brainevent.binary_csrmv_p

brainevent.binary_csrmv_p = <brainevent.XLACustomKernel object>

Low-level XLA custom-kernel primitive for binary_csrmv.

This XLACustomKernel instance dispatches the binary (event-driven) CSR sparse matrix-vector multiplication operation to registered backends (numba, pallas, cuda_raw), using runtime shape/dtype metadata provided by the high-level wrapper.

Only entries of v that are True (boolean) or positive (float) are considered active events and contribute to the output, enabling efficient event-driven sparse-dense products commonly used in spiking neural networks.

Beyond backend dispatch, the primitive stores JAX transformation bindings (JVP, transpose, batching, and call registration) so the operation integrates correctly with jit, vmap, and autodiff.

Available backends can be queried with binary_csrmv_p.available_backends(platform), and the default backend can be configured with binary_csrmv_p.set_default(platform, backend).

See also

binary_csrmv

High-level user-facing function wrapper.