How to use getdist - 10 common examples
To help you get started, we’ve selected a few getdist examples, based on popular ways it is used in public projects.
Secure your code as it's written. Use Snyk Code to scan source code in minutes - no build needed - and fix issues immediately.
cmbant / getdist / getdist / mcsamples.py View on Github 
else:
smooth_1D = smooth_scale_1D * width / fine_width
if smooth_1D < 2:
logging.warning('fine_bins not large enough to well sample smoothing scale - ' + par.name)
smooth_1D = min(max(1., smooth_1D), fine_bins // 2)
logging.debug("%s 1D sigma_range, std: %s, %s; smooth_1D_bins: %s ", par.name, par.sigma_range, par.err,
smooth_1D)
winw = min(int(round(2.5 * smooth_1D)), fine_bins // 2 - 2)
kernel = Kernel1D(winw, smooth_1D)
cache = {}
conv = convolve1D(bins, kernel.Win, 'same', cache=cache)
fine_x = np.linspace(binmin, binmax, fine_bins)
density1D = Density1D(fine_x, P=conv, view_ranges=[par.range_min, par.range_max])
if meanlikes: rawbins = conv.copy()
if par.has_limits and boundary_correction_order >= 0:
# correct for cuts allowing for normalization over window
prior_mask = np.ones(fine_bins + 2 * winw)
if par.has_limits_bot:
prior_mask[winw] = 0.5
prior_mask[: winw] = 0
if par.has_limits_top:
prior_mask[-(winw + 1)] = 0.5
prior_mask[-winw:] = 0
a0 = convolve1D(prior_mask, kernel.Win, 'valid', cache=cache)
ix = np.nonzero(a0 * density1D.P)def thin_indices(self, factor, weights=None):
"""
Indices to make single weight 1 samples. Assumes integer weights.
:param factor: The factor to thin by, should be int.
:param weights: The weights to thin, None if this should use the weights stored in the object.
:return: array of indices of samples to keep
"""
if weights is None: weights = self.weights
numrows = len(weights)
norm1 = np.sum(weights)
weights = weights.astype(np.int)
norm = np.sum(weights)
if abs(norm - norm1) > 1e-4:
raise WeightedSampleError('Can only thin with integer weights')
if factor != int(factor):
raise WeightedSampleError('Thin factor must be integer')
factor = int(factor)
if factor >= np.max(weights):
cumsum = np.cumsum(weights) // factor
# noinspection PyTupleAssignmentBalance
_, thin_ix = np.unique(cumsum, return_index=True)
else:
tot = 0
i = 0
thin_ix = np.empty(norm // factor, dtype=np.int)
ix = 0
mult = weights[i]
while i < numrows:
if mult + tot < factor:
tot += multdef __init__(self):
self.gauss = Gaussian2D([0, 0], (0.7, 1, 0.3), label='Gaussian')
self.bending = Mixture2D([[0, 0], [2, 1.8]], [(np.sqrt(0.5), 1, 0.9), (1, 1, 0.8)], [0.6, 0.4], xmin=-1, label='bending')
self.hammer = Mixture2D([[0, 0], [1, 1.8]], [(np.sqrt(0.5), 1, 0.9), (0.3, 1, -0.7)], [0.5, 0.5], label='hammer')
cov = make_2D_Cov(np.sqrt(0.5), 1, 0.1)
self.skew = Mixture2D([[0, 0], [0, 1.2]], [cov, cov / 4], [0.5, 0.5], label='skew')
