Return to Question

edited body

edited Jan 19, 2024 at 18:35

I am using the GraphBLAS C API (https://graphblas.org/) which provides an interface for performing mathematical operations on sparse matrices. Given an adjacency matrix $\mathbf{A}: \mathbb{R}^{n \times n}$, I would like to remove duplicate rows from the matrix. Instead of implementing this in the standard $\mathcal{O}(n^2)$$\mathcal{O}(n^3)$ way where each row must be checked against every subsequent row, I would like to try and take advantage of built-in mathematical operations from the API.

For instance in the matrix

$$ \begin{bmatrix} 1 & 0 & 0 & 1 \\ 0 & 1 & 1 & 0 \\ 1 & 0 & 0 & 1 \\ 0 & 0 & 0 & 0 \end{bmatrix} $$ either the first or third row should be deleted (which one matters not).

Does anyone know of such a mathematical operation that can take $\mathbf{A}$ and multiply/divide/<operate> with either itself or another $n \times n$ matrix (or vector) such that duplicate rows are deleted (but one of the duplicates is kept)?

If this is off-topic, let me know and I can move over to Math SE (although they will probably tell me to bug off :)

I am using the GraphBLAS C API (https://graphblas.org/) which provides an interface for performing mathematical operations on sparse matrices. Given an adjacency matrix $\mathbf{A}: \mathbb{R}^{n \times n}$, I would like to remove duplicate rows from the matrix. Instead of implementing this in the standard $\mathcal{O}(n^2)$ way where each row must be checked against every subsequent row, I would like to try and take advantage of built-in mathematical operations from the API.

For instance in the matrix

$$ \begin{bmatrix} 1 & 0 & 0 & 1 \\ 0 & 1 & 1 & 0 \\ 1 & 0 & 0 & 1 \\ 0 & 0 & 0 & 0 \end{bmatrix} $$ either the first or third row should be deleted (which one matters not).

If this is off-topic, let me know and I can move over to Math SE (although they will probably tell me to bug off :)

I am using the GraphBLAS C API (https://graphblas.org/) which provides an interface for performing mathematical operations on sparse matrices. Given an adjacency matrix $\mathbf{A}: \mathbb{R}^{n \times n}$, I would like to remove duplicate rows from the matrix. Instead of implementing this in the standard $\mathcal{O}(n^3)$ way where each row must be checked against every subsequent row, I would like to try and take advantage of built-in mathematical operations from the API.

For instance in the matrix

$$ \begin{bmatrix} 1 & 0 & 0 & 1 \\ 0 & 1 & 1 & 0 \\ 1 & 0 & 0 & 1 \\ 0 & 0 & 0 & 0 \end{bmatrix} $$ either the first or third row should be deleted (which one matters not).

If this is off-topic, let me know and I can move over to Math SE (although they will probably tell me to bug off :)

Source Link

asked Jan 19, 2024 at 18:17

codeing_monkey

Mathematical operation for removing duplicate rows in a matrix

I am using the GraphBLAS C API (https://graphblas.org/) which provides an interface for performing mathematical operations on sparse matrices. Given an adjacency matrix $\mathbf{A}: \mathbb{R}^{n \times n}$, I would like to remove duplicate rows from the matrix. Instead of implementing this in the standard $\mathcal{O}(n^2)$ way where each row must be checked against every subsequent row, I would like to try and take advantage of built-in mathematical operations from the API.

For instance in the matrix

$$ \begin{bmatrix} 1 & 0 & 0 & 1 \\ 0 & 1 & 1 & 0 \\ 1 & 0 & 0 & 1 \\ 0 & 0 & 0 & 0 \end{bmatrix} $$ either the first or third row should be deleted (which one matters not).

If this is off-topic, let me know and I can move over to Math SE (although they will probably tell me to bug off :)