I have recently attempted to concisely draw several graphs in a plot using gnuplot and the plot for ... syntax. In this case, I needed nested loops because I wanted to pass something like the following index combinations (simplified here) to the plot expression:
i = 0, j = 0
i = 1, j = 0
i = 1, j = 1
i = 2, j = 0
i = 2, j = 1
i = 2, j = 2
and so on.
So i loops from 0 to some upper limit N and for each iteration of i, j loops from 0 to i (so i <= j). I tried doing this with the following:
# f(i, j, x) = ...
N = 5
plot for [i=0:N] for [j=0:i] f(i, j, x) title sprintf('j = %d', j)
but this only gives five iterations with j = 0 every time (as shown by the title). So it seems that gnuplot only evaluates the for expressions once, fixing i = 0 at the beginning and not re-evaluating to keep up with changing i values. Something like this has already been hinted at in this answer (“in the plot for ... structure the second index cannot depend on the first one.”).
Is there a simple way to do what I want in gnuplot (i.e. use the combinations of indices given above with some kind of loop)? There is the do for { ... } structure since gnuplot 4.6, but that requires individual statements in its body, so it can’t be used to assemble a single plot statement. I suppose one could use multiplot to get around this, but I’d like to avoid multiplot if possible because it makes things more complicated than seems necessary.
I took your problem personally. For your specific problem you can use a mathematical trick. Remap your indices (i,j) to a single index k, such that
(0,0) -> (0)
(1,0) -> (1)
(1,1) -> (2)
(2,0) -> (3)
...
It can be shown that the relation between i and j and k is
k = i*(i+1)/2 + j
which can be inverted with a bit of algebra
i(k)=floor((sqrt(1+8.*k)-1.)/2.)
j(k)=k-i(k)*(i(k)+1)/2
Now, you can use a single index k in your loop
N = 5
kmax = N*(N+1)/2 + N
plot for [k=0:kmax] f(i(k), j(k), x) title sprintf('j = %d', j(k))
Related
Given n pairs of integers. Split into two subsets A and B to minimize sum(maximum difference among first values of A, maximum difference among second values of B).
Example : n = 4
{0, 0}; {5;5}; {1; 1}; {3; 4}
A = {{0; 0}; {1; 1}}
B = {{5; 5}; {3; 4}}
(maximum difference among first values of A, maximum difference among second values of B).
(maximum difference among first values of A) = fA_max - fA_min = 1 - 0 = 1
(maximum difference among second values of B) = sB_max - sB_min = 5 - 4 = 1
Therefore, the answer if 1 + 1 = 2. And this is the best way.
Obviously, maximum difference among the values equals to (maximum value - minimum value). Hence, what we need to do is find the minimum of (fA_max - fA_min) + (sB_max - sB_min)
Suppose the given array is arr[], first value if arr[].first and second value is arr[].second.
I think it is quite easy to solve this in quadratic complexity. You just need to sort the array by the first value. Then all the elements in subset A should be picked consecutively in the sorted array. So, you can loop for all ranges [L;R] of the sorted. Each range, try to add all elements in that range into subset A and add all the remains into subset B.
For more detail, this is my C++ code
int calc(pair<int, int> a[], int n){
int m = 1e9, M = -1e9, res = 2e9; //m and M are min and max of all the first values in subset A
for (int l = 1; l <= n; l++){
int g = m, G = M; //g and G are min and max of all the second values in subset B
for(int r = n; r >= l; r--) {
if (r - l + 1 < n){
res = min(res, a[r].first - a[l].first + G - g);
}
g = min(g, a[r].second);
G = max(G, a[r].second);
}
m = min(m, a[l].second);
M = max(M, a[l].second);
}
return res;
}
Now, I want to improve my algorithm down to loglinear complexity. Of course, sort the array by the first value. After that, if I fixed fA_min = a[i].first, then if the index i increase, the fA_max will increase while the (sB_max - sB_min) decrease.
But now I am still stuck here, is there any ways to solve this problem in loglinear complexity?
The following approach is an attempt to escape the n^2, using an argmin list for the second element of the tuples (lets say the y-part). Where the points are sorted regarding x.
One Observation is that there is an optimum solution where A includes index argmin[0] or argmin[n-1] or both.
in get_best_interval_min_max we focus once on including argmin[0] and the next smallest element on y and so one. The we do the same from the max element.
