Related
I have two codes that theoretically should return the exact same output. However, this does not happen. The issue is that the two codes handle very small numbers (doubles) to the order of 1e-100 or so. I suspect that there could be some numerical issues which are related to that, and lead to the two outputs being different even though they should be theoretically the same.
Does it indeed make sense that handling numbers on the order of 1e-100 cause such problems? I don't mind the difference in output, if I could safely assume that the source is numerical issues. Does anyone have a good source/reference that talks about issues that come up with stability of algorithms when they handle numbers in such order?
Thanks.
Does anyone have a good source/reference that talks about issues that come up with stability of algorithms when they handle numbers in such order?
The first reference that comes to mind is What Every Computer Scientist Should Know About Floating-Point Arithmetic. It covers floating-point maths in general.
As far as numerical stability is concerned, the best references probably depend on the numerical algorithm in question. Two wide-ranging works that come to mind are:
Numerical Recipes by Press et al;
Matrix Computations by Golub and Van Loan.
It is not necessarily the small numbers that are causing the problem.
How do you check whether the outputs are the "exact same"?
I would check equality with tolerance. You may consider the floating point numbers x and y equal if either fabs(x-y) < 1.0e-6 or fabs(x-y) < fabs(x)*1.0e-6 holds.
Usually, there is a HUGE difference between the two algorithms if there are numerical issues. Often, a small change in the input may result in extreme changes in the output, if the algorithm suffers from numerical issues.
What makes you think that there are "numerical issues"?
If possible, change your algorithm to use Kahan Summation (aka compensated summation). From Wikipedia:
function KahanSum(input)
var sum = 0.0
var c = 0.0 //A running compensation for lost low-order bits.
for i = 1 to input.length do
y = input[i] - c //So far, so good: c is zero.
t = sum + y //Alas, sum is big, y small, so low-order digits of y are lost.
c = (t - sum) - y //(t - sum) recovers the high-order part of y; subtracting y recovers -(low part of y)
sum = t //Algebraically, c should always be zero. Beware eagerly optimising compilers!
//Next time around, the lost low part will be added to y in a fresh attempt.
return sum
This works by keeping a second running total of the cumulative error, similar to the Bresenham line drawing algorithm. The end result is that you get precision that is nearly double the data type's advertised precision.
Another technique I use is to sort my numbers from small to large (by manitude, ignoring sign) and add or subtract the small numbers first, then the larger ones. This has the virtue that if you add and subtract the same value multiple times, such numbers may cancel exactly and can be removed from the list.
When people talk about the use of "magic numbers" in computer programming, what do they mean?
Magic numbers are any number in your code that isn't immediately obvious to someone with very little knowledge.
For example, the following piece of code:
sz = sz + 729;
has a magic number in it and would be far better written as:
sz = sz + CAPACITY_INCREMENT;
Some extreme views state that you should never have any numbers in your code except -1, 0 and 1 but I prefer a somewhat less dogmatic view since I would instantly recognise 24, 1440, 86400, 3.1415, 2.71828 and 1.414 - it all depends on your knowledge.
However, even though I know there are 1440 minutes in a day, I would probably still use a MINS_PER_DAY identifier since it makes searching for them that much easier. Whose to say that the capacity increment mentioned above wouldn't also be 1440 and you end up changing the wrong value? This is especially true for the low numbers: the chance of dual use of 37197 is relatively low, the chance of using 5 for multiple things is pretty high.
Use of an identifier means that you wouldn't have to go through all your 700 source files and change 729 to 730 when the capacity increment changed. You could just change the one line:
#define CAPACITY_INCREMENT 729
to:
#define CAPACITY_INCREMENT 730
and recompile the lot.
Contrast this with magic constants which are the result of naive people thinking that just because they remove the actual numbers from their code, they can change:
x = x + 4;
to:
#define FOUR 4
x = x + FOUR;
That adds absolutely zero extra information to your code and is a total waste of time.
"magic numbers" are numbers that appear in statements like
if days == 365
Assuming you didn't know there were 365 days in a year, you'd find this statement meaningless. Thus, it's good practice to assign all "magic" numbers (numbers that have some kind of significance in your program) to a constant,
DAYS_IN_A_YEAR = 365
And from then on, compare to that instead. It's easier to read, and if the earth ever gets knocked out of alignment, and we gain an extra day... you can easily change it (other numbers might be more likely to change).
There's more than one meaning. The one given by most answers already (an arbitrary unnamed number) is a very common one, and the only thing I'll say about that is that some people go to the extreme of defining...
#define ZERO 0
#define ONE 1
If you do this, I will hunt you down and show no mercy.
