def print_fisher_info():
"""Prints Fisher information for logging purposes"""
fisher = FisherInformation(data_file, include_nuisance_parameters=False)
fisher_obs = fisher_info["observable"]
fisher_bins = fisher_info["histogram_bins"]
fisher_range = fisher_info["histogram_range"]
fisher_theta = fisher_info["theta_true"]
info, _ = fisher.full_information(
theta=fisher_theta,
luminosity=luminosity,
model_file=f"{model_dir}/{gen_method}/{gen_method}",
)
logger.info(f"Fisher information in rates:")
logger.info(f"{info}")
info_histo, _ = fisher.histo_information(
theta=fisher_theta,
luminosity=luminosity,
observable=fisher_obs,
bins=fisher_bins,
histrange=fisher_range,
)
logger.info(f"Fisher information in 1D histogram:")
logger.info(f"{info_histo}")
def print_fisher_info():
""" Prints Fisher information for logging purposes """
fisher = FisherInformation(data_file, include_nuisance_parameters=False)
fisher_obs = fisher_info['observable']
fisher_bins = fisher_info['histogram_bins']
fisher_range = fisher_info['histogram_range']
fisher_theta = fisher_info['theta_true']
info, _ = fisher.full_information(
theta=fisher_theta,
luminosity=luminosity,
model_file=f'{model_dir}/{gen_method}/{gen_method}',
)
logger.info(f'Fisher information in rates:')
logger.info(f'\n{info}')
info_histo, _ = fisher.histo_information(
theta=fisher_theta,
luminosity=luminosity,
observable=fisher_obs,
bins=fisher_bins,
histrange=fisher_range,
)
logger.info(f'Fisher information in 1D histogram:')
logger.info(f'\n{info_histo}')
def get_ratio_estimator(estimator_name: str):
"""
Returns the desired ratio estimator class defined by its string name
:param estimator_name: name to identify the desired estimator
:return: ConditionalEstimator
"""
try:
estimator_cls = ratio_estimators[estimator_name]
except KeyError:
logger.info(
f"Invalid name: {estimator_name}. Defaulting to Parameterized")
estimator_cls = ParameterizedRatioEstimator
return estimator_cls
def parse_theta_ranges():
"""
Parses the minimum and maximum values for thetas
in order to build a theta ranges.
:return: list
"""
ranges = []
for asymp_theta in asymp_info["region"].keys():
ranges.append((
asymp_info["region"][asymp_theta]["min"],
asymp_info["region"][asymp_theta]["max"],
))
logger.info(f"Theta ranges: {ranges}")
return ranges
def parse_theta_ranges():
"""
Parses the minimum and maximum values for thetas
in order to build a theta ranges.
:return: list
"""
ranges = []
for asymp_theta in asymp_info['region'].keys():
ranges.append((
asymp_info['region'][asymp_theta]['min'],
asymp_info['region'][asymp_theta]['max'],
))
logger.info(f'Theta ranges: {ranges}')
return ranges
#############################
sampler = SampleAugmenter(data_file, include_nuisance_parameters=nuisance)
#############################
## Create training samples ##
#############################
# Different methods have different arguments
train_ratio_methods = {'alice', 'alices', 'cascal', 'carl', 'rolr', 'rascal'}
train_local_methods = {'sally', 'sallino'}
train_global_methods = {'scandal'}
# Iterate through the methods
for method in methods:
logger.info(f'Sampling from method: {method}')
training_params = inputs[method]
for i in range(n_sampling_runs):
if method in train_ratio_methods:
theta_0_spec = training_params['theta_0']
theta_1_spec = training_params['theta_1']
theta_0_vals = get_theta_values(theta_0_spec)
theta_1_vals = get_theta_values(theta_1_spec)
sampler.sample_train_ratio(
theta0=theta_0_vals,
theta1=theta_1_vals,
n_samples=n_samples_train,
folder=f'{data_dir}/Samples_{method}_{i}',
sampler = SampleAugmenter(data_file, include_nuisance_parameters=nuisance)
#############################
## Create training samples ##
#############################
# Different methods have different arguments
train_ratio_methods = {"alice", "alices", "cascal", "carl", "rolr", "rascal"}
train_local_methods = {"sally", "sallino"}
train_global_methods = {"scandal"}
# Iterate through the methods
for method in methods:
logger.info(f"Sampling from method: {method}")
training_params = inputs[method]
for i in range(n_sampling_runs):
if method in train_ratio_methods:
theta_0_spec = training_params["theta_0"]
theta_1_spec = training_params["theta_1"]
theta_0_vals = get_theta_values(theta_0_spec)
theta_1_vals = get_theta_values(theta_1_spec)
sampler.sample_train_ratio(
theta0=theta_0_vals,
theta1=theta_1_vals,
n_samples=n_samples_train,
folder=f"{data_dir}/Samples_{method}_{i}",