def fixture_model_with_plates(fixture_distribution_parameters,
fixture_pm_model_decorate):
batch_shape, observed = fixture_distribution_parameters
expected_obs_shape = (() if isinstance(observed, float) else
observed.shape[:len(observed.shape) -
len(batch_shape)])
if fixture_pm_model_decorate:
expected_rv_shapes = {
"model/loc": (),
"model/obs": expected_obs_shape,
}
else:
expected_rv_shapes = {"loc": (), "obs": expected_obs_shape}
def model():
loc = yield pm.Normal("loc", 0, 1)
obs = yield pm.Normal("obs",
loc,
1,
plate=batch_shape,
observed=observed)
return obs
if fixture_pm_model_decorate:
model = pm.model(model)
return model, expected_rv_shapes
def mc3_approach(T, width, height, N):
shape = (height, width)
x0 = np.random.randint(2, size=shape)
with pm.model() as model:
x = pm.Bernoulli('x', 0.5, shape=shape, testval=x0)
magnetization = pm.Potential(
'm',
-get_H(to_spins(x)) / T
)
scaling = .0006
mul = int(height * width * 1.75)
step = pm.BinaryMetropolis([x], scaling=scaling)
trace = pm.sample(N * mul * 5, step=step, chains=1, tune=False)
dataset = [to_two_color(2 * t['x'] - 1) for t in trace[::mul * 5]]
# Print out the final percent magnetization
lattice = 2 * trace[-1]['x'] - 1
print('Finished. Net magnetization: {:3.0%}'
.format(abs(lattice.sum()) / lattice.size))
return dataset