Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

verbose feature added in likelihood #43

Merged
merged 4 commits into from
Aug 13, 2024
Merged

verbose feature added in likelihood #43

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
b0c3508
verbose feature for better diagnostics for the likelihood
sibirrer Aug 13, 2024
11bdde7
Merge remote-tracking branch 'origin/dsp_msd' into dsp_msd
sibirrer Aug 13, 2024
232c5ca
[pre-commit.ci] auto fixes from pre-commit.com hooks
pre-commit-ci[bot] Aug 13, 2024
da20372
testing improved with verbose
sibirrer Aug 13, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 11 additions & 2 deletions hierarc/Likelihood/cosmo_likelihood.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -104,17 +104,24 @@ def __init__(
z_max = kwargs_lens["z_source2"]
self._z_max = z_max

def likelihood(self, args, kwargs_cosmo_interp=None):
def likelihood(self, args, kwargs_cosmo_interp=None, verbose=False):
"""

:param args: list of sampled parameters
:param kwargs_cosmo_interp: interpolated angular diameter distances with
'ang_diameter_distances' and 'redshifts', and optionally 'ok' and 'K' in none-flat scenarios
:type kwargs_cosmo_interp: dict
:param verbose: If true, prints intermediate outputs of likelihood calculation
:type verbose: bool
:return: log likelihood of the combined lenses
"""
for i in range(0, len(args)):
if args[i] < self._lower_limit[i] or args[i] > self._upper_limit[i]:
if verbose:
print(
"Parameter %i with value %s out of range [%s, %s]"
% (i, args[i], self._lower_limit[i], self._upper_limit[i])
)
return -np.inf

kwargs_cosmo, kwargs_lens, kwargs_kin, kwargs_source, kwargs_los = (
Expand All @@ -135,6 +142,8 @@ def likelihood(self, args, kwargs_cosmo_interp=None):
return -np.inf
# make sure that Omega_DE is not negative...
if 1.0 - om - ok <= 0:
if verbose:
print("curvature unphysical with 1-Om - Ok <= 0")
return -np.inf
if kwargs_cosmo_interp is not None:
kwargs_cosmo = {**kwargs_cosmo, **kwargs_cosmo_interp}
Expand All @@ -145,6 +154,7 @@ def likelihood(self, args, kwargs_cosmo_interp=None):
kwargs_kin=kwargs_kin,
kwargs_source=kwargs_source,
kwargs_los=kwargs_los,
verbose=verbose,
)

if self._sne_evaluate is True:
Expand Down Expand Up @@ -176,7 +186,6 @@ def cosmo_instance(self, kwargs_cosmo):
:param kwargs_cosmo: cosmology parameter keyword argument list
:return: ~astropy.cosmology (or equivalent interpolation scheme class)
"""
print(kwargs_cosmo, "test kwargs_cosmo")
if "ang_diameter_distances" in kwargs_cosmo and "redshifts" in kwargs_cosmo:
# in that case we use directly the interpolation mode to approximate angular diameter distances
cosmo = CosmoInterp(
Expand Down
9 changes: 7 additions & 2 deletions hierarc/Likelihood/hierarchy_likelihood.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -175,16 +175,19 @@ def lens_log_likelihood(
kwargs_kin=None,
kwargs_source=None,
kwargs_los=None,
verbose=False,
):
"""Log likelihood of the data of a lens given a model (defined with hyper-
parameters) and cosmology.

:param cosmo: astropy.cosmology instance
:param kwargs_lens: keywords of the hyper parameters of the lens model
:param kwargs_kin: keyword arguments of the kinematic model hyper parameters
:param kwargs_lens: keywords of the hyperparameters of the lens model
:param kwargs_kin: keyword arguments of the kinematic model hyperparameters
:param kwargs_source: keyword argument of the source model (such as SNe)
:param kwargs_los: list of keyword arguments of global line of sight
distributions
:param verbose: If true, prints intermediate outputs of likelihood calculation
:type verbose: bool
:return: log likelihood of the data given the model
"""

