def _sample_pymc3(cls, dist, size, seed):
"""Sample from PyMC3."""
import pymc3
pymc3_rv_map = {
'MatrixNormalDistribution':
lambda dist: pymc3.MatrixNormal(
'X',
mu=matrix2numpy(dist.location_matrix, float),
rowcov=matrix2numpy(dist.scale_matrix_1, float),
colcov=matrix2numpy(dist.scale_matrix_2, float),
shape=dist.location_matrix.shape),
'WishartDistribution':
lambda dist: pymc3.WishartBartlett(
'X', nu=int(dist.n), S=matrix2numpy(dist.scale_matrix, float))
}
dist_list = pymc3_rv_map.keys()
if dist.__class__.__name__ not in dist_list:
return None
with pymc3.Model():
pymc3_rv_map[dist.__class__.__name__](dist)
return pymc3.sample(size, chains=1, progressbar=False)[:]['X']
def _sample_pymc3(cls, dist, size, seed):
"""Sample from PyMC3."""
import pymc3
pymc3_rv_map = {
'MatrixNormalDistribution':
lambda dist: pymc3.MatrixNormal(
'X',
mu=matrix2numpy(dist.location_matrix, float),
rowcov=matrix2numpy(dist.scale_matrix_1, float),
colcov=matrix2numpy(dist.scale_matrix_2, float),
shape=dist.location_matrix.shape),
'WishartDistribution':
lambda dist: pymc3.WishartBartlett(
'X', nu=int(dist.n), S=matrix2numpy(dist.scale_matrix, float))
}
sample_shape = {
'WishartDistribution': lambda dist: dist.scale_matrix.shape,
'MatrixNormalDistribution': lambda dist: dist.location_matrix.shape
}
dist_list = pymc3_rv_map.keys()
if dist.__class__.__name__ not in dist_list:
return None
import logging
logging.getLogger("pymc3").setLevel(logging.ERROR)
with pymc3.Model():
pymc3_rv_map[dist.__class__.__name__](dist)
samps = pymc3.sample(draws=prod(size),
chains=1,
progressbar=False,
random_seed=seed,
return_inferencedata=False,
compute_convergence_checks=False)['X']
return samps.reshape(size +
sample_shape[dist.__class__.__name__](dist))
prec = np.linalg.inv(covariance)
mean = [.5, 1, .2]
data = scipy.stats.multivariate_normal(mean, covariance).rvs(5000)
plt.scatter(data[:, 0], data[:, 1])
with pm.Model() as model:
S = np.eye(3)
nu = 5
mu = pm.Normal('mu', mu=0, sd=1, shape=3)
# Use the transformed Wishart distribution
# Under the hood this will do a Cholesky decomposition
# of S and add two RVs to the sampler: c and z
prec = pm.WishartBartlett('prec', S, nu)
# To be able to compare it to truth, convert precision to covariance
cov = pm.Deterministic('cov', tt.nlinalg.matrix_inverse(prec))
lp = pm.MvNormal('likelihood', mu=mu, tau=prec, observed=data)
start = pm.find_MAP()
step = pm.NUTS(scaling=start)
def run(n=3000):
if n == "short":
n = 50
with model:
trace = pm.sample(n, step, start)