Computing wrong result for Column Summation

[ShidehHashemian]

Hi,

I want to compute the column summation of a sparse matrix in CSR format. Here is the code snippet I wrote in cpp.

    int I = 5, J = 4;
    taco::Format B_format({taco::Dense, taco::Sparse});
    taco::Tensor<int> B({I, J}, B_format);
    B.insert({0, 1}, 1);
    B.insert({2, 0}, 2);
    B.insert({2, 2}, 5);
    B.insert({4, 2}, 3);
    B.pack();
    Format A_format({taco::Sparse});
    taco::Tensor<int> A({J}, A_format);
    taco::IndexVar i, j;
    A(j) = B(i, j);
    A.compile();
    A.assemble();
    A.compute();
    taco::write("A_taco.mtx", A);

And here is how A_taco.mtx looks like:

1  1
0  2
2  5
2  3

My operating system is Ubuntu 22.04 LTS.
I have built the version of TACO on the master branch as instructed. I didn't change the default compiler and optimization flags.

I would be grateful if you could help me with this problem.