Related
Imagine you have 3 buckets, but each of them has a hole in it. I'm trying to fill a bath tub. The bath tub has a minimum level of water it needs and a maximum level of water it can contain. By the time you reach the tub with the bucket it is not clear how much water will be in the bucket, but you have a range of possible values.
Is it possible to adequately fill the tub with water?
Pretty much you have 3 ranges (min,max), is there some sum of them that will fall within a 4th range?
For example:
Bucket 1 : 5-10L
Bucket 2 : 15-25L
Bucket 3 : 10-50L
Bathtub 100-150L
Is there some guaranteed combination of 1 2 and 3 that will fill the bathtub within the requisite range? Multiples of each bucket can be used.
EDIT: Now imagine there are 50 different buckets?
If the capacity of the tub is not very large ( not greater than 10^6 for an example), we can solve it using dynamic programming.
Approach:
Initialization: memo[X][Y] is an array to memorize the result. X = number of buckets, Y = maximum capacity of the tub. Initialize memo[][] with -1.
Code:
bool dp(int bucketNum, int curVolume){
if(curVolume > maxCap)return false; // pruning extra branches
if(curVolume>=minCap && curVolume<=maxCap){ // base case on success
return true;
}
int &ret = memo[bucketNum][curVolume];
if(ret != -1){ // this state has been visited earlier
return false;
}
ret = false;
for(int i = minC[bucketNum]; i < = maxC[bucketNum]; i++){
int newVolume = curVolume + i;
for(int j = bucketNum; j <= 3; j++){
ret|=dp(j,newVolume);
if(ret == true)return ret;
}
}
return ret;
}
Warning: Code not tested
Here's a naïve recursive solution in python that works just fine (although it doesn't find an optimal solution):
def match_helper(lower, upper, units, least_difference, fail = dict()):
if upper < lower + least_difference:
return None
if fail.get((lower,upper)):
return None
exact_match = [ u for u in units if u['lower'] >= lower and u['upper'] <= upper ]
if exact_match:
return [ exact_match[0] ]
for unit in units:
if unit['upper'] > upper:
continue
recursive_match = match_helper(lower - unit['lower'], upper - unit['upper'], units, least_difference)
if recursive_match:
return [unit] + recursive_match
else:
fail[(lower,upper)] = 1
return None
def match(lower, upper):
units = [
{ 'name': 'Bucket 1', 'lower': 5, 'upper': 10 },
{ 'name': 'Bucket 2', 'lower': 15, 'upper': 25 },
{ 'name': 'Bucket 3', 'lower': 10, 'upper': 50 }
]
least_difference = min([ u['upper'] - u['lower'] for u in units ])
return match_helper(
lower = lower,
upper = upper,
units = sorted(units, key = lambda u: u['upper']),
least_difference = min([ u['upper'] - u['lower'] for u in units ]),
)
result = match(100, 175)
if result:
lower = sum([ u['lower'] for u in result ])
upper = sum([ u['upper'] for u in result ])
names = [ u['name'] for u in result ]
print lower, "-", upper
print names
else:
print "No solution"
It prints "No solution" for 100-150, but for 100-175 it comes up with a solution of 5x bucket 1, 5x bucket 2.
Assuming you are saying that the "range" for each bucket is the amount of water that it may have when it reaches the tub, and all you care about is if they could possibly fill the tub...
Just take the "max" of each bucket and sum them. If that is in the range of what you consider the tub to be "filled" then it can.
Updated:
Given that buckets can be used multiple times, this seems to me like we're looking for solutions to a pair of equations.
Given buckets x, y and z we want to find a, b and c:
a*x.min + b*y.min + c*z.min >= bathtub.min
and
a*x.max + b*y.max + c*z.max <= bathtub.max
Re: http://en.wikipedia.org/wiki/Diophantine_equation
If bathtub.min and bathtub.max are both multiples of the greatest common divisor of a,b and c, then there are infinitely many solutions (i.e. we can fill the tub), otherwise there are no solutions (i.e. we can never fill the tub).
This can be solved with multiple applications of the change making problem.
Each Bucket.Min value is a currency denomination, and Bathtub.Min is the target value.
When you find a solution via a change-making algorithm, then apply one more constraint:
sum(each Bucket.Max in your solution) <= Bathtub.max
If this constraint is not met, throw out this solution and look for another. This will probably require a change to a standard change-making algorithm that allows you to try other solutions when one is found to not be suitable.
Initially, your target range is Bathtub.Range.
Each time you add an instance of a bucket to the solution, you reduce the target range for the remaining buckets.
For example, using your example buckets and tub:
Target Range = 100..150
Let's say we want to add a Bucket1 to the candidate solution. That then gives us
Target Range = 95..140
because if the rest of the buckets in the solution total < 95, then this Bucket1 might not be sufficient to fill the tub to 100, and if the rest of the buckets in the solution total > 140, then this Bucket1 might fill the tub over 150.
So, this gives you a quick way to check if a candidate solution is valid:
TargetRange = Bathtub.Range
foreach Bucket in CandidateSolution
TargetRange.Min -= Bucket.Min
TargetRange.Max -= Bucket.Max
if TargetRange.Min == 0 AND TargetRange.Max >= 0 then solution found
if TargetRange.Min < 0 or TargetRange.Max < 0 then solution is invalid
This still leaves the question - How do you come up with the set of candidate solutions?
Brute force would try all possible combinations of buckets.