Another kind of magic number, though, is used in file formats. It's just a value included as typically the first thing in the file which helps identify the file format, the version of the file format and/or the endian-ness of the particular file.
For example, you might have a magic number of 0x12345678. If you see that magic number, it's a fair guess you're seeing a file of the correct format. If you see, on the other hand, 0x78563412, it's a fair guess that you're seeing an endian-swapped version of the same file format.
The term "magic number" gets abused a bit, though, referring to almost anything that identifies a file format - including quite long ASCII strings in the header.
http://en.wikipedia.org/wiki/File_format#Magic_number
Wikipedia is your friend (Magic Number article)
Most of the answers so far have described a magic number as a constant that isn't self describing. Being a little bit of an "old-school" programmer myself, back in the day we described magic numbers as being any constant that is being assigned some special purpose that influences the behaviour of the code. For example, the number 999999 or MAX_INT or something else completely arbitrary.
The big problem with magic numbers is that their purpose can easily be forgotten, or the value used in another perfectly reasonable context.
As a crude and terribly contrived example:
while (int i != 99999)
{
DoSomeCleverCalculationBasedOnTheValueOf(i);
if (escapeConditionReached)
{
i = 99999;
}
}
The fact that a constant is used or not named isn't really the issue. In the case of my awful example, the value influences behaviour, but what if we need to change the value of "i" while looping?
Clearly in the example above, you don't NEED a magic number to exit the loop. You could replace it with a break statement, and that is the real issue with magic numbers, that they are a lazy approach to coding, and without fail can always be replaced by something less prone to either failure, or to losing meaning over time.
Anything that doesn't have a readily apparent meaning to anyone but the application itself.
if (foo == 3) {
// do something
} else if (foo == 4) {
// delete all users
}
Magic numbers are special value of certain variables which causes the program to behave in an special manner.
For example, a communication library might take a Timeout parameter and it can define the magic number "-1" for indicating infinite timeout.
The term magic number is usually used to describe some numeric constant in code. The number appears without any further description and thus its meaning is esoteric.
The use of magic numbers can be avoided by using named constants.
Using numbers in calculations other than 0 or 1 that aren't defined by some identifier or variable (which not only makes the number easy to change in several places by changing it in one place, but also makes it clear to the reader what the number is for).
In simple and true words, a magic number is a three-digit number, whose sum of the squares of the first two digits is equal to the third one.
Ex-202,
as, 2*2 + 0*0 = 2*2.
Now, WAP in java to accept an integer and print whether is a magic number or not.
It may seem a bit banal, but there IS at least one real magic number in every programming language.
0
I argue that it is THE magic wand to rule them all in virtually every programmer's quiver of magic wands.
FALSE is inevitably 0
TRUE is not(FALSE), but not necessarily 1! Could be -1 (0xFFFF)
NULL is inevitably 0 (the pointer)
And most compilers allow it unless their typechecking is utterly rabid.
0 is the base index of array elements, except in languages that are so antiquated that the base index is '1'. One can then conveniently code for(i = 0; i < 32; i++), and expect that 'i' will start at the base (0), and increment to, and stop at 32-1... the 32nd member of an array, or whatever.
0 is the end of many programming language strings. The "stop here" value.
0 is likewise built into the X86 instructions to 'move strings efficiently'. Saves many microseconds.
0 is often used by programmers to indicate that "nothing went wrong" in a routine's execution. It is the "not-an-exception" code value. One can use it to indicate the lack of thrown exceptions.
Zero is the answer most often given by programmers to the amount of work it would take to do something completely trivial, like change the color of the active cell to purple instead of bright pink. "Zero, man, just like zero!"
0 is the count of bugs in a program that we aspire to achieve. 0 exceptions unaccounted for, 0 loops unterminated, 0 recursion pathways that cannot be actually taken. 0 is the asymptote that we're trying to achieve in programming labor, girlfriend (or boyfriend) "issues", lousy restaurant experiences and general idiosyncracies of one's car.
Yes, 0 is a magic number indeed. FAR more magic than any other value. Nothing ... ahem, comes close.
rlynch#datalyser.com
Is there an upper limit to the number of bugs contained in a given program? If the number of instructions are known, could one say the program cannot contain more than 'n' bugs? For example, how many bugs could the following function contain?
double calcInterest(double amount) {
return -O.07 / amount;
}
A parser would count four terms in the function, and I could count these errors:
wrong number syntax
wrong interest rate (business requirements error)
wrong calculation (should be multiply)
Potential divide by zero
Clearly the number of bugs is not infinite given a finite number of instructions. Alternatively, one could say the function accepts 2^64 inputs, and of those, how many produce the correct output. However, is there any way to formally prove an upper limit?