cov = make_2D_Cov(np.sqrt(0.5), 1, 0.1)
self.broadtail = Mixture2D([[0, 0], [0, 0.2]], [cov, cov * 8], [0.9, 0.1], label='broad tail')
self.tensorlike = Mixture2D([[0, 0.03], [0, 0.03]], [(0.03, 0.03, 0.1), (0.03, 0.06, 0.1)], [0.85, 0.15], ymin=0, label='tensor like')
self.rotating = Mixture2D([[0, 0], [0, 0.2]], [(1, 1, 0.5), (2, 2, -0.5)], [0.6, 0.4], label='rotating')
self.tight = Mixture2D([[0, 0], [2.5, 3.5]], [(1, 1, 0.99), (1, 1.5, 0.98)], [0.6, 0.4], label='tight')
self.cut_correlated = Gaussian2D([0, 0], (0.7, 1, 0.95), ymin=0.3, xmax=1.2, label='cut correlated')
self.shape_set = [self.gauss, self.bending, self.hammer, self.skew, self.broadtail, self.rotating, self.tight,def __init__(self):
self.gauss = Gaussian2D([0, 0], (0.7, 1, 0.3), label='Gaussian')
self.bending = Mixture2D([[0, 0], [2, 1.8]], [(np.sqrt(0.5), 1, 0.9), (1, 1, 0.8)], [0.6, 0.4], xmin=-1, label='bending')
self.hammer = Mixture2D([[0, 0], [1, 1.8]], [(np.sqrt(0.5), 1, 0.9), (0.3, 1, -0.7)], [0.5, 0.5], label='hammer')
cov = make_2D_Cov(np.sqrt(0.5), 1, 0.1)
self.skew = Mixture2D([[0, 0], [0, 1.2]], [cov, cov / 4], [0.5, 0.5], label='skew')
cov = make_2D_Cov(np.sqrt(0.5), 1, 0.1)
self.broadtail = Mixture2D([[0, 0], [0, 0.2]], [cov, cov * 8], [0.9, 0.1], label='broad tail')
self.tensorlike = Mixture2D([[0, 0.03], [0, 0.03]], [(0.03, 0.03, 0.1), (0.03, 0.06, 0.1)], [0.85, 0.15], ymin=0, label='tensor like')
self.rotating = Mixture2D([[0, 0], [0, 0.2]], [(1, 1, 0.5), (2, 2, -0.5)], [0.6, 0.4], label='rotating')
self.tight = Mixture2D([[0, 0], [2.5, 3.5]], [(1, 1, 0.99), (1, 1.5, 0.98)], [0.6, 0.4], label='tight')
self.cut_correlated = Gaussian2D([0, 0], (0.7, 1, 0.95), ymin=0.3, xmax=1.2, label='cut correlated')
self.shape_set = [self.gauss, self.bending, self.hammer, self.skew, self.broadtail, self.rotating, self.tight,
self.cut_correlated, self.tensorlike]
self.cut_gaussians = self.cutGaussians((0.7, 1, 0.3))
# these examples are from Wand and Jones 93
self.bimodal = []
self.bimodal.append(Mixture2D([[-1, 0], [1, 0]], [(2. / 3, 2. / 3, 0), (2. / 3, 2. / 3, 0)], label='bimodal WJ1'))
self.bimodal.append(Mixture2D([[-3. / 2, 0], [3. / 2, 0]], [(1. / 4, 1, 0), (1. / 4, 1, 0)], label='bimodal WJ2'))# these examples are from Wand and Jones 93
self.bimodal = []
self.bimodal.append(
Mixture2D([[-1, 0], [1, 0]], [(2. / 3, 2. / 3, 0), (2. / 3, 2. / 3, 0)], label='bimodal WJ1'))
self.bimodal.append(
Mixture2D([[-3. / 2, 0], [3. / 2, 0]], [(1. / 4, 1, 0), (1. / 4, 1, 0)], label='bimodal WJ2'))
self.bimodal.append(
Mixture2D([[-1, 1], [1, -1]], [(2. / 3, 2. / 3, 3. / 5), (2. / 3, 2. / 3, 3. / 5)], label='bimodal WJ3'))
self.bimodal.append(
Mixture2D([[1, -1], [-1, 1]], [(2. / 3, 2. / 3, 7. / 10), (2. / 3, 2. / 3, 0)], label='bimodal WJ4'))
self.trimodal = []