Here is my solution for finding the optimal solution (least number of buckets). It compares the ratio of the maximums to the ratio of the minimums, to figure out the optimal number of buckets to fill the tub.
private static void BucketProblem()
{
Range bathTub = new Range(100, 175);
List<Range> buckets = new List<Range> {new Range(5, 10), new Range(15, 25), new Range(10, 50)};
Dictionary<Range, int> result;
bool canBeFilled = SolveBuckets(bathTub, buckets, out result);
}
private static bool BucketHelper(Range tub, List<Range> buckets, Dictionary<Range, int> results)
{
Range bucket;
int startBucket = -1;
int fills = -1;
for (int i = buckets.Count - 1; i >=0 ; i--)
{
bucket = buckets[i];
double maxRatio = (double)tub.Maximum / bucket.Maximum;
double minRatio = (double)tub.Minimum / bucket.Minimum;
if (maxRatio >= minRatio)
{
startBucket = i;
if (maxRatio - minRatio > 1)
fills = (int) minRatio + 1;
else
fills = (int) maxRatio;
break;
}
}
if (startBucket < 0)
return false;
bucket = buckets[startBucket];
tub.Maximum -= bucket.Maximum * fills;
tub.Minimum -= bucket.Minimum * fills;
results.Add(bucket, fills);
return tub.Maximum == 0 || tub.Minimum <= 0 || startBucket == 0 || BucketHelper(tub, buckets.GetRange(0, startBucket), results);
}
public static bool SolveBuckets(Range tub, List<Range> buckets, out Dictionary<Range, int> results)
{
results = new Dictionary<Range, int>();
buckets = buckets.OrderBy(b => b.Minimum).ToList();
return BucketHelper(new Range(tub.Minimum, tub.Maximum), buckets, results);
}
In a programming contest, a problem was:
Count all solutions to the equation: x + 4y + 4z = n. You will be
given n and you will determine the count of solutions. Assume x, y and z are positive integers.
I have considered using triple for loops (brute force), but it was unefficient, causing TIME LIMIT EXCEED. (since the n may be = 1000,000):
int sol = 0;
for (int i = 1; i <= n; i++)
{
for (int j = 1; j <= n / 4; j++)
{
for (int k = 1; k <= n / 4; k++)
{
if (i + 4 * j + 4 * k == n)
sol++;
}
}
}
My friend could solve the problem. When I asked him, he said that he didn't use brute force at all. Instead, he converted the equation to a 'series' (i.e. summition). I asked him to tell how me but he refused :)
Can I know how?
This is particular case of coin change problem, which is solved in general by dynamic programming.
But here we can elaborate simple solution. I consider x,y,z > 0
x + 4*(y+z)=n
Let y + z = q = p + 1 (q > 1, p > 0)
x+4*q=n
x+4*p=n-4
There are M = Floor((n-5)/4) variants for x and p, hence there are M possible values of
q = 2..M+1
For every q>1 there are (q-1) variants of y and z: q = 1 + (q-1) = 2 + (q-2) +..+(q-1)+1
So we have N=1 + 2 + 3 + ... + M = M * (M + 1)/2 solutions
Example:
n = 15;
M = (15 - 5) div 4 = 2
N = 3
(3,1,2),(3,2,1),(7,1,1)
First note that n-x must be divisible by 4. Start by finding the smallest value that x can take:
start = 4
while ((n - start) % 4 != 0)
{
start = start + 1
}
From now on, you know that x will take values from [start, start+4, start+8 ...]. Now you can count the number of solutions by a simple counting loop:
count = 0
for (x = start; x < n - 4; x = x + 4)
{
y_z_sum = (n - x) / 4
count = count + y_z_sum - 1
}
For each choice of x, we can compute the value of y+z. For each value for y+z, there are y+z-1 possible choices (since y ranges from 1 to y+z-1, assuming that y and z are both positive integers).
Instead of a brute force solution with O(n3) running time, you can achieve O(n) this way.
This is a classic linear algebra problem. Please refer to any linear algebra textbook on how to solve a system of linear equations. One such method is called Gaussian Elimination.
I know that there is an algorithm that permits, given a combination of number (no repetitions, no order), calculates the index of the lexicographic order.
It would be very useful for my application to speedup things...
For example:
combination(10, 5)
1 - 1 2 3 4 5
2 - 1 2 3 4 6
3 - 1 2 3 4 7
....
251 - 5 7 8 9 10
252 - 6 7 8 9 10
I need that the algorithm returns the index of the given combination.
es: index( 2, 5, 7, 8, 10 ) --> index
EDIT: actually I'm using a java application that generates all combinations C(53, 5) and inserts them into a TreeMap.
My idea is to create an array that contains all combinations (and related data) that I can index with this algorithm.
Everything is to speedup combination searching.
However I tried some (not all) of your solutions and the algorithms that you proposed are slower that a get() from TreeMap.
If it helps: my needs are for a combination of 5 from 53 starting from 0 to 52.
Thank you again to all :-)
Here is a snippet that will do the work.
#include <iostream>
int main()
{
const int n = 10;
const int k = 5;
int combination[k] = {2, 5, 7, 8, 10};
int index = 0;
int j = 0;
for (int i = 0; i != k; ++i)
{
for (++j; j != combination[i]; ++j)
{
index += c(n - j, k - i - 1);
}
}
std::cout << index + 1 << std::endl;
return 0;
}
It assumes you have a function
int c(int n, int k);
that will return the number of combinations of choosing k elements out of n elements.
The loop calculates the number of combinations preceding the given combination.
By adding one at the end we get the actual index.
For the given combination there are
c(9, 4) = 126 combinations containing 1 and hence preceding it in lexicographic order.
Of the combinations containing 2 as the smallest number there are
c(7, 3) = 35 combinations having 3 as the second smallest number
c(6, 3) = 20 combinations having 4 as the second smallest number
All of these are preceding the given combination.
