Questions tagged [reduction]
481 questions
-1
votes
1 answer
Search Minimum/Maximum from n Arrays parallel in CUDA (Reduction Problem)
Is there a performant way in CUDA to get out of multiple arrays (which exist in different structures)
to find the maximum/minimum in parallel? The structures are structured according to the Structure of Arrays format.
A simple idea would be to…
Compty_
- 1
- 1
- 1
-1
votes
1 answer
Construct a ciruct from a graph that outputs 1 if the graph has an independent set
Let's say I have a graph,
and from this graph I want to create a circuit K whose inputs can be set so that K outputs true iff the graph has an independent set of size ≥2.
I've seen some decent stuff on the internet / youtube about how to go about…
David Dennis
- 702
- 2
- 9
- 26
-1
votes
1 answer
How do I refactor this code to make it shorter?
import math
def roundup(x):
return int(math.ceil(x / 10.0)) * 10
w=0
while w == 5:
print("Would you like to *work out* a missing letter in a GTIN-8 code, or *check* a code?")
response = input(":")
if response == 'work out':
print("Input a 7…
Sacad Muhumed
- 11
- 3
-1
votes
1 answer
comparing Matlab vs CUDA correlation and reduction on a 2D array
I am trying to compare cross-correlation using FFT vs using windowing method.
My Matlab code is:
isize = 20;
n = 7;
for i = 1:n %%7x7 xcorr
for j = 1:n
xcout(i,j) = sum(sum(ffcorr1 .* ref(i:i+isize-1,j:j+isize-1))); %%ref is 676 element array…
vivekv80
- 119
- 5
- 16
-1
votes
1 answer
How to use reduction vector variable in OpenMP?
I want to use the reduction directive in openmp, but it's not working. Compiler error says:
"reduction: OpenMP 'parallel for' empty factor in directive"
(Visual studio 2015 community)
or
"reduction:^ needed scalar variable"
This is my code…
hyun seung Kim
- 3
- 3
-1
votes
2 answers
N largest elements of a vector along with their indices
I have a thrust::device_vector vec. Assume that vec.size() = L and that N < L. I want to find the largest N elements in vec along with their indices. How can we do this efficiently using raw CUDA or thrust?
Hieu Pham
- 97
- 1
- 8
-1
votes
1 answer
OpenMP Dot Product and Pointers
I'm trying to implement dotproduct in OpenMP with large arrays allocated with malloc. However, when I use reduction(+:result) it produces different results for each program run. Why do I get different results? How can I remedy that? And how can this…
neckutrek
- 353
- 2
- 14
-1
votes
1 answer
R: Reducing data set and creating a conditional mean
So I have a large data set, with many columns(10) and 100,000 rows. One of the columns is the date of observation with two other corresponding columns, one species and the other year. First, I want to create a new column that will give me the mean…
John
- 59
- 1
- 9
-1
votes
2 answers
array_walk only partially removes matches
Ran into a weird situation were using array_walk() will only partially remove matches from my method, not certain exactly what is going on. I am currently using PHP v5.6.4. The issue almost seems to be that it is only removing every secondary…
ehime
- 8,025
- 14
- 51
- 110
-1
votes
2 answers
Can I use OpenMP reduction when the variable is an array element?
The OpenMP manual says
A type name in a declare reduction directive cannot be a function type, an array type, a reference type, or a type qualified with const, volatile or restrict.
What can I do to produce my results into array elements? I…
YOung
- 938
- 2
- 10
- 14
-2
votes
1 answer
Blockwise/Strided reduction using CUDA
TLDR: I am trying to write a GPU code that computes a blockwise reduction on an array. The input looks like [block_0, trash_0, block_1, trash_1, ..., block_n, trash_n], and I want to compute block_0 + block_1 + ... + block_n. Relevant sizes: each…
s769
- 3
- 3
-2
votes
1 answer
GPU reduction code only runs one time
I have been using the code sample supplied by Robert Crovella:
thrust::max_element slow in comparison cublasIsamax - More efficient implementation?
Which is a very fast reduction code. I modified it to also return the index of the max in the input…
brumby
- 1
- 1
-2
votes
1 answer
Cuda Reduction in 2d Array
I want to calculate the average of the values over the whole image in Cuda. To test how reduction in 2D array work, I write this kernel below. The final output o should be the sum of all the image values. The input g is a 2D array with value 1 in…
Lv Zhaoyang
- 107
- 1
- 4
- 10
-3
votes
2 answers
reducing a fraction in c
#include
#include
int main(){
system("color f0");
int a,b,c,d,e,f,g,h,z;
printf("First numerator:");
scanf("%d",&a);
printf("First denominator:");
scanf("%d",&b);
printf("Second numerator:");
…
Sanee Reimoo
- 19
- 2
-3
votes
1 answer
LU Decomposition Crout Reduction
This is my code for LU Decomposition Crout Method:
function [L, U] = croutreduction(A)
[row,column]=size(A);
L=eye(row,column);
//A = 3x3
if row==3 then
U(1,1)=A(1,1); U(1,2)=A(1,2); U(1,3)=A(1,3);
…
Elra Ghifary
- 113
- 1
- 14