COO format (Coordinate) in SciPy - Cheat Sheet & Quick Revision

Recall & Review

beginner

What does COO stand for in sparse matrix formats?

COO stands for Coordinate format. It stores a sparse matrix by listing the row and column coordinates of non-zero elements along with their values.

beginner

How does COO format store data internally?

COO format stores three arrays: one for row indices, one for column indices, and one for the non-zero values. Each position in these arrays corresponds to one non-zero element.

intermediate

Why is COO format useful for sparse matrices?

COO format is simple and efficient for constructing sparse matrices incrementally and for matrices with many non-zero elements scattered randomly.

beginner

How do you create a COO sparse matrix using scipy?

Use scipy.sparse.coo_matrix with three arrays: data, (row indices, column indices), and optionally the shape of the matrix.

intermediate

What is a limitation of COO format compared to CSR or CSC formats?

COO format is less efficient for arithmetic operations and matrix-vector products because it does not store data sorted by rows or columns.

Which arrays are stored in COO format for a sparse matrix?

A Only values

B Row pointers and values

C Row indices, column indices, and values

D Column pointers and values

COO format stores row indices, column indices, and the corresponding non-zero values.

What is the main advantage of COO format?

A Fast matrix multiplication

B Easy to build incrementally

C Sorted data by rows

D Uses less memory than CSR

COO format is simple and good for building sparse matrices incrementally.

Which scipy function creates a COO sparse matrix?

A scipy.sparse.coo_matrix

B scipy.sparse.csr_matrix

C scipy.sparse.dok_matrix

D scipy.sparse.lil_matrix

The function scipy.sparse.coo_matrix creates a COO format sparse matrix.

What is a drawback of COO format compared to CSR?

A Less efficient for arithmetic operations

B Cannot store zeros

C Requires more memory for dense matrices

D Does not support row indexing

COO format is less efficient for arithmetic and matrix-vector operations than CSR.

In COO format, how are the non-zero elements stored?

A Grouped by columns only

B Sorted by row and column

C Grouped by rows only

D In any order with matching row and column arrays

COO format stores non-zero elements in any order, with corresponding row and column indices.

Explain how COO format represents a sparse matrix and why it might be chosen over other formats.

Think about how you would list non-zero elements with their positions.

Describe the steps to create a COO sparse matrix using scipy and how to access its data.

Focus on the function and the arrays it needs.