Of the combinations containing 2 and 5 as the two smallest numbers there are
c(4, 2) = 6 combinations having 6 as the third smallest number.
All of these are preceding the given combination.
Etc.
If you put a print statement in the inner loop you will get the numbers
126, 35, 20, 6, 1.
Hope that explains the code.
Convert your number selections to a factorial base number. This number will be the index you want. Technically this calculates the lexicographical index of all permutations, but if you only give it combinations, the indexes will still be well ordered, just with some large gaps for all the permutations that come in between each combination.
Edit: pseudocode removed, it was incorrect, but the method above should work. Too tired to come up with correct pseudocode at the moment.
Edit 2: Here's an example. Say we were choosing a combination of 5 elements from a set of 10 elements, like in your example above. If the combination was 2 3 4 6 8, you would get the related factorial base number like so:
Take the unselected elements and count how many you have to pass by to get to the one you are selecting.
1 2 3 4 5 6 7 8 9 10
2 -> 1
1 3 4 5 6 7 8 9 10
3 -> 1
1 4 5 6 7 8 9 10
4 -> 1
1 5 6 7 8 9 10
6 -> 2
1 5 7 8 9 10
8 -> 3
So the index in factorial base is 1112300000
In decimal base, it's
1*9! + 1*8! + 1*7! + 2*6! + 3*5! = 410040
This is Algorithm 2.7 kSubsetLexRank on page 44 of Combinatorial Algorithms by Kreher and Stinson.
r = 0
t[0] = 0
for i from 1 to k
if t[i - 1] + 1 <= t[i] - 1
for j from t[i - 1] to t[i] - 1
r = r + choose(n - j, k - i)
return r
The array t holds your values, for example [5 7 8 9 10]. The function choose(n, k) calculates the number "n choose k". The result value r will be the index, 251 for the example. Other inputs are n and k, for the example they would be 10 and 5.
zero-base,
# v: array of length k consisting of numbers between 0 and n-1 (ascending)
def index_of_combination(n,k,v):
idx = 0
for p in range(k-1):
if p == 0: arrg = range(1,v[p]+1)
else: arrg = range(v[p-1]+2, v[p]+1)
for a in arrg:
idx += combi[n-a, k-1-p]
idx += v[k-1] - v[k-2] - 1
return idx
Null Set has the right approach. The index corresponds to the factorial-base number of the sequence. You build a factorial-base number just like any other base number, except that the base decreases for each digit.
Now, the value of each digit in the factorial-base number is the number of elements less than it that have not yet been used. So, for combination(10, 5):
(1 2 3 4 5) == 0*9!/5! + 0*8!/5! + 0*7!/5! + 0*6!/5! + 0*5!/5!
== 0*3024 + 0*336 + 0*42 + 0*6 + 0*1
== 0
(10 9 8 7 6) == 9*3024 + 8*336 + 7*42 + 6*6 + 5*1
== 30239
It should be pretty easy to calculate the index incrementally.
If you have a set of positive integers 0<=x_1 < x_2< ... < x_k , then you could use something called the squashed order:
I = sum(j=1..k) Choose(x_j,j)
The beauty of the squashed order is that it works independent of the largest value in the parent set.
The squashed order is not the order you are looking for, but it is related.
To use the squashed order to get the lexicographic order in the set of k-subsets of {1,...,n) is by taking
1 <= x1 < ... < x_k <=n
compute
0 <= n-x_k < n-x_(k-1) ... < n-x_1
Then compute the squashed order index of (n-x_k,...,n-k_1)
Then subtract the squashed order index from Choose(n,k) to get your result, which is the lexicographic index.
If you have relatively small values of n and k, you can cache all the values Choose(a,b) with a
See Anderson, Combinatorics on Finite Sets, pp 112-119
I needed also the same for a project of mine and the fastest solution I found was (Python):
import math
def nCr(n,r):
f = math.factorial
return f(n) / f(r) / f(n-r)
def index(comb,n,k):
r=nCr(n,k)
for i in range(k):
if n-comb[i]<k-i:continue
r=r-nCr(n-comb[i],k-i)
return r
My input "comb" contained elements in increasing order You can test the code with for example:
import itertools
k=3
t=[1,2,3,4,5]
for x in itertools.combinations(t, k):
print x,index(x,len(t),k)
It is not hard to prove that if comb=(a1,a2,a3...,ak) (in increasing order) then:
index=[nCk-(n-a1+1)Ck] + [(n-a1)C(k-1)-(n-a2+1)C(k-1)] + ... =
nCk -(n-a1)Ck -(n-a2)C(k-1) - .... -(n-ak)C1
There's another way to do all this. You could generate all possible combinations and write them into a binary file where each comb is represented by it's index starting from zero. Then, when you need to find an index, and the combination is given, you apply a binary search on the file. Here's the function. It's written in VB.NET 2010 for my lotto program, it works with Israel lottery system so there's a bonus (7th) number; just ignore it.
Public Function Comb2Index( _
ByVal gAr() As Byte) As UInt32
Dim mxPntr As UInt32 = WHL.AMT.WHL_SYS_00 '(16.273.488)
Dim mdPntr As UInt32 = mxPntr \ 2
Dim eqCntr As Byte
Dim rdAr() As Byte
modBinary.OpenFile(WHL.WHL_SYS_00, _
FileMode.Open, FileAccess.Read)
Do
modBinary.ReadBlock(mdPntr, rdAr)
RP: If eqCntr = 7 Then GoTo EX
If gAr(eqCntr) = rdAr(eqCntr) Then
eqCntr += 1
GoTo RP
ElseIf gAr(eqCntr) < rdAr(eqCntr) Then
If eqCntr > 0 Then eqCntr = 0
mxPntr = mdPntr
mdPntr \= 2
ElseIf gAr(eqCntr) > rdAr(eqCntr) Then
If eqCntr > 0 Then eqCntr = 0
mdPntr += (mxPntr - mdPntr) \ 2
End If
Loop Until eqCntr = 7
EX: modBinary.CloseFile()
Return mdPntr
End Function
P.S. It takes 5 to 10 mins to generate 16 million combs on a Core 2 Duo. To find the index using binary search on file takes 397 milliseconds on a SATA drive.