If bug is "a requirement not met by the program", then there is no limit on the number of bugs (per line or otherwise), since there is no limit on the number of requirements.
print "hello world"
Might contain a million bugs. It doesn't create a pink elephant. I leave it to the reader to come up with 999999 other requirements not satisfied by this program.
Number of instructions have nothing to do with whether the program does what the user wants it to do. I mean, look at how poorly GCC does balancing my check book. Buggy as all get out, down right useless!
This would all depend on how you define a 'bug'.
If you define a program as a function from some input to some output, and a specification as a definition of that function, and a bug as any difference in output from the specification on a given input, then yes, you can conceivably have countably infinite bugs - however this is a somewhat useless definition of a bug.
The upper limit is the number of states your program can be in. Since this number is finite on real machines you could number the states from 1 to n. For each state you could label if this state is a bug or not. So yes, but even a small program having 16 bytes of memory has 2^128 states and the problem of analyzing all the different states is intractable.
There is a theoretical upper limit for bugs, but for all but the most trivial programs it is very nearly impossible to calculate, although engines such as Pex do give it the old college try.
Law of programming:
"If You will find all compile-time bugs, then n logical ones are still hidden, waiting to surprise You at run-time."
Depends on how you count bugs, which leads me to say "nope, no limit." I don't know about you, but I can easily write several bugs in the same line of code. For instance, how many bugs are in this Java code? :-P
public int addTwoNumbers(int x, String y)
{{
z == x + y;
return y;
}
As little as one if the bug is significant enough.
What is the best way to constrain the values of a PRNG to a smaller range? If you use modulus and the old max number is not evenly divisible by the new max number you bias toward the 0 through (old_max - new_max - 1). I assume the best way would be something like this (this is floating point, not integer math)
random_num = PRNG() / max_orginal_range * max_smaller_range
But something in my gut makes me question that method (maybe floating point implementation and representation differences?).
The random number generator will produce consistent results across hardware and software platforms, and the constraint needs to as well.
I was right to doubt the pseudocode above (but not for the reasons I was thinking). MichaelGG's answer got me thinking about the problem in a different way. I can model it using smaller numbers and test every outcome. So, let's assume we have a PRNG that produces a random number between 0 and 31 and you want the smaller range to be 0 to 9. If you use modulus you bias toward 0, 1, 2, and 3. If you use the pseudocode above you bias toward 0, 2, 5, and 7. I don't think there can be a good way to map one set into the other. The best that I have come up with so far is to regenerate the random numbers that are greater than old_max/new_max, but that has deep problems as well (reducing the period, time to generate new numbers until one is in the right range, etc.).
I think I may have naively approached this problem. It may be time to start some serious research into the literature (someone has to have tackled this before).
I know this might not be a particularly helpful answer, but I think the best way would be to conceive of a few different methods, then trying them out a few million times, and check the result sets.
When in doubt, try it yourself.
EDIT
It should be noted that many languages (like C#) have built in limiting in their functions
int maximumvalue = 20;
Random rand = new Random();
rand.Next(maximumvalue);
And whenever possible, you should use those rather than any code you would write yourself. Don't Reinvent The Wheel.
This problem is akin to rolling a k-sided die given only a p-sided die, without wasting randomness.
In this sense, by Lemma 3 in "Simulating a dice with a dice" by B. Kloeckner, this waste is inevitable unless "every prime number dividing k also divides p". Thus, for example, if p is a power of 2 (and any block of random bits is the same as rolling a die with a power of 2 number of faces) and k has prime factors other than 2, the best you can do is get arbitrarily close to no waste of randomness, such as by batching multiple rolls of the p-sided die until p^n is "close enough" to a power of k.
Let me also go over some of your concerns about regenerating random numbers:
"Reducing the period": Besides batching of bits, this concern can be dealt with in several ways:
Use a PRNG with a bigger "period" (maximum cycle length).
Add a Bays–Durham shuffle to the PRNG's implementation.
Use a "true" random number generator; this is not trivial.
Employ randomness extraction, which is discussed in Devroye and Gravel 2015-2020 and in my Note on Randomness Extraction. However, randomness extraction is pretty involved.
Ignore the problem, especially if it isn't a security application or serious simulation.
"Time to generate new numbers until one is in the right range": If you want unbiased random numbers, then any algorithm that does so will generally have to run forever in the worst case. Again, by Lemma 3, the algorithm will run forever in the worst case unless "every prime number dividing k also divides p", which is not the case if, say, k is 10 and p is 32.
See also the question: How to generate a random integer in the range [0,n] from a stream of random bits without wasting bits?, especially my answer there.