We get two dictionaries {(i,j):(profit, idx)}, telling us how much we gain in y when including points[i:j+1] in A, towards min or max on y. idx is the idx in the argmin array.
calculate the objective for each dict assuming max/min or y is not in A.
combine the results of both dictionaries, : (i1,j1): (v1, idx1) and (i2,j2): (v2, idx2). result : j2 - i1 + max_y - min_y - v1 - v2.
Constraint: idx1 < idx2. Because the indices in the argmin array can not intersect, otherwise some profit in y might be counted twice.
On average the dictionaries (dmin,dmax) are smaller than n, but in the worst case when x and y correlate [(i,i) for i in range(n)] they are exactly n, and we do not win any time. Anyhow on random instances this approach is much faster. Maybe someone can improve upon this.
import numpy as np
from random import randrange
import time
def get_best_interval_min_max(points):# sorted input according to x dim
L = len(points)
argmin_b = np.argsort([p[1] for p in points])
b_min,b_max = points[argmin_b[0]][1], points[argmin_b[L-1]][1]
arg = [argmin_b[0],argmin_b[0]]
res_min = dict()
for i in range(1,L):
res_min[tuple(arg)] = points[argmin_b[i]][1] - points[argmin_b[0]][1],i # the profit in b towards min
if arg[0] > argmin_b[i]: arg[0]=argmin_b[i]
elif arg[1] < argmin_b[i]: arg[1]=argmin_b[i]
arg = [argmin_b[L-1],argmin_b[L-1]]
res_max = dict()
for i in range(L-2,-1,-1):
res_max[tuple(arg)] = points[argmin_b[L-1]][1]-points[argmin_b[i]][1],i # the profit in b towards max
if arg[0]>argmin_b[i]: arg[0]=argmin_b[i]
elif arg[1]<argmin_b[i]: arg[1]=argmin_b[i]
# return the two dicts, difference along y,
return res_min, res_max, b_max-b_min
def argmin_algo(points):
# return the objective value, sets A and B, and the interval for A in points.
points.sort()
# get the profits for different intervals on the sorted array for max and min
dmin, dmax, y_diff = get_best_interval_min_max(points)
key = [None,None]
res_min = 2e9
# the best result when only the min/max b value is includes in A
for d in [dmin,dmax]:
for k,(v,i) in d.items():
res = points[k[1]][0]-points[k[0]][0] + y_diff - v
if res < res_min:
key = k
res_min = res
# combine the results for max and min.
for k1,(v1,i) in dmin.items():
for k2,(v2,j) in dmax.items():
if i > j: break # their argmin_b indices can not intersect!
idx_l, idx_h = min(k1[0], k2[0]), max(k1[1],k2[1]) # get index low and idx hight for combination
res = points[idx_h][0]-points[idx_l][0] -v1 -v2 + y_diff
if res < res_min:
key = (idx_l, idx_h) # new merged interval
res_min = res
return res_min, points[key[0]:key[1]+1], points[:key[0]]+points[key[1]+1:], key
def quadratic_algorithm(points):
points.sort()
m, M, res = 1e9, -1e9, 2e9
idx = (0,0)
for l in range(len(points)):
g, G = m, M
for r in range(len(points)-1,l-1,-1):
if r-l+1 < len(points):
res_n = points[r][0] - points[l][0] + G - g
if res_n < res:
res = res_n
idx = (l,r)
g = min(g, points[r][1])
G = max(G, points[r][1])
m = min(m, points[l][1])
M = max(M, points[l][1])
return res, points[idx[0]:idx[1]+1], points[:idx[0]]+points[idx[1]+1:], idx
# let's try it and compare running times to the quadratic_algorithm
# get some "random" points
c1=0
c2=0
for i in range(100):
points = [(randrange(100), randrange(100)) for i in range(1,200)]
points.sort() # sorted for x dimention
s = time.time()
r1 = argmin_algo(points)
e1 = time.time()
r2 = quadratic_algorithm(points)
e2 = time.time()
c1 += (e1-s)
c2 += (e2-e1)
if not r1[0] == r2[0]:
print(r1,r2)
raise Exception("Error, results are not equal")
print("time of argmin_algo", c1, "time of quadratic_algorithm",c2)
UPDATE: #Luka proved the algorithm described in this answer is not exact. But I will keep it here because it's a good performance heuristics and opens the way to many probabilistic methods.
