def _likelihood(self, args):
"""
routine to compute X2 given variable parameters for a MCMC/PSO chainF
"""
#generate image and computes likelihood
multi_band_list = self.update_multi_band(args)
image_model = SingleBandMultiModel(multi_band_list, self._kwargs_model,
likelihood_mask_list=self._likelihood_mask_list, band_index=self._band_index)
logL = image_model.likelihood_data_given_model(source_marg=self._source_marg, **self._kwargs_params)
return logL
class TestSingleBandMultiModel(object):
def setup(self):
# data specifics
sigma_bkg = 0.05 # background noise per pixel
exp_time = 100 # exposure time (arbitrary units, flux per pixel is in units #photons/exp_time unit)
numPix = 10 # cutout pixel size
deltaPix = 0.1 # pixel size in arcsec (area per pixel = deltaPix**2)
fwhm = 0.5 # full width half max of PSF
# PSF specification
kwargs_data = simulation_util.data_configure_simple(
numPix, deltaPix, exp_time, sigma_bkg)
data_class = ImageData(**kwargs_data)
kwargs_psf = {
'psf_type': 'GAUSSIAN',
'fwhm': fwhm,
'pixel_size': deltaPix
}
psf_class = PSF(**kwargs_psf)
kwargs_spemd = {
'theta_E': 1.,
'gamma': 1.8,
'center_x': 0,
'center_y': 0,
'e1': 0.1,
'e2': 0.1
}
lens_model_list = ['SPEP']
kwargs_lens = [kwargs_spemd]
lens_model_class = LensModel(lens_model_list=lens_model_list)
kwargs_sersic = {
'amp': 1.,
'R_sersic': 0.1,
'n_sersic': 2,
'center_x': 0,
'center_y': 0
}
# 'SERSIC_ELLIPSE': elliptical Sersic profile
kwargs_sersic_ellipse = {
'amp': 1.,
'R_sersic': .6,
'n_sersic': 3,
'center_x': 0,
'center_y': 0,
'e1': 0.1,
'e2': 0.1
}
lens_light_model_list = ['SERSIC']
kwargs_lens_light = [kwargs_sersic]
lens_light_model_class = LightModel(
light_model_list=lens_light_model_list)
source_model_list = ['SERSIC_ELLIPSE']
kwargs_source = [kwargs_sersic_ellipse]
source_model_class = LightModel(light_model_list=source_model_list)
# Point Source
point_source_model_list = ['UNLENSED']
kwargs_ps = [{
'ra_image': [0.4],
'dec_image': [-0.2],
'point_amp': [2]
}]
point_source_class = PointSource(
point_source_type_list=point_source_model_list)
kwargs_numerics = {
'supersampling_factor': 1,
'supersampling_convolution': False,
'compute_mode': 'regular'
}
imageModel = ImageModel(data_class,
psf_class,
lens_model_class,
source_model_class,
lens_light_model_class,
point_source_class=point_source_class,
kwargs_numerics=kwargs_numerics)
image_sim = simulation_util.simulate_simple(imageModel, kwargs_lens,
kwargs_source,
kwargs_lens_light,
kwargs_ps)
data_class.update_data(image_sim)
kwargs_data['image_data'] = image_sim
self.multi_band_list = [[kwargs_data, kwargs_psf, kwargs_numerics]]
self.kwargs_model = {
'lens_model_list': lens_model_list,
'source_light_model_list': source_model_list,
'lens_light_model_list': lens_light_model_list,
'point_source_model_list': point_source_model_list,
'fixed_magnification_list': [False],
'index_lens_model_list': [[0]],
'index_lens_light_model_list': [[0]],
'index_source_light_model_list': [[0]],
'index_point_source_model_list': [[0]],
}
self.kwargs_params = {
'kwargs_lens': kwargs_lens,
'kwargs_source': kwargs_source,
'kwargs_lens_light': kwargs_lens_light,
'kwargs_ps': kwargs_ps
}
self.single_band = SingleBandMultiModel(
multi_band_list=self.multi_band_list,
kwargs_model=self.kwargs_model,
linear_solver=True)
self.single_band_no_linear = SingleBandMultiModel(
multi_band_list=self.multi_band_list,
kwargs_model=self.kwargs_model,
linear_solver=False)
def test_likelihood_data_given_model(self):
logl = self.single_band.likelihood_data_given_model(
**self.kwargs_params)
logl_no_linear = self.single_band_no_linear.likelihood_data_given_model(
**self.kwargs_params)
npt.assert_almost_equal(logl / logl_no_linear, 1, decimal=4)
def test_num_param_linear(self):
num_linear = self.single_band.num_param_linear(**self.kwargs_params)
assert num_linear == 3
num_linear = self.single_band_no_linear.num_param_linear(
**self.kwargs_params)
assert num_linear == 0