Expand All @@ -209,6 +212,8 @@ def lens_log_likelihood(
kwargs_los=kwargs_los,
cosmo=cosmo,
)
if verbose:
print("log likelihood of lens %s = %s" % (self._name, a))
return np.nan_to_num(a)

def hyper_param_likelihood(
Expand Down
4 changes: 4 additions & 0 deletions hierarc/Likelihood/lens_sample_likelihood.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -53,6 +53,7 @@ def log_likelihood(
kwargs_kin=None,
kwargs_source=None,
kwargs_los=None,
verbose=False,
):
"""

Expand All @@ -61,6 +62,8 @@ def log_likelihood(
:param kwargs_kin: keyword arguments of the kinematic model
:param kwargs_source: keyword argument of the source model (such as SNe)
:param kwargs_los: line of sight keyword argument list
:param verbose: If true, prints intermediate outputs of likelihood calculation
:type verbose: bool
:return: log likelihood of the combined lenses
"""
log_likelihood = 0
Expand All @@ -71,6 +74,7 @@ def log_likelihood(
kwargs_kin=kwargs_kin,
kwargs_source=kwargs_source,
kwargs_los=kwargs_los,
verbose=verbose,
)
return log_likelihood

Expand Down
16 changes: 8 additions & 8 deletions test/test_Likelihood/test_cosmo_likelihood.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -99,29 +99,29 @@ def custom_prior(

kwargs_cosmo = {"h0": self.H0_true, "om": self.omega_m_true, "ok": 0}
args = cosmoL.param.kwargs2args(kwargs_cosmo=kwargs_cosmo)
logl = cosmoL.likelihood(args=args)
logl_prior = cosmoL_prior.likelihood(args=args)
logl = cosmoL.likelihood(args=args, verbose=True)
logl_prior = cosmoL_prior.likelihood(args=args, verbose=True)
npt.assert_almost_equal(logl - logl_prior, 1, decimal=8)

kwargs = {"h0": self.H0_true * 0.99, "om": self.omega_m_true, "ok": 0}
args = cosmoL.param.kwargs2args(kwargs_cosmo=kwargs)
logl_sigma = cosmoL.likelihood(args=args)
logl_sigma = cosmoL.likelihood(args=args, verbose=True)
npt.assert_almost_equal(logl - logl_sigma, 0.5, decimal=2)

kwargs = {"h0": 100, "om": 1.0, "ok": 0.1}
args = cosmoL.param.kwargs2args(kwargs)
logl = cosmoL.likelihood(args=args)
logl = cosmoL.likelihood(args=args, verbose=True)
assert logl == -np.inf

kwargs = {"h0": 100, "om": 0.1, "ok": -0.6}
args = cosmoL.param.kwargs2args(kwargs)
logl = cosmoL.likelihood(args=args)
logl = cosmoL.likelihood(args=args, verbose=True)
assert logl == -np.inf

# outside the prior limit
kwargs = {"h0": 1000, "om": 0.3, "ok": -0.1}
args = cosmoL.param.kwargs2args(kwargs)
logl = cosmoL.likelihood(args=args)
logl = cosmoL.likelihood(args=args, verbose=True)
assert logl == -np.inf

def test_cosmo_instance(self):
Expand Down Expand Up @@ -232,7 +232,7 @@ def test_sne_likelihood_integration(self):
)
kwargs_cosmo = {"h0": self.H0_true, "om": self.omega_m_true, "ok": 0}
args = cosmoL.param.kwargs2args(kwargs_cosmo=kwargs_cosmo)
logl = cosmoL.likelihood(args=args)
logl = cosmoL.likelihood(args=args, verbose=True)
assert logl < 0

def test_kde_likelihood_integration(self):
Expand All @@ -253,7 +253,7 @@ def test_kde_likelihood_integration(self):
)
kwargs_cosmo = {"h0": self.H0_true, "om": self.omega_m_true, "ok": 0}
args = cosmoL.param.kwargs2args(kwargs_cosmo=kwargs_cosmo)
logl = cosmoL.likelihood(args=args)
logl = cosmoL.likelihood(args=args, verbose=True)
assert logl < 0


Expand Down
Loading