Assuming the maximum setSize is not too large, you can simply generate a lookup table, where the inputs are encoded this way:
int index(a,b,c,...)
{
int key = 0;
key |= 1<<a;
key |= 1<<b;
key |= 1<<c;
//repeat for all arguments
return Lookup[key];
}
To generate the lookup table, look at this "banker's order" algorithm. Generate all the combinations, and also store the base index for each nItems. (For the example on p6, this would be [0,1,5,11,15]). Note that by you storing the answers in the opposite order from the example (LSBs set first) you will only need one table, sized for the largest possible set.
Populate the lookup table by walking through the combinations doing Lookup[combination[i]]=i-baseIdx[nItems]
EDIT: Never mind. This is completely wrong.
Let your combination be (a1, a2, ..., ak-1, ak) where a1 < a2 < ... < ak. Let choose(a,b) = a!/(b!*(a-b)!) if a >= b and 0 otherwise. Then, the index you are looking for is
choose(ak-1, k) + choose(ak-1-1, k-1) + choose(ak-2-1, k-2) + ... + choose (a2-1, 2) + choose (a1-1, 1) + 1
The first term counts the number of k-element combinations such that the largest element is less than ak. The second term counts the number of (k-1)-element combinations such that the largest element is less than ak-1. And, so on.
Notice that the size of the universe of elements to be chosen from (10 in your example) does not play a role in the computation of the index. Can you see why?
Sample solution:
class Program
{
static void Main(string[] args)
{
// The input
var n = 5;
var t = new[] { 2, 4, 5 };
// Helping transformations
ComputeDistances(t);
CorrectDistances(t);
// The algorithm
var r = CalculateRank(t, n);
Console.WriteLine("n = 5");
Console.WriteLine("t = {2, 4, 5}");
Console.WriteLine("r = {0}", r);
Console.ReadKey();
}
static void ComputeDistances(int[] t)
{
var k = t.Length;
while (--k >= 0)
t[k] -= (k + 1);
}
static void CorrectDistances(int[] t)
{
var k = t.Length;
while (--k > 0)
t[k] -= t[k - 1];
}
static int CalculateRank(int[] t, int n)
{
int k = t.Length - 1, r = 0;
for (var i = 0; i < t.Length; i++)
{
if (t[i] == 0)
{
n--;
k--;
continue;
}
for (var j = 0; j < t[i]; j++)
{
n--;
r += CalculateBinomialCoefficient(n, k);
}
n--;
k--;
}
return r;
}
static int CalculateBinomialCoefficient(int n, int k)
{
int i, l = 1, m, x, y;
if (n - k < k)
{
x = k;
y = n - k;
}
else
{
x = n - k;
y = k;
}
for (i = x + 1; i <= n; i++)
l *= i;
m = CalculateFactorial(y);
return l/m;
}
static int CalculateFactorial(int n)
{
int i, w = 1;
for (i = 1; i <= n; i++)
w *= i;
return w;
}
}
The idea behind the scenes is to associate a k-subset with an operation of drawing k-elements from the n-size set. It is a combination, so the overall count of possible items will be (n k). It is a clue that we could seek the solution in Pascal Triangle. After a while of comparing manually written examples with the appropriate numbers from the Pascal Triangle, we will find the pattern and hence the algorithm.
I used user515430's answer and converted to python3. Also this supports non-continuous values so you could pass in [1,3,5,7,9] as your pool instead of range(1,11)
from itertools import combinations
from scipy.special import comb
from pandas import Index
debugcombinations = False
class IndexedCombination:
def __init__(self, _setsize, _poolvalues):
self.setsize = _setsize
self.poolvals = Index(_poolvalues)
self.poolsize = len(self.poolvals)
self.totalcombinations = 1
fast_k = min(self.setsize, self.poolsize - self.setsize)
for i in range(1, fast_k + 1):
self.totalcombinations = self.totalcombinations * (self.poolsize - fast_k + i) // i
#fill the nCr cache
self.choose_cache = {}
n = self.poolsize
k = self.setsize
for i in range(k + 1):
for j in range(n + 1):
if n - j >= k - i:
self.choose_cache[n - j,k - i] = comb(n - j,k - i, exact=True)
if debugcombinations:
print('testnth = ' + str(self.testnth()))
def get_nth_combination(self,index):
n = self.poolsize
r = self.setsize
c = self.totalcombinations
#if index < 0 or index >= c:
# raise IndexError
result = []
while r:
c, n, r = c*r//n, n-1, r-1
while index >= c:
index -= c
c, n = c*(n-r)//n, n-1
result.append(self.poolvals[-1 - n])
return tuple(result)
def get_n_from_combination(self,someset):
n = self.poolsize
k = self.setsize
index = 0
j = 0
for i in range(k):
setidx = self.poolvals.get_loc(someset[i])
for j in range(j + 1, setidx + 1):
index += self.choose_cache[n - j, k - i - 1]
j += 1
return index
#just used to test whether nth_combination from the internet actually works
def testnth(self):
n = 0
_setsize = self.setsize
mainset = self.poolvals
for someset in combinations(mainset, _setsize):
nthset = self.get_nth_combination(n)
n2 = self.get_n_from_combination(nthset)
if debugcombinations:
print(str(n) + ': ' + str(someset) + ' vs ' + str(n2) + ': ' + str(nthset))
if n != n2:
return False
for x in range(_setsize):
if someset[x] != nthset[x]:
return False
n += 1
return True
setcombination = IndexedCombination(5, list(range(1,10+1)))
print( str(setcombination.get_n_from_combination([2,5,7,8,10])))
returns 188
An interview question:
Given a function f(x) that 1/4 times returns 0, 3/4 times returns 1.