self.trimodal.append(Mixture2D([[-6. / 5, 6. / 5], [6. / 5, -6. / 5], [0, 0]],
[(3. / 5, 3. / 5, 3. / 10), (3. / 5, 3. / 5, -3. / 5), (0.25, 0.25, 0.2)],
weights=[9, 9, 2], label='trimodal WJ1'))
self.trimodal.append(Mixture2D([[-6. / 5, 0], [6. / 5, 0], [0, 0]],
[(3. / 5, 3. / 5, 0.7), (3. / 5, 3. / 5, 0.7), (0.25, 0.25, -0.7)],
label='trimodal WJ2'))
self.trimodal.append(Mixture2D([[-1, 0], [1, 2 * np.sqrt(3) / 3], [1, -2 * np.sqrt(3) / 3]],
[(0.6, 0.7, 0.6), (0.6, 0.7, 0), (0.4, 0.7, 0)], weights=[3, 3, 1],
label='trimodal WJ3'))
self.quadrimodal = []
self.quadrimodal.append(Mixture2D([[-1, 1], [-1, -1], [1, -1], [1, 1]],
[(2. / 3, 2. / 3, 2. / 5), (2. / 3, 2. / 3, 3. / 5), (2. / 3, 2. / 3, -0.7),
(2. / 3, 2. / 3, -0.5)],
weights=[1, 3, 1, 3], label='quadrimodal'))
self.all = self.shape_set + self.bimodal + self.trimodal + self.quadrimodal + self.cut_gaussiansself.bending = Mixture2D([[0, 0], [2, 1.8]], [(np.sqrt(0.5), 1, 0.9), (1, 1, 0.8)], [0.6, 0.4], xmin=-1, label='bending')
self.hammer = Mixture2D([[0, 0], [1, 1.8]], [(np.sqrt(0.5), 1, 0.9), (0.3, 1, -0.7)], [0.5, 0.5], label='hammer')
cov = make_2D_Cov(np.sqrt(0.5), 1, 0.1)
self.skew = Mixture2D([[0, 0], [0, 1.2]], [cov, cov / 4], [0.5, 0.5], label='skew')
cov = make_2D_Cov(np.sqrt(0.5), 1, 0.1)
self.broadtail = Mixture2D([[0, 0], [0, 0.2]], [cov, cov * 8], [0.9, 0.1], label='broad tail')
self.tensorlike = Mixture2D([[0, 0.03], [0, 0.03]], [(0.03, 0.03, 0.1), (0.03, 0.06, 0.1)], [0.85, 0.15], ymin=0, label='tensor like')
self.rotating = Mixture2D([[0, 0], [0, 0.2]], [(1, 1, 0.5), (2, 2, -0.5)], [0.6, 0.4], label='rotating')
self.tight = Mixture2D([[0, 0], [2.5, 3.5]], [(1, 1, 0.99), (1, 1.5, 0.98)], [0.6, 0.4], label='tight')
self.cut_correlated = Gaussian2D([0, 0], (0.7, 1, 0.95), ymin=0.3, xmax=1.2, label='cut correlated')
self.shape_set = [self.gauss, self.bending, self.hammer, self.skew, self.broadtail, self.rotating, self.tight,
self.cut_correlated, self.tensorlike]
self.cut_gaussians = self.cutGaussians((0.7, 1, 0.3))
# these examples are from Wand and Jones 93
self.bimodal = []
self.bimodal.append(Mixture2D([[-1, 0], [1, 0]], [(2. / 3, 2. / 3, 0), (2. / 3, 2. / 3, 0)], label='bimodal WJ1'))
self.bimodal.append(Mixture2D([[-3. / 2, 0], [3. / 2, 0]], [(1. / 4, 1, 0), (1. / 4, 1, 0)], label='bimodal WJ2'))
self.bimodal.append(Mixture2D([[-1, 1], [1, -1]], [(2. / 3, 2. / 3, 3. / 5), (2. / 3, 2. / 3, 3. / 5)], label='bimodal WJ3'))
self.bimodal.append(Mixture2D([[1, -1], [-1, 1]], [(2. / 3, 2. / 3, 7. / 10), (2. / 3, 2. / 3, 0)], label='bimodal WJ4'))
self.trimodal = []def __init__(self):
self.gauss = Gaussian2D([0, 0], (0.7, 1, 0.3), label='Gaussian')
self.bending = Mixture2D([[0, 0], [2, 1.8]], [(np.sqrt(0.5), 1, 0.9), (1, 1, 0.8)], [0.6, 0.4], xmin=-1, label='bending')
