def test_with_cov(self):
simdict = Simulations()
meadict = Measurements()
covdict = Covariances()
# mock measurements
arr_a = np.random.rand(1, 4*mpisize)
comm.Bcast(arr_a, root=0)
mea = Observable(arr_a, 'measured')
meadict.append(name=('test', None, 4*mpisize, None),
data=mea, otype='plain')
# mock sims
arr_b = np.random.rand(5, 4*mpisize)
sim = Observable(arr_b, 'simulated')
simdict.append(name=('test', None, 4*mpisize, None),
data=sim, otype='plain')
# mock covariance
arr_c = np.random.rand(4, 4*mpisize)
cov = Observable(arr_c, 'covariance')
covdict.append(name=('test', None, 4*mpisize, None),
cov_data=cov)
# with covariance
lh = SimpleLikelihood(meadict, covdict)
# calc by likelihood
rslt = lh(simdict) # feed variable value, not parameter value
# calc by hand
full_b = np.vstack(comm.allgather(arr_b)) # global arr_b
diff = (np.mean(full_b, axis=0) - arr_a)
full_cov = np.vstack(comm.allgather(arr_c)) # global covariance
(sign, logdet) = np.linalg.slogdet(full_cov*2.*np.pi)
baseline = -0.5*(np.vdot(diff, np.linalg.solve(full_cov, diff.T))+sign*logdet)
assert np.allclose(rslt, baseline)
def test_without_simcov(self):
simdict = Simulations()
meadict = Measurements()
covdict = Covariances()
# mock measurements
arr_a = np.random.rand(1, 4*mpisize)
comm.Bcast(arr_a, root=0)
mea = Observable(arr_a, 'measured')
meadict.append(name=('test', None, 4*mpisize, None),
data=mea, otype='plain')
# mock covariance
arr_c = np.random.rand(4, 4*mpisize)
cov = Observable(arr_c, 'covariance')
covdict.append(name=('test', None, 4*mpisize, None),
cov_data=cov)
# mock observable with repeated single realisation
arr_b = np.random.rand(1, 4*mpisize)
comm.Bcast(arr_b, root=0)
arr_ens = np.zeros((2, 4*mpisize))
for i in range(len(arr_ens)):
arr_ens[i] = arr_b
sim = Observable(arr_ens, 'simulated')
simdict.append(name=('test', None, 4*mpisize, None),
data=sim, otype='plain')
# simplelikelihood
lh_simple = SimpleLikelihood(meadict, covdict)
rslt_simple = lh_simple(simdict)
# ensemblelikelihood
lh_ensemble = EnsembleLikelihood(meadict, covdict)
rslt_ensemble = lh_ensemble(simdict)
assert rslt_ensemble == rslt_simple
def test_diag(self):
simdict = Simulations()
meadict = Measurements()
covdict = Covariances()
# mock measurements
arr_a = np.random.rand(1, 40)
mea = Observable(arr_a, 'measured')
meadict.append(name=('test', None, 40, None),
data=mea, otype='plain')
# mock (diagonal) covariance
arr_var = np.random.rand(40)
cov = Observable(np.diag(arr_var), 'covariance')
covdict.append(name=('test', None, 40, None),
cov_data=cov)
# mock observable
arr_ens = np.random.rand(10, 40)
sim = Observable(arr_ens, 'simulated')
simdict.append(name=('test', None, 40, None),
data=sim, otype='plain')
# ensemblelikelihood + diagonal_covcov
lh_ens = EnsembleLikelihood(meadict, covdict, cov_func=diagonal_mcov)
result_ens = lh_ens(simdict)
# EnsembleLikelihoodDiagonal
lh_diag = EnsembleLikelihoodDiagonal(meadict, covdict)
result_diag = lh_diag(simdict)
assert np.allclose(result_diag, result_ens)
def test_with_trace_approximation(self):
simdict = Simulations()
meadict = Measurements()
covdict = Covariances()
# mock measurements
arr_a = np.random.rand(1, 4*mpisize)
comm.Bcast(arr_a, root=0)
mea = Observable(arr_a, 'measured')
meadict.append(name=('test', None, 4*mpisize, None),
data=mea, otype='plain')
# mock covariance (NB for the trace approximation to work, the data
# covariance needs to be diagonal)
arr_c = np.diag(np.random.rand(4))
cov = Observable(arr_c, 'covariance')
covdict.append(name=('test', None, 4*mpisize, None),
cov_data=cov)
# mock observable with repeated single realisation
arr_b = np.random.rand(1, 4*mpisize)
comm.Bcast(arr_b, root=0)
arr_ens = np.zeros((2, 4*mpisize))
for i in range(len(arr_ens)):
arr_ens[i] = arr_b
sim = Observable(arr_ens, 'simulated')
simdict.append(name=('test', None, 4*mpisize, None),
data=sim, otype='plain')
# simplelikelihood
lh_simple = SimpleLikelihood(meadict, covdict)
result_simple = lh_simple(simdict)
# ensemblelikelihood
lh_ensemble = EnsembleLikelihood(meadict, covdict,
use_trace_approximation=True)
result_ensemble = lh_ensemble(simdict)
assert result_ensemble == result_simple
def test_covdict_apply_mask(self):
msk = np.random.randint(0, 2, 2 * mpisize).reshape(1, -1)
mskdict = Masks()
comm.Bcast(msk, root=0)
mskdict.append(name=('test', None, 2 * mpisize, None),
data=msk,
otype='plain')
cov = np.random.rand(2, 2 * mpisize)
covdict = Covariances()
covdict.append(name=('test', None, 2 * mpisize, None), cov_data=cov)
covdict = mskdict(covdict)
pix_num = msk.sum()
assert ('test', None, pix_num, None) in covdict.keys()
def test_covdict_append_observable(self):
cov = Observable(np.random.rand(2, 2 * mpisize), 'covariance')
covdict = Covariances()
covdict.append(name=('test', None, 2 * mpisize, None),
cov_data=cov) # plain covariance
assert np.allclose(covdict[('test', None, 2 * mpisize, None)].data,
cov.data)
cov = Observable(np.random.rand(12 * mpisize, 12 * mpisize * mpisize),
'covariance')
covdict.append(name=('test', None, mpisize, None),
cov_data=cov) # healpix covariance
assert np.allclose(covdict[('test', None, mpisize, None)].data,
cov.data)
def test_covdict_append_array(self):
cov = np.random.rand(2, 2 * mpisize)
covdict = Covariances()
covdict.append(name=('test', None, 2 * mpisize, None),
cov_data=cov) # plain covariance
assert covdict[('test', None, 2 * mpisize,
None)].shape == (2 * mpisize, 2 * mpisize)
assert np.allclose(covdict[('test', None, 2 * mpisize, None)].data,
cov)
cov = np.random.rand(12 * mpisize, 12 * mpisize * mpisize)
covdict.append(name=('test', None, mpisize, None),
cov_data=cov) # healpix covariance
assert covdict[('test', None, mpisize,
None)].shape == (12 * mpisize * mpisize,
12 * mpisize * mpisize)
assert np.allclose(covdict[('test', None, mpisize, None)].data, cov)