Write a function g(x) using f(x) that 1/2 times returns 0, 1/2 times returns 1.
My implementation is:
function g(x) = {
if (f(x) == 0){ // 1/4
var s = f(x)
if( s == 1) {// 3/4 * 1/4
return s // 3/16
} else {
g(x)
}
} else { // 3/4
var k = f(x)
if( k == 0) {// 1/4 * 3/4
return k // 3/16
} else {
g(x)
}
}
}
Am I right? What's your solution?(you can use any language)
If you call f(x) twice in a row, the following outcomes are possible (assuming that
successive calls to f(x) are independent, identically distributed trials):
00 (probability 1/4 * 1/4)
01 (probability 1/4 * 3/4)
10 (probability 3/4 * 1/4)
11 (probability 3/4 * 3/4)
01 and 10 occur with equal probability. So iterate until you get one of those
cases, then return 0 or 1 appropriately:
do
a=f(x); b=f(x);
while (a == b);
return a;
It might be tempting to call f(x) only once per iteration and keep track of the two
most recent values, but that won't work. Suppose the very first roll is 1,
with probability 3/4. You'd loop until the first 0, then return 1 (with probability 3/4).
The problem with your algorithm is that it repeats itself with high probability. My code:
function g(x) = {
var s = f(x) + f(x) + f(x);
// s = 0, probability: 1/64
// s = 1, probability: 9/64
// s = 2, probability: 27/64
// s = 3, probability: 27/64
if (s == 2) return 0;
if (s == 3) return 1;
return g(x); // probability to go into recursion = 10/64, with only 1 additional f(x) calculation
}
I've measured average number of times f(x) was calculated for your algorithm and for mine. For yours f(x) was calculated around 5.3 times per one g(x) calculation. With my algorithm this number reduced to around 3.5. The same is true for other answers so far since they are actually the same algorithm as you said.
P.S.: your definition doesn't mention 'random' at the moment, but probably it is assumed. See my other answer.
Your solution is correct, if somewhat inefficient and with more duplicated logic. Here is a Python implementation of the same algorithm in a cleaner form.
def g ():
while True:
a = f()
if a != f():
return a
If f() is expensive you'd want to get more sophisticated with using the match/mismatch information to try to return with fewer calls to it. Here is the most efficient possible solution.
def g ():
lower = 0.0
upper = 1.0
while True:
if 0.5 < lower:
return 1
elif upper < 0.5:
return 0
else:
middle = 0.25 * lower + 0.75 * upper
if 0 == f():
lower = middle
else:
upper = middle
This takes about 2.6 calls to g() on average.
The way that it works is this. We're trying to pick a random number from 0 to 1, but we happen to stop as soon as we know whether the number is 0 or 1. We start knowing that the number is in the interval (0, 1). 3/4 of the numbers are in the bottom 3/4 of the interval, and 1/4 are in the top 1/4 of the interval. We decide which based on a call to f(x). This means that we are now in a smaller interval.
If we wash, rinse, and repeat enough times we can determine our finite number as precisely as possible, and will have an absolutely equal probability of winding up in any region of the original interval. In particular we have an even probability of winding up bigger than or less than 0.5.
If you wanted you could repeat the idea to generate an endless stream of bits one by one. This is, in fact, provably the most efficient way of generating such a stream, and is the source of the idea of entropy in information theory.
Given a function f(x) that 1/4 times returns 0, 3/4 times returns 1
Taking this statement literally, f(x) if called four times will always return zero once and 1 3 times. This is different than saying f(x) is a probabalistic function and the 0 to 1 ratio will approach 1 to 3 (1/4 vs 3/4) over many iterations. If the first interpretation is valid, than the only valid function for f(x) that will meet the criteria regardless of where in the sequence you start from is the sequence 0111 repeating. (or 1011 or 1101 or 1110 which are the same sequence from a different starting point). Given that constraint,
g()= (f() == f())
should suffice.
As already mentioned your definition is not that good regarding probability. Usually it means that not only probability is good but distribution also. Otherwise you can simply write g(x) which will return 1,0,1,0,1,0,1,0 - it will return them 50/50, but numbers won't be random.
Another cheating approach might be:
var invert = false;
function g(x) {
invert = !invert;
if (invert) return 1-f(x);
return f(x);
}
This solution will be better than all others since it calls f(x) only one time. But the results will not be very random.
A refinement of the same approach used in btilly's answer, achieving an average ~1.85 calls to f() per g() result (further refinement documented below achieves ~1.75, tbilly's ~2.6, Jim Lewis's accepted answer ~5.33). Code appears lower in the answer.
Basically, I generate random integers in the range 0 to 3 with even probability: the caller can then test bit 0 for the first 50/50 value, and bit 1 for a second. Reason: the f() probabilities of 1/4 and 3/4 map onto quarters much more cleanly than halves.
Description of algorithm
btilly explained the algorithm, but I'll do so in my own way too...
