for val=%.20f\n", val);
}
Which produces:
test(0.0d);
> val != (val - Double.MIN_NORMAL*1e50d) for val=0.00000000000000000000
test(1.0d);
> val == (val - Double.MIN_NORMAL*1e50d) for val=1.00000000000000000000
test(0.1d);
> val == (val - Double.MIN_NORMAL*1e50d) for val=0.10000000000000000000
Somebody pls explain what's going against my logic here, that even if I add MIN_NORMAL times 1e50d, I still get the same number.
I checked binary representations and 1 * Double.MIN_NORMAL is different from 2 * Double.MIN_NORMAL, but subtracting those from anything except zero does not change the original number.
MIN_NORMAL is, as the Javadoc says, only the smallest normalized double value. But that doesn't mean it is something like 1 for ints: For floating-point point values, there simply is no standard "eps" you can add to change to the next representable value -- the "eps" always depends on the exponent of the given floating-point value. That's why they're called floating points, in the end :)
However, Java 1.6+ provides Math.nextAfter() which returns, for any given double, the next or previous representable double.
On older Java versions, you can always mess around with Double.doubleToLongBits(), incrementing or decrementing its result, and converting back by Double.longBitsToDouble(); this gets you the next or previous representable double value -- in most cases: there are a few special cases (NaN, infinite values), so this is not recommended to floating-point newbies :)
Double has a limited precision. MIN_NORMAL is 2e-1022. It will be dropped unless the number you add it to is also in the ballpark of 2e-1000.

Smallest epsilon so that comparison result change

What is the smallest float value A so that (x < x + A) == true?
I tried with Float.MIN_VALUE but surprisingly(? [1]) it doesn't work (except for values of 0.)
Knowing how the IEEE 754 standard stores float values, I could just add 1 to the mantissa of the float in question, but this seams really hackish. I don't want to put byte arrays and bit operations in my code for such a trivial matter, especially with Java. In addition if I simply add 1 to the Float.floatToIntBits() and the mantissa is all 1, it will increase the exponent by 1 and set the mantissa to 0. I don't want to implements all the handling of this cases if it is not necessary.
Isn't there some sort of function (hopefully build-in) that given the float x, it returns the smallest float A such that (x < x + A) == true?
If there isn't, what would be the cleanest way to implement it?
I'm using this because of how I'm iterating over a line of vertices
// return the next vertices strictly at the left of pNewX
float otherLeftX = pOppositeToCurrentCave.leftVertexTo(pNewX);
// we add MIN_VALUE so that the next call to leftVertexTo will return the same vertex returned by leftVertexTo(pNewX)
otherLeftX += Float.MIN_VALUE;
while(otherLeftX >= 0 && pOppositeToCurrentCave.hasLeftVertexTo(otherLeftX)) {
otherLeftX = pOppositeToCurrentCave.leftVertexTo(otherLeftX);
//stuff
}
Right now because of this problem the first vertex is always skipped because the second call to leftVertexTo(otherLeftX) doesn't return the same value it returned on the first call
[1] Not so surprising. I happened to realize after I noticed the problem that since the gap between floats is relative, for whatever number != 0 the MIN_VALUE is so small that it will be truncated and (x = x + FLOAT.MIN_VALUE) == true
You can try Math.nextUp(x)
Here is the doc:
Returns the floating-point value adjacent to f in the direction of positive infinity. This method is semantically equivalent to nextAfter(f, Float.POSITIVE_INFINITY); however, a nextUp implementation may run faster than its equivalent nextAfter call.
Special Cases:
If the argument is NaN, the result is NaN.
If the argument is positive infinity, the result is positive infinity.
If the argument is zero, the result is Float.MIN_VALUE
Parameters:
f - starting floating-point value
Returns:
The adjacent floating-point value closer to positive infinity.

How to implement infinity in Java?

