brainevent.csr_to_coo_index

brainevent.csr_to_coo_index(indptr, indices)

Convert CSR format index arrays to COO format index arrays.

Transforms the Compressed Sparse Row representation of a sparse matrix (given by indptr and indices) into the Coordinate representation, which uses explicit row and column index arrays for each non-zero element.

Parameters:

• indptr (Array | ndarray) – Row pointer array in CSR format. For a matrix with m rows, this has length m + 1. Element indptr[i] gives the index into indices where row i starts, and indptr[i+1] - indptr[i] is the number of non-zero entries in row i.

• indices (Array | ndarray) – Column index array in CSR format. Contains the column index for each non-zero element. Length equals the number of stored elements.

Returns:

• pre_ids (jax.Array or numpy.ndarray) – Row indices in COO format, with the same length as indices.

• post_ids (jax.Array or numpy.ndarray) – Column indices in COO format (identical to the input indices).

See also

coo_to_csc_index

Convert COO indices to CSC format.

csr_to_csc_index

Convert CSR indices directly to CSC format.

Notes

The function automatically selects NumPy or JAX operations based on the type of the input arrays. When JAX arrays are provided, the computation is wrapped in jax.ensure_compile_time_eval() so that it runs at trace time.

Examples

>>> import numpy as np
>>> from brainevent._misc import csr_to_coo_index
>>> indptr = np.array([0, 2, 3, 5])
>>> indices = np.array([0, 2, 1, 0, 3])
>>> row_ids, col_ids = csr_to_coo_index(indptr, indices)
>>> print(row_ids)
[0 0 1 2 2]
>>> print(col_ids)
[0 2 1 0 3]