I will describe a loglinear algorithm. I couldn't find a counter example. But I also couldn't find a proof :/
Let set A be ordered by first element and set B be ordered by second element. They are initially empty. Take floor(n/2) random points of your set of points and put in set A. Put the remaining points in set B. Define this as a partition.
Let's call a partition stable if you can't take an element of set A, put it in B and decrease the objective function and if you can't take an element of set B, put it in A and decrease the objective function. Otherwise, let's call the partition unstable.
For an unstable partition, the only moves that are interesting are the ones that take the first or the last element of A and move to B or take the first or the last element of B and move to A. So, we can find all interesting moves for a given unstable partition in O(1). If an interesting move decreases the objective function, do it. Go like that until the partition becomes stable. I conjecture that it takes at most O(n) moves for the partition to become stable. I also conjecture that at the moment the partition becomes stable, you will have a solution.
I am trying to generate a matrix, that has all unique combinations of [0 0 1 1], I wrote this code for this:
v1 = [0 0 1 1];
M1 = unique(perms([0 0 1 1]),'rows');
• This isn't ideal, because perms() is seeing each vector element as unique and doing:
4! = 4 * 3 * 2 * 1 = 24 combinations.
• With unique() I tried to delete all the repetitive entries so I end up with the combination matrix M1 →
only [4!/ 2! * (4-2)!] = 6 combinations!
Now, when I try to do something very simple like:
n = 15;
i = 1;
v1 = [zeros(1,n-i) ones(1,i)];
M = unique(perms(vec_1),'rows');
• Instead of getting [15!/ 1! * (15-1)!] = 15 combinations, the perms() function is trying to do
15! = 1.3077e+12 combinations and it's interrupted.
• How would you go about doing in a much better way? Thanks in advance!
You can use nchoosek to return the indicies which should be 1, I think in your heart you knew this must be possible because you were using the definition of nchoosek to determine the expected final number of permutations! So we can use:
idx = nchoosek( 1:N, k );
Where N is the number of elements in your array v1, and k is the number of elements which have the value 1. Then it's simply a case of creating the zeros array and populating the ones.
v1 = [0, 0, 1, 1];
N = numel(v1); % number of elements in array
k = nnz(v1); % number of non-zero elements in array
colidx = nchoosek( 1:N, k ); % column index for ones
rowidx = repmat( 1:size(colidx,1), k, 1 ).'; % row index for ones
M = zeros( size(colidx,1), N ); % create output
M( rowidx(:) + size(M,1) * (colidx(:)-1) ) = 1;
This works for both of your examples without the need for a huge intermediate matrix.
Aside: since you'd have the indicies using this approach, you could instead create a sparse matrix, but whether that's a good idea or not would depend what you're doing after this point.
I am using Python 2.7. From previous posts, I am learning Python and I have moved from arrays and now I am working on loops. I am also trying to work with operations using arrays.
A1 = np.random.random_integers(35, size=(10.,5.))
A = np.array(A1)
B1 = np.random.random_integers(68, size=(10.,5.))
B = np.array(B1)
D = np.zeros(10,5) #array has 10 rows and 5 columns filled with zeros to give me the array size I want
for j in range (1,5):
for k in range (1,5):
D[j,k] = 0
for el in range (1,10):
D[j,k] = D[j,k] + A[j] * B[k]
The error I am getting is : setting an array element with a sequence
Is my formatting incorrect?
Because A, B and D are all 2D arrays, then D[j,k]
is a single element, while A[j] (the same as A[j,:]) is a 1D array which, in this case, has 5 elements. Similar for B[k] = B[k,:], i.e. also a 5 element array.
A[j] * B[k] is therefore also five element array, which can not be stored in the place of a single element, and you therefore get the error: setting an array element with a sequence.
If you want to select single elements from A and B, then the last line should be
D[j,k] = D[j,k] + A[j,k] * B[j,k]
Some further comments on your code:
# A is already a numpy array, so 'A = np.array(A1)' is redundant and can be omitted
A = np.random.random_integers(35, size=(10.,5.))
# Same as above
B = np.random.random_integers(68, size=(10.,5.))
D = np.zeros([10,5]) # This is the correct syntax for creating a 2D array with the np.zeros() function
for j in range(1,5):
for k in range(1,5):
# D[j,k] = 0 You have already defined D to be zero for all elements with the np.zeros function, so there is no need to do it again
for el in range(1,75):
D[j,k] = D[j,k] + A[j] * B[k]
EDIT:
Well, I do not have enough reputation to comment on your post #Caroline.py, so I will do it here instead:
First of all, remember that python uses zero indexing, so 'range(1,5)' gives you '[1,2,3,4]', which means that you would not reach the first index, i.e. index 0. Thus you would probably want to use 'range(0,5)', which is the same as just 'range(5)', instead.