The algorithm basically generates a random real number x between 0 and 1, then returns a result depending on which "result bucket" that number falls in:
result bucket result
x < 0.25 0
0.25 <= x < 0.5 1
0.5 <= x < 0.75 2
0.75 <= x 3
But, generating a random real number given only f() is difficult. We have to start with the knowledge that our x value should be in the range 0..1 - which we'll call our initial "possible x" space. We then hone in on an actual value for x:
each time we call f():
if f() returns 0 (probability 1 in 4), we consider x to be in the lower quarter of the "possible x" space, and eliminate the upper three quarters from that space
if f() returns 1 (probability 3 in 4), we consider x to be in the upper three-quarters of the "possible x" space, and eliminate the lower quarter from that space
when the "possible x" space is completely contained by a single result bucket, that means we've narrowed x down to the point where we know which result value it should map to and have no need to get a more specific value for x.
It may or may not help to consider this diagram :-):
"result bucket" cut-offs 0,.25,.5,.75,1
0=========0.25=========0.5==========0.75=========1 "possible x" 0..1
| | . . | f() chooses x < vs >= 0.25
| result 0 |------0.4375-------------+----------| "possible x" .25..1
| | result 1| . . | f() chooses x < vs >= 0.4375
| | | . ~0.58 . | "possible x" .4375..1
| | | . | . | f() chooses < vs >= ~.58
| | ||. | | . | 4 distinct "possible x" ranges
Code
int g() // return 0, 1, 2, or 3
{
if (f() == 0) return 0;
if (f() == 0) return 1;
double low = 0.25 + 0.25 * (1.0 - 0.25);
double high = 1.0;
while (true)
{
double cutoff = low + 0.25 * (high - low);
if (f() == 0)
high = cutoff;
else
low = cutoff;
if (high < 0.50) return 1;
if (low >= 0.75) return 3;
if (low >= 0.50 && high < 0.75) return 2;
}
}
If helpful, an intermediary to feed out 50/50 results one at a time:
int h()
{
static int i;
if (!i)
{
int x = g();
i = x | 4;
return x & 1;
}
else
{
int x = i & 2;
i = 0;
return x ? 1 : 0;
}
}
NOTE: This can be further tweaked by having the algorithm switch from considering an f()==0 result to hone in on the lower quarter, to having it hone in on the upper quarter instead, based on which on average resolves to a result bucket more quickly. Superficially, this seemed useful on the third call to f() when an upper-quarter result would indicate an immediate result of 3, while a lower-quarter result still spans probability point 0.5 and hence results 1 and 2. When I tried it, the results were actually worse. A more complex tuning was needed to see actual benefits, and I ended up writing a brute-force comparison of lower vs upper cutoff for second through eleventh calls to g(). The best result I found was an average of ~1.75, resulting from the 1st, 2nd, 5th and 8th calls to g() seeking low (i.e. setting low = cutoff).
Here is a solution based on central limit theorem, originally due to a friend of mine:
/*
Given a function f(x) that 1/4 times returns 0, 3/4 times returns 1. Write a function g(x) using f(x) that 1/2 times returns 0, 1/2 times returns 1.
*/
#include <iostream>
#include <cstdlib>
#include <ctime>
#include <cstdio>
using namespace std;
int f() {
if (rand() % 4 == 0) return 0;
return 1;
}
int main() {
srand(time(0));
int cc = 0;
for (int k = 0; k < 1000; k++) { //number of different runs
int c = 0;
int limit = 10000; //the bigger the limit, the more we will approach %50 percent
for (int i=0; i<limit; ++i) c+= f();
cc += c < limit*0.75 ? 0 : 1; // c will be 0, with probability %50
}
printf("%d\n",cc); //cc is gonna be around 500
return 0;
}
Since each return of f() represents a 3/4 chance of TRUE, with some algebra we can just properly balance the odds. What we want is another function x() which returns a balancing probability of TRUE, so that
function g() {
return f() && x();
}
returns true 50% of the time.
So let's find the probability of x (p(x)), given p(f) and our desired total probability (1/2):
p(f) * p(x) = 1/2
3/4 * p(x) = 1/2
p(x) = (1/2) / 3/4
p(x) = 2/3
So x() should return TRUE with a probability of 2/3, since 2/3 * 3/4 = 6/12 = 1/2;
Thus the following should work for g():
function g() {
return f() && (rand() < 2/3);
}
Assuming
P(f[x] == 0) = 1/4
P(f[x] == 1) = 3/4
and requiring a function g[x] with the following assumptions
P(g[x] == 0) = 1/2
P(g[x] == 1) = 1/2
I believe the following definition of g[x] is sufficient (Mathematica)
g[x_] := If[f[x] + f[x + 1] == 1, 1, 0]
or, alternatively in C
int g(int x)
{
return f(x) + f(x+1) == 1
? 1
: 0;
}
This is based on the idea that invocations of {f[x], f[x+1]} would produce the following outcomes
{
{0, 0},
{0, 1},
{1, 0},
{1, 1}
}
Summing each of the outcomes we have
{
0,
1,
1,
2
}
where a sum of 1 represents 1/2 of the possible sum outcomes, with any other sum making up the other 1/2.
Edit.
As bdk says - {0,0} is less likely than {1,1} because
1/4 * 1/4 < 3/4 * 3/4
However, I am confused myself because given the following definition for f[x] (Mathematica)
f[x_] := Mod[x, 4] > 0 /. {False -> 0, True -> 1}
or alternatively in C
int f(int x)
{
return (x % 4) > 0
? 1
: 0;
}
then the results obtained from executing f[x] and g[x] seem to have the expected distribution.