Does Java have anything to represent infinity for every numerical data type? How is it implemented such that I can do mathematical operations with it?
E.g.
int myInf = infinity; //However it is done
myInf + 5; //returns infinity
myInf*(-1); //returns negative infinity
I have tried using very large numbers, but I want a proper, easy solution.
double supports Infinity
double inf = Double.POSITIVE_INFINITY;
System.out.println(inf + 5);
System.out.println(inf - inf); // same as Double.NaN
System.out.println(inf * -1); // same as Double.NEGATIVE_INFINITY
prints
Infinity
NaN
-Infinity
note: Infinity - Infinity is Not A Number.
I'm supposing you're using integer math for a reason. If so, you can get a result that's functionally nearly the same as POSITIVE_INFINITY by using the MAX_VALUE field of the Integer class:
Integer myInf = Integer.MAX_VALUE;
(And for NEGATIVE_INFINITY you could use MIN_VALUE.) There will of course be some functional differences, e.g., when comparing myInf to a value that happens to be MAX_VALUE: clearly this number isn't less than myInf. Also, as noted in the comments below, incrementing positive infinity will wrap you back around to negative numbers (and decrementing negative infinity will wrap you back to positive).
There's also a library that actually has fields POSITIVE_INFINITY and NEGATIVE_INFINITY, but they are really just new names for MAX_VALUE and MIN_VALUE.
To use Infinity, you can use Double which supports Infinity: -
System.out.println(Double.POSITIVE_INFINITY);
System.out.println(Double.POSITIVE_INFINITY * -1);
System.out.println(Double.NEGATIVE_INFINITY);
System.out.println(Double.POSITIVE_INFINITY - Double.NEGATIVE_INFINITY);
System.out.println(Double.POSITIVE_INFINITY - Double.POSITIVE_INFINITY);
OUTPUT: -
Infinity
-Infinity
-Infinity
Infinity
NaN
The Double and Float types have the POSITIVE_INFINITY constant.
Integer Infinity :
Integer maxNumber = Integer.MAX_VALUE
Double Infinity
Double maxNumber = Double.MAX_VALUE;
Double positiveInf = Double.POSITIVE_INFINITY;
Double negativeInf = Double.NEGATIVE_INFINITY
Float infinity
Float positiveInf = Float.POSITIVE_INFINITY;
Float negativeInf = Float.NEGATIVE_INFINITY
Float maxNumber = Float.MAX_VALUE;
I'm not sure that Java has infinity for every numerical type but for some numerical data types the answer is positive:
Float.POSITIVE_INFINITY
Float.NEGATIVE_INFINITY
or
Double.POSITIVE_INFINITY
Double.NEGATIVE_INFINITY
Also you may find useful the following article which represents some mathematical operations involving +/- infinity: Java Floating-Point Number Intricacies.
Only Double and Float type support POSITIVE_INFINITY constant.
A generic solution is to introduce a new type. It may be more involved, but it has the advantage of working for any type that doesn't define its own infinity.
If T is a type for which lteq is defined, you can define InfiniteOr<T> with lteq something like this:
class InfiniteOr with type parameter T:
field the_T of type null-or-an-actual-T
isInfinite()
return this.the_T == null
getFinite():
assert(!isInfinite());
return this.the_T
lteq(that)
if that.isInfinite()
return true
if this.isInfinite()
return false
return this.getFinite().lteq(that.getFinite())
I'll leave it to you to translate this to exact Java syntax. I hope the ideas are clear; but let me spell them out anyways.
The idea is to create a new type which has all the same values as some already existing type, plus one special value which—as far as you can tell through public methods—acts exactly the way you want infinity to act, e.g. it's greater than anything else. I'm using null to represent infinity here, since that seems the most straightforward in Java.
If you want to add arithmetic operations, decide what they should do, then implement that. It's probably simplest if you handle the infinite cases first, then reuse the existing operations on finite values of the original type.
There might or might not be a general pattern to whether or not it's beneficial to adopt a convention of handling left-hand-side infinities before right-hand-side infinities or vice versa; I can't tell without trying it out, but for less-than-or-equal (lteq) I think it's simpler to look at right-hand-side infinity first. I note that lteq is not commutative, but add and mul are; maybe that is relevant.
Note: coming up with a good definition of what should happen on infinite values is not always easy. It is for comparison, addition and multiplication, but maybe not subtraction. Also, there is a distinction between infinite cardinal and ordinal numbers which you may want to pay attention to.
For the numeric wrapper types.
e.g Double.POSITIVE_INFINITY
Hope this might help you.
Since the class Number is not final, here is an
idea, that I don't find yet in the other posts.
Namely to subclass the class Number.
This would somehow deliver an object that can be treated
as infinity for Integer, Long, Double, Float,
BigInteger and BigDecimal.
Since there are only two values, we could use the singleton pattern:
public final class Infinity extends Number {
public final static Infinity POSITIVE = new Infinity(false);
public final static Infinity NEGATIVE = new Infinity(true);
private boolean negative;
private Infinity(boolean n) {
negative = n;
}
}
Somehow I think the remaining methods intValue(), longValue()
etc.. should then be overriden to throw an exceptions. So that
the infinity value cannot be used without further precautions.
I'm a beginner in Java...
I found another implementation for the infinity in the Java documentation, for the boolean and double types.
https://docs.oracle.com/javase/specs/jls/se7/html/jls-4.html#jls-4.2.3
Positive zero and negative zero compare equal; thus the result of the
expression 0.0==-0.0 is true and the result of 0.0>-0.0 is false. But
other operations can distinguish positive and negative zero; for
example, 1.0/0.0 has the value positive infinity, while the value of
1.0/-0.0 is negative infinity.
It looks ugly, but it works.
public class Main {
public static void main(String[] args) {
System.out.println(1.0/0.0);
System.out.println(-1.0/0.0);
}
}

Categories