spires_read_reservoir_state

Copy the reservoir’s current neuron state vector into a caller-provided buffer.

Signature

spires_status spires_read_reservoir_state(spires_reservoir *r, double *buffer);

Parameters

Returns

spires_status — SPIRES_OK on success, or SPIRES_ERR_INVALID_ARG if r or buffer is NULL.

Example

size_t N = spires_num_neurons(r);
double *buf = malloc(N * sizeof(double));
 
for (size_t t = 0; t < series_length; t++) {
    spires_step(r, &input[t * num_inputs]);
    spires_read_reservoir_state(r, buf);
 
    /* Process the state without allocation overhead */
    double norm = 0.0;
    for (size_t j = 0; j < N; j++) {
        norm += buf[j] * buf[j];
    }
    printf("t=%zu  ||x|| = %.6f\n", t, sqrt(norm));
}
 
free(buf);

Notes

• No allocation. Unlike spires_copy_reservoir_state, this function performs no heap allocation. It is suitable for use in performance-critical loops where allocation overhead or fragmentation is a concern.
• Buffer sizing. The buffer must hold at least num_neurons doubles. Use spires_num_neurons to query the required size. Writing beyond the buffer boundary is undefined behavior.
• Snapshot semantics. The data is copied out of the reservoir. Subsequent mutations to the reservoir do not affect the buffer contents.
• Thread safety. Do not call this function on a reservoir that is concurrently being mutated by another thread.

See Also

• spires_copy_reservoir_state — returns a malloc’d copy (simpler API, but allocates).
• spires_compute_output — compute the readout output.
• spires_num_neurons — query the required buffer size.