4fb00229644a14226db9028e63e8caba9a0158f2,arviz/stats/diagnostics.py,,gelman_rubin,#Any#Any#Any#,151

Before Change

            x = trace[var].values.T
            num_samples = x.shape[1]
            // Calculate between-chain variance
            between_chain_variance = num_samples * np.var(np.mean(x, axis=1), axis=0, ddof=1)
            // Calculate within-chain variance
            within_chain_variance = np.mean(np.var(x, axis=1, ddof=1), axis=0)
            // Estimate of marginal posterior variance
            v_hat = (within_chain_variance * (num_samples - 1) / num_samples +

After Change

    r_hat = pd.Series(name="Rhat")

    for var in varnames:
        r_hat[var] = _get_rhat(trace[var].values.T, round_to)
    return r_hat


def _get_rhat(values, round_to=2):

Instances