-
Notifications
You must be signed in to change notification settings - Fork 195
Open
Description
I'm using the taco C++ API (built with OpenMP ON) to compute the contraction between two sparse tensors using this code:
Format csf({Sparse, Sparse, Sparse, Sparse});
Tensor<double> X = read("x.tns", csf);
Tensor<double> Y = read("y.tns", csf);
Tensor<double> Z({X.getDimension(0),X.getDimension(1),Y.getDimension(0),Y.getDimension(1)}, csf);
IndexVar i, j, k, l, m, n;
Z(i,j,m,n) = X(i,j,k,l) * Y(m,n,k,l);
Z.compile();
Z.printComputeIR(std::cout);
Z.assemble();
Z.compute();
And I found the printed code is not omp parallel in the loop. However, I test the spmv computation and it is generated with omp parallel.
Metadata
Metadata
Assignees
Labels
No labels