self.hammer = Mixture2D([[0, 0], [1, 1.8]], [(np.sqrt(0.5), 1, 0.9), (0.3, 1, -0.7)], [0.5, 0.5], label='hammer')
cov = make_2D_Cov(np.sqrt(0.5), 1, 0.1)
self.skew = Mixture2D([[0, 0], [0, 1.2]], [cov, cov / 4], [0.5, 0.5], label='skew')
cov = make_2D_Cov(np.sqrt(0.5), 1, 0.1)
self.broadtail = Mixture2D([[0, 0], [0, 0.2]], [cov, cov * 8], [0.9, 0.1], label='broad tail')
self.tensorlike = Mixture2D([[0, 0.03], [0, 0.03]], [(0.03, 0.03, 0.1), (0.03, 0.06, 0.1)], [0.85, 0.15], ymin=0, label='tensor like')
self.rotating = Mixture2D([[0, 0], [0, 0.2]], [(1, 1, 0.5), (2, 2, -0.5)], [0.6, 0.4], label='rotating')
self.tight = Mixture2D([[0, 0], [2.5, 3.5]], [(1, 1, 0.99), (1, 1.5, 0.98)], [0.6, 0.4], label='tight')
self.cut_correlated = Gaussian2D([0, 0], (0.7, 1, 0.95), ymin=0.3, xmax=1.2, label='cut correlated')
self.shape_set = [self.gauss, self.bending, self.hammer, self.skew, self.broadtail, self.rotating, self.tight,
self.cut_correlated, self.tensorlike]
self.cut_gaussians = self.cutGaussians((0.7, 1, 0.3))self.gauss = Gaussian2D([0, 0], (0.7, 1, 0.3), label='Gaussian')
self.bending = Mixture2D([[0, 0], [2, 1.8]], [(np.sqrt(0.5), 1, 0.9), (1, 1, 0.8)], [0.6, 0.4], xmin=-1, label='bending')
self.hammer = Mixture2D([[0, 0], [1, 1.8]], [(np.sqrt(0.5), 1, 0.9), (0.3, 1, -0.7)], [0.5, 0.5], label='hammer')
cov = make_2D_Cov(np.sqrt(0.5), 1, 0.1)
self.skew = Mixture2D([[0, 0], [0, 1.2]], [cov, cov / 4], [0.5, 0.5], label='skew')
cov = make_2D_Cov(np.sqrt(0.5), 1, 0.1)
self.broadtail = Mixture2D([[0, 0], [0, 0.2]], [cov, cov * 8], [0.9, 0.1], label='broad tail')
self.tensorlike = Mixture2D([[0, 0.03], [0, 0.03]], [(0.03, 0.03, 0.1), (0.03, 0.06, 0.1)], [0.85, 0.15], ymin=0, label='tensor like')
self.rotating = Mixture2D([[0, 0], [0, 0.2]], [(1, 1, 0.5), (2, 2, -0.5)], [0.6, 0.4], label='rotating')
self.tight = Mixture2D([[0, 0], [2.5, 3.5]], [(1, 1, 0.99), (1, 1.5, 0.98)], [0.6, 0.4], label='tight')
self.cut_correlated = Gaussian2D([0, 0], (0.7, 1, 0.95), ymin=0.3, xmax=1.2, label='cut correlated')
self.shape_set = [self.gauss, self.bending, self.hammer, self.skew, self.broadtail, self.rotating, self.tight,
self.cut_correlated, self.tensorlike]
self.cut_gaussians = self.cutGaussians((0.7, 1, 0.3))
# these examples are from Wand and Jones 93
self.bimodal = []
self.bimodal.append(Mixture2D([[-1, 0], [1, 0]], [(2. / 3, 2. / 3, 0), (2. / 3, 2. / 3, 0)], label='bimodal WJ1'))
self.bimodal.append(Mixture2D([[-3. / 2, 0], [3. / 2, 0]], [(1. / 4, 1, 0), (1. / 4, 1, 0)], label='bimodal WJ2'))
self.bimodal.append(Mixture2D([[-1, 1], [1, -1]], [(2. / 3, 2. / 3, 3. / 5), (2. / 3, 2. / 3, 3. / 5)], label='bimodal WJ3'))
self.bimodal.append(Mixture2D([[1, -1], [-1, 1]], [(2. / 3, 2. / 3, 7. / 10), (2. / 3, 2. / 3, 0)], label='bimodal WJ4'))def __init__(self):