Table[f[x], {x, 0, 20}]
{0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0}
Table[g[x], {x, 0, 20}]
{1, 0, 0, 1, 1, 0, 0, 1, 1, 0, 0, 1, 1, 0, 0, 1, 1, 0, 0, 1, 1}
This is much like the Monty Hall paradox.
In general.
Public Class Form1
'the general case
'
'twiceThis = 2 is 1 in four chance of 0
'twiceThis = 3 is 1 in six chance of 0
'
'twiceThis = x is 1 in 2x chance of 0
Const twiceThis As Integer = 7
Const numOf As Integer = twiceThis * 2
Private Sub Button1_Click(ByVal sender As System.Object, _
ByVal e As System.EventArgs) Handles Button1.Click
Const tries As Integer = 1000
y = New List(Of Integer)
Dim ct0 As Integer = 0
Dim ct1 As Integer = 0
Debug.WriteLine("")
''show all possible values of fx
'For x As Integer = 1 To numOf
' Debug.WriteLine(fx)
'Next
'test that gx returns 50% 0's and 50% 1's
Dim stpw As New Stopwatch
stpw.Start()
For x As Integer = 1 To tries
Dim g_x As Integer = gx()
'Debug.WriteLine(g_x.ToString) 'used to verify that gx returns 0 or 1 randomly
If g_x = 0 Then ct0 += 1 Else ct1 += 1
Next
stpw.Stop()
'the results
Debug.WriteLine((ct0 / tries).ToString("p1"))
Debug.WriteLine((ct1 / tries).ToString("p1"))
Debug.WriteLine((stpw.ElapsedTicks / tries).ToString("n0"))
End Sub
Dim prng As New Random
Dim y As New List(Of Integer)
Private Function fx() As Integer
'1 in numOf chance of zero being returned
If y.Count = 0 Then
'reload y
y.Add(0) 'fx has only one zero value
Do
y.Add(1) 'the rest are ones
Loop While y.Count < numOf
End If
'return a random value
Dim idx As Integer = prng.Next(y.Count)
Dim rv As Integer = y(idx)
y.RemoveAt(idx) 'remove the value selected
Return rv
End Function
Private Function gx() As Integer
'a function g(x) using f(x) that 50% of the time returns 0
' that 50% of the time returns 1
Dim rv As Integer = 0
For x As Integer = 1 To twiceThis
fx()
Next
For x As Integer = 1 To twiceThis
rv += fx()
Next
If rv = twiceThis Then Return 1 Else Return 0
End Function
End Class
I'm looking for an algorithm that places tick marks on an axis, given a range to display, a width to display it in, and a function to measure a string width for a tick mark.
For example, given that I need to display between 1e-6 and 5e-6 and a width to display in pixels, the algorithm would determine that I should put tickmarks (for example) at 1e-6, 2e-6, 3e-6, 4e-6, and 5e-6. Given a smaller width, it might decide that the optimal placement is only at the even positions, i.e. 2e-6 and 4e-6 (since putting more tickmarks would cause them to overlap).
A smart algorithm would give preference to tickmarks at multiples of 10, 5, and 2. Also, a smart algorithm would be symmetric around zero.
As I didn't like any of the solutions I've found so far, I implemented my own. It's in C# but it can be easily translated into any other language.
It basically chooses from a list of possible steps the smallest one that displays all values, without leaving any value exactly in the edge, lets you easily select which possible steps you want to use (without having to edit ugly if-else if blocks), and supports any range of values. I used a C# Tuple to return three values just for a quick and simple demonstration.
private static Tuple<decimal, decimal, decimal> GetScaleDetails(decimal min, decimal max)
{
// Minimal increment to avoid round extreme values to be on the edge of the chart
decimal epsilon = (max - min) / 1e6m;
max += epsilon;
min -= epsilon;
decimal range = max - min;
// Target number of values to be displayed on the Y axis (it may be less)
int stepCount = 20;
// First approximation
decimal roughStep = range / (stepCount - 1);
// Set best step for the range
decimal[] goodNormalizedSteps = { 1, 1.5m, 2, 2.5m, 5, 7.5m, 10 }; // keep the 10 at the end
// Or use these if you prefer: { 1, 2, 5, 10 };
// Normalize rough step to find the normalized one that fits best
decimal stepPower = (decimal)Math.Pow(10, -Math.Floor(Math.Log10((double)Math.Abs(roughStep))));
var normalizedStep = roughStep * stepPower;
var goodNormalizedStep = goodNormalizedSteps.First(n => n >= normalizedStep);
decimal step = goodNormalizedStep / stepPower;
// Determine the scale limits based on the chosen step.
decimal scaleMax = Math.Ceiling(max / step) * step;
decimal scaleMin = Math.Floor(min / step) * step;
return new Tuple<decimal, decimal, decimal>(scaleMin, scaleMax, step);
}
static void Main()
{
// Dummy code to show a usage example.
var minimumValue = data.Min();
var maximumValue = data.Max();
var results = GetScaleDetails(minimumValue, maximumValue);
chart.YAxis.MinValue = results.Item1;
chart.YAxis.MaxValue = results.Item2;
chart.YAxis.Step = results.Item3;
}
Take the longest of the segments about zero (or the whole graph, if zero is not in the range) - for example, if you have something on the range [-5, 1], take [-5,0].
Figure out approximately how long this segment will be, in ticks. This is just dividing the length by the width of a tick. So suppose the method says that we can put 11 ticks in from -5 to 0. This is our upper bound. For the shorter side, we'll just mirror the result on the longer side.