I can see that you changed the el range from 75 to 10. If you don't use el to anything, it just means that you add perform the last line 10 times.
I don't know what you want to do, but if you want to store the multiple of A and B in D, then this should be right:
for j in range(10):
for k in range(5):
D[j,k] = A[j,k] * B[j,k]
or just
D = A * B
pvec = 1:3;
for i = 1:3
p=pvec(i);
for m = 1:p
erfun=erfc(5/(2*sqrt(p-m)));
suma(m) = sum(erfun)
end
end
I want to save sum of all values of erfun for every p i.e I want to have 3 values in final array but every value in the array should be sum of all the values of erfun for one p.
Similar questions have been addressed but I could not apply them in my case.
Minimal fix method (to your own code)
The following modification to your code will yield your requested results
suma = zeros(3,1);
pvec = 1:3;
for i = 1:3
p=pvec(i);
for m = 1:p
erfun=erfc(5/(2*sqrt(p-m)));
suma(i) = suma(i) + erfun; %// <-- modified here
end
end
Where I've also included suma = zeros(3,1), which I assume that you also have in your code (however not shown in your question); pre-allocating suma with sufficient entries.
Alternative method (arrayfun)
Another solution, you can make use of the arrayfun command to get rid of the inner for loop:
suma = zeros(1,3);
pvec = 1:3;
for i = 1:3
p=pvec(i);
suma(i) = sum(arrayfun(#(x) erfc(5/(2*sqrt(p-x))), 1:p));
end
Alternative method #2 (arrayfun)
An even more condensed solution, including also the purpose of the outer for loop in your arrayfun call:
suma = arrayfun(#(x) ...
sum(erfc(5./(2*sqrt(kron(x, ones(1,x-1)) - 1:(x-1))))), pvec)
Here we've made use of the kron command, which will be implicitly used in the arrayfun command above as follows
kron(1, []) = [] %// empty array
kron(2, [1]) = 2
kron(3, [1 1]) = [3 3]
and used the fact that erfc addition from 1/sqrt(0) is always 0 (i.e., erfc(Inf) = 0, and hence we needn't evaluate the case m=p as it yields no addition to our sum).
Result
All of the above methods yield the result
suma =
0
0.0004
0.0128
I have the following problem. Say I have a vector:
v = [1,2,3,4,5,1,2,3,4,...]
I want to sequentially sample points from the vector, that have an absolute maginute difference higher than a threshold from a previously sampled point. So say my threshold is 2.
I start at the index 1, and sample the first point 1. Then my condition is met at v[3], and I sample 3 (since 3-1 >= 2). Then 3, the new sampled point becomes the reference, that I check against. The next sampled point is 5 which is v[5] (5-3 >= 2). Then the next point is 1 which is v[6] (abs(1-5) >= 2).
Unfortunately my code in R, is taking too long. Basically I am scanning the array repeatedly and looking for matches. I think that this approach is naive though. I have a feeling that I can accomplish this task in a single pass through the array. I dont know how though. Any help appreciated. I guess the problem I am running into is that the location of the next sample point can be anywhere in the array, and I need to scan the array from the current point to the end to find it.
Thanks.
I don't see a way this can be done without a loop, so here is one:
my.sample <- function(x, thresh) {
out <- x
i <- 1
for (j in seq_along(x)[-1]) {
if (abs(x[i]-x[j]) >= thresh) {
i <- j
} else {
out[j] <- NA
}
}
out[!is.na(out)]
}
my.sample(x = c(1:5,1:4), thresh = 2)
# [1] 1 3 5 1 3
You can do this without a loop using a bit of recursion:
vsearch = function(v, x, fun=NULL) {
# v: input vector
# x: threshold level
if (!length(v) > 0) return(NULL)
y = v-rep(v[1], times=length(v))
if (!is.null(fun)) y = fun(y)
i = which(y >= x)
if (!length(i) > 0) return(NULL)
i = i[1]
return(c(v[i], vsearch(v[-(1:(i-1))], x, fun=fun)))
}
With your vector above:
> vsearch(c(1,2,3,4,5,1,2,3,4), 2, abs)
[1] 3 5 1 3