Function bootstrap_stderr

test/general/lm_eval/metrics.py:207–233

(f, xs, iters)

def bootstrap_stderr(f, xs, iters):
    import multiprocessing as mp

    pool = mp.Pool(mp.cpu_count())
    # this gives a biased estimate of the stderr (i.e w/ the mean, it gives something
    # equivalent to stderr calculated without Bessel's correction in the stddev.
    # Unfortunately, I haven't been able to figure out what the right correction is
    # to make the bootstrap unbiased - i considered multiplying by sqrt(n/(n-1)) but
    # that would be ad-hoc and I can't prove that that would actually be an unbiased estimator)
    # Thankfully, shouldn't matter because our samples are pretty big usually anyways
    res = []
    chunk_size = min(1000, iters)
    from tqdm import tqdm

    print("bootstrapping for stddev:", f.__name__)
    for bootstrap in tqdm(
        pool.imap(
            _bootstrap_internal(f, chunk_size),
            [(i, xs) for i in range(iters // chunk_size)],
        ),
        total=iters // chunk_size,
    ):
        # sample w replacement
        res.extend(bootstrap)

    pool.close()
    return sample_stddev(res)


def stderr_for_metric(metric, bootstrap_iters):