If PRNG() is generating uniformly distributed random numbers then the above looks good. In fact (if you want to scale the mean etc.) the above should be fine for all purposes. I guess you need to ask what the error associated with the original PRNG() is, and whether further manipulating will add to that substantially.
If in doubt, generate an appropriately sized sample set, and look at the results in Excel or similar (to check your mean / std.dev etc. for what you'd expect)
If you have access to a PRNG function (say, random()) that'll generate numbers in the range 0 <= x < 1, can you not just do:
random_num = (int) (random() * max_range);
to give you numbers in the range 0 to max_range?
Here's how the CLR's Random class works when limited (as per Reflector):
long num = maxValue - minValue;
if (num <= 0x7fffffffL) {
return (((int) (this.Sample() * num)) + minValue);
}
return (((int) ((long) (this.GetSampleForLargeRange() * num))) + minValue);
Even if you're given a positive int, it's not hard to get it to a double. Just multiply the random int by (1/maxint). Going from a 32-bit int to a double should provide adequate precision. (I haven't actually tested a PRNG like this, so I might be missing something with floats.)
Psuedo random number generators are essentially producing a random series of 1s and 0s, which when appended to each other, are an infinitely large number in base two. each time you consume a bit from you're prng, you are dividing that number by two and keeping the modulus. You can do this forever without wasting a single bit.
If you need a number in the range [0, N), then you need the same, but instead of base two, you need base N. It's basically trivial to convert the bases. Consume the number of bits you need, return the remainder of those bits back to your prng to be used next time a number is needed.
So far I've seen many posts dealing with equality of floating point numbers. The standard answer to a question like "how should we decide if x and y are equal?" is
abs(x - y) < epsilon
where epsilon is a fixed, small constant. This is because the "operands" x and y are often the results of some computation where a rounding error is involved, hence the standard equality operator == is not what we mean, and what we should really ask is whether x and y are close, not equal.
Now, I feel that if x is "almost equal" to y, then also x*10^20 should be "almost equal" to y*10^20, in the sense that the relative error should be the same (but "relative" to what?). But with these big numbers, the above test would fail, i.e. that solution does not "scale".
How would you deal with this issue? Should we rescale the numbers or rescale epsilon? How?
(Or is my intuition wrong?)
Here is a related question, but I don't like its accepted answer, for the reinterpret_cast thing seems a bit tricky to me, I don't understand what's going on. Please try to provide a simple test.
It all depends on the specific problem domain. Yes, using relative error will be more correct in the general case, but it can be significantly less efficient since it involves an extra floating-point division. If you know the approximate scale of the numbers in your problem, using an absolute error is acceptable.
This page outlines a number of techniques for comparing floats. It also goes over a number of important issues, such as those with subnormals, infinities, and NaNs. It's a great read, I highly recommend reading it all the way through.
As an alternative solution, why not just round or truncate the numbers and then make a straight comparison? By setting the number of significant digits in advance, you can be certain of the accuracy within that bound.
The problem is that with very big numbers, comparing to epsilon will fail.
Perhaps a better (but slower) solution would be to use division, example:
div(max(a, b), min(a, b)) < eps + 1
Now the 'error' will be relative.
Using relative error is at least not as bad as using absolute errors, but it has subtle problems for values near zero due to rounding issues. A far from perfect, but somewhat robust algorithm combines absolute and relative error approaches:
boolean approxEqual(float a, float b, float absEps, float relEps) {
// Absolute error check needed when comparing numbers near zero.
float diff = abs(a - b);
if (diff <= absEps) {
return true;
}
// Symmetric relative error check without division.
return (diff <= relEps * max(abs(a), abs(b)));
}
I adapted this code from Bruce Dawson's excellent article Comparing Floating Point Numbers, 2012 Edition, a required read for anyone doing floating-point comparisons -- an amazingly complex topic with many pitfalls.
Most of the time when code compares values, it is doing so to answer some sort of question. For example:
If I know what a function returned when given a value of X, can I assume it will return the same thing if given Y?
If I have a method of computing a function which is slow but accurate, I am willing to accept some inaccuracy in exchange for speed, and I want to test a candidate function which seems to fit the bill, are the outputs from that function close enough to the known-accurate one to be considered "correct".
To answer the first question, code should ideally do a bit-wise comparison on the value, though unless a language supports the new operators added to IEEE-754 in 2009 that may be less efficient than ideal. To answer the second question, one should define what degree of accuracy is required and test against that.
I don't think there's much merit in a general-purpose method which regards as equal things which are close, since different applications will have differing requirements for both absolute and relative tolerance, based upon what exact questions the tests are supposed to answer.