def log_likelihood(z, x, P, H, R):
"""
Returns log-likelihood of the measurement z given the Gaussian
posterior (x, P) using measurement function H and measurement
covariance error R
"""
S = np.dot(H, np.dot(P, H.T)) + R
return logpdf(z, np.dot(H, x), S)
def runner(chunk):
##
res = np.empty_like(chunk)
wrapper = tqdm.tqdm if verbose else (lambda x, **kwarg: x)
##
for i in wrapper(range(len(chunk))):
drv = chain[i]
drl = lprobs[i]
##
if (drv - cmean) @ cicov @ (drv - cmean) < ss.chi2.ppf(1-alpha, df=chain.shape[-1]):
res[i] = logpdf(drv, cmean, ccov) - drl
else:
res[i] = -np.inf
return res