Now try to put in as many (up to 11) ticks in, such that the marker for each tick in the form i*10*10^n, i*5*10^n, i*2*10^n, where n is an integer, and i is the index of the tick. Now it's an optimization problem - we want to maximize the number of ticks we can put in, while at the same time minimizing the distance between the last tick and the end of the result. So assign a score for getting as many ticks as we can, less than our upper bound, and assign a score to getting the last tick close to n - you'll have to experiment here.
In the above example, try n = 1. We get 1 tick (at i=0). n = 2 gives us 1 tick, and we're further from the lower bound, so we know that we have to go the other way. n = 0 gives us 6 ticks, at each integer point point. n = -1 gives us 12 ticks (0, -0.5, ..., -5.0). n = -2 gives us 24 ticks, and so on. The scoring algorithm will give them each a score - higher means a better method.
Do this again for the i * 5 * 10^n, and i*2*10^n, and take the one with the best score.
(as an example scoring algorithm, say that the score is the distance to the last tick times the maximum number of ticks minus the number needed. This will likely be bad, but it'll serve as a decent starting point).
Funnily enough, just over a week ago I came here looking for an answer to the same question, but went away again and decided to come up with my own algorithm. I am here to share, in case it is of any use.
I wrote the code in Python to try and bust out a solution as quickly as possible, but it can easily be ported to any other language.
The function below calculates the appropriate interval (which I have allowed to be either 10**n, 2*10**n, 4*10**n or 5*10**n) for a given range of data, and then calculates the locations at which to place the ticks (based on which numbers within the range are divisble by the interval). I have not used the modulo % operator, since it does not work properly with floating-point numbers due to floating-point arithmetic rounding errors.
Code:
import math
def get_tick_positions(data: list):
if len(data) == 0:
return []
retpoints = []
data_range = max(data) - min(data)
lower_bound = min(data) - data_range/10
upper_bound = max(data) + data_range/10
view_range = upper_bound - lower_bound
num = lower_bound
n = math.floor(math.log10(view_range) - 1)
interval = 10**n
num_ticks = 1
while num <= upper_bound:
num += interval
num_ticks += 1
if num_ticks > 10:
if interval == 10 ** n:
interval = 2 * 10 ** n
elif interval == 2 * 10 ** n:
interval = 4 * 10 ** n
elif interval == 4 * 10 ** n:
interval = 5 * 10 ** n
else:
n += 1
interval = 10 ** n
num = lower_bound
num_ticks = 1
if view_range >= 10:
copy_interval = interval
else:
if interval == 10 ** n:
copy_interval = 1
elif interval == 2 * 10 ** n:
copy_interval = 2
elif interval == 4 * 10 ** n:
copy_interval = 4
else:
copy_interval = 5
first_val = 0
prev_val = 0
times = 0
temp_log = math.log10(interval)
if math.isclose(lower_bound, 0):
first_val = 0
elif lower_bound < 0:
if upper_bound < -2*interval:
if n < 0:
copy_ub = round(upper_bound*10**(abs(temp_log) + 1))
times = copy_ub // round(interval*10**(abs(temp_log) + 1)) + 2
else:
times = upper_bound // round(interval) + 2
while first_val >= lower_bound:
prev_val = first_val
first_val = times * copy_interval
if n < 0:
first_val *= (10**n)
times -= 1
first_val = prev_val
times += 3
else:
if lower_bound > 2*interval:
if n < 0:
copy_ub = round(lower_bound*10**(abs(temp_log) + 1))
times = copy_ub // round(interval*10**(abs(temp_log) + 1)) - 2
else:
times = lower_bound // round(interval) - 2
while first_val < lower_bound:
first_val = times*copy_interval
if n < 0:
first_val *= (10**n)
times += 1
if n < 0:
retpoints.append(first_val)
else:
retpoints.append(round(first_val))
val = first_val
times = 1
while val <= upper_bound:
val = first_val + times * interval
if n < 0:
retpoints.append(val)
else:
retpoints.append(round(val))
times += 1
retpoints.pop()
return retpoints
When passing in the following three data-points to the function
points = [-0.00493, -0.0003892, -0.00003292]
... the output I get (as a list) is as follows:
[-0.005, -0.004, -0.003, -0.002, -0.001, 0.0]
When passing this:
points = [1.399, 38.23823, 8309.33, 112990.12]
... I get:
[0, 20000, 40000, 60000, 80000, 100000, 120000]
When passing this:
points = [-54, -32, -19, -17, -13, -11, -8, -4, 12, 15, 68]
... I get:
[-60, -40, -20, 0, 20, 40, 60, 80]
... which all seem to be a decent choice of positions for placing ticks.
The function is written to allow 5-10 ticks, but that could easily be changed if you so please.
Whether the list of data supplied contains ordered or unordered data it does not matter, since it is only the minimum and maximum data points within the list that matter.
This simple algorithm yields an interval that is multiple of 1, 2, or 5 times a power of 10. And the axis range gets divided in at least 5 intervals. The code sample is in java language:
protected double calculateInterval(double range) {
double x = Math.pow(10.0, Math.floor(Math.log10(range)));
if (range / x >= 5)
return x;
else if (range / (x / 2.0) >= 5)
return x / 2.0;
else
return x / 5.0;
}
This is an alternative, for minimum 10 intervals:
protected double calculateInterval(double range) {
double x = Math.pow(10.0, Math.floor(Math.log10(range)));
if (range / (x / 2.0) >= 10)
return x / 2.0;
else if (range / (x / 5.0) >= 10)
return x / 5.0;
else
return x / 10.0;
}
I've been using the jQuery flot graph library. It's open source and does axis/tick generation quite well. I'd suggest looking at it's code and pinching some ideas from there.