self.gauss = Gaussian2D([0, 0], (0.7, 1, 0.3), label='Gaussian')
self.bending = Mixture2D([[0, 0], [2, 1.8]], [(np.sqrt(0.5), 1, 0.9), (1, 1, 0.8)], [0.6, 0.4], xmin=-1, label='bending')
self.hammer = Mixture2D([[0, 0], [1, 1.8]], [(np.sqrt(0.5), 1, 0.9), (0.3, 1, -0.7)], [0.5, 0.5], label='hammer')
cov = make_2D_Cov(np.sqrt(0.5), 1, 0.1)
self.skew = Mixture2D([[0, 0], [0, 1.2]], [cov, cov / 4], [0.5, 0.5], label='skew')
cov = make_2D_Cov(np.sqrt(0.5), 1, 0.1)
self.broadtail = Mixture2D([[0, 0], [0, 0.2]], [cov, cov * 8], [0.9, 0.1], label='broad tail')
self.tensorlike = Mixture2D([[0, 0.03], [0, 0.03]], [(0.03, 0.03, 0.1), (0.03, 0.06, 0.1)], [0.85, 0.15], ymin=0, label='tensor like')
self.rotating = Mixture2D([[0, 0], [0, 0.2]], [(1, 1, 0.5), (2, 2, -0.5)], [0.6, 0.4], label='rotating')
self.tight = Mixture2D([[0, 0], [2.5, 3.5]], [(1, 1, 0.99), (1, 1.5, 0.98)], [0.6, 0.4], label='tight')
self.cut_correlated = Gaussian2D([0, 0], (0.7, 1, 0.95), ymin=0.3, xmax=1.2, label='cut correlated')def run_test_program(plots=['dists_2D', 'dists_1D', 'ISE_1D', 'ISE_2D'], sims=100, nsamp=default_nsamp, mbc=1, bco=1):
import time
chains.print_load_details = False
plt.rc("ytick", direction="in")
plt.rc("xtick", direction="in")
test1D = Test1DDistributions()
test2D = Test2DDistributions()
test_settings = {'mult_bias_correction_order': mbc, 'boundary_correction_order': bco,
'smooth_scale_1D': -1, 'smooth_scale_2D': -1}
g = getSubplotPlotter(subplot_size=2)
colors = ['k', 'C0', 'C1', 'C2', 'C3', 'C4']
if 'ISE_1D' in plots:
compare_method(test1D.distributions(), nx=3,
test_settings=[{'mult_bias_correction_order': 1, 'boundary_correction_order': 1},
{'mult_bias_correction_order': 2, 'boundary_correction_order': 1},
{'mult_bias_correction_order': 0, 'boundary_correction_order': 0},
{'mult_bias_correction_order': 0, 'boundary_correction_order': 1},
{'mult_bias_correction_order': 0, 'boundary_correction_order': 2},
], colors=colors, linestyles=['-', '-', ':', '-.', '--'],
fname='compare_method_1d_N%s.pdf' % nsamp,
sims=sims, nsamp=nsamp
)
if 'ISE_2D' in plots:getdist
GetDist Monte Carlo sample analysis, plotting and GUI
Package Health Score
73 / 100Popular getdist functions
- getdist.chains.WeightedSampleError
- getdist.convolve.convolve1D
- getdist.convolve.convolve2D
- getdist.densities.Density2D
- getdist.gaussian_mixtures.Gaussian2D
- getdist.gaussian_mixtures.Mixture1D
- getdist.gaussian_mixtures.Mixture2D
- getdist.IniFile
- getdist.MCSamples
- getdist.mcsamples.loadMCSamples
- getdist.paramnames.makeList
- getdist.paramnames.ParamInfo
- getdist.paramnames.ParamNames
- getdist.plots
- getdist.plots.GetDistPlotError
- getdist.plots.GetDistPlotter
- getdist.plots.getSubplotPlotter
- getdist.plots.set_active_style
- getdist.types
- getdist.types.BestFit