def test_empty_population(db_path, sampler):
def make_model(theta):
def model(args):
return {"result": 1 if random.random() > theta else 0}
return model
theta1 = .2
theta2 = .6
model1 = make_model(theta1)
model2 = make_model(theta2)
models = [model1, model2]
models = list(map(SimpleModel, models))
population_size = ConstantPopulationSize(1500)
parameter_given_model_prior_distribution = [Distribution(), Distribution()]
abc = ABCSMC(models, parameter_given_model_prior_distribution,
MinMaxDistanceFunction(measures_to_use=["result"]),
population_size,
eps=MedianEpsilon(0),
sampler=sampler)
abc.new(db_path, {"result": 0})
minimum_epsilon = -1
history = abc.run(minimum_epsilon, max_nr_populations=3)
mp = history.get_model_probabilities(history.max_t)
expected_p1, expected_p2 = theta1 / (theta1 + theta2), theta2 / (theta1 +
theta2)
assert abs(mp.p[0] - expected_p1) + abs(mp.p[1] - expected_p2) < .05
def test_two_competing_gaussians_multiple_population_adaptive_populatin_size(db_path, sampler):
# Define a gaussian model
sigma = .5
def model(args):
return {"y": st.norm(args['x'], sigma).rvs()}
# We define two models, but they are identical so far
models = [model, model]
models = list(map(SimpleModel, models))
# The prior over the model classes is uniform
model_prior = RV("randint", 0, 2)
# However, our models' priors are not the same. Their mean differs.
mu_x_1, mu_x_2 = 0, 1
parameter_given_model_prior_distribution = [Distribution(x=st.norm(mu_x_1, sigma)),
Distribution(x=st.norm(mu_x_2, sigma))]
# Particles are perturbed in a Gaussian fashion
parameter_perturbation_kernels = [MultivariateNormalTransition() for _ in range(2)]
# We plug all the ABC setup together
nr_populations = 3
population_size = AdaptivePopulationSize(400, mean_cv=0.05,
max_population_size=1000)
abc = ABCSMC(models, parameter_given_model_prior_distribution,
MinMaxDistanceFunction(measures_to_use=["y"]),
population_size,
model_prior=model_prior,
eps=MedianEpsilon(.2),
sampler=sampler)
# Finally we add meta data such as model names and define where to store the results
# y_observed is the important piece here: our actual observation.
y_observed = 1
abc.new(db_path, {"y": y_observed})
# We run the ABC with 3 populations max
minimum_epsilon = .05
history = abc.run(minimum_epsilon, max_nr_populations=3)
# Evaluate the model probabililties
mp = history.get_model_probabilities(history.max_t)
def p_y_given_model(mu_x_model):
return st.norm(mu_x_model, sp.sqrt(sigma ** 2 + sigma ** 2)).pdf(y_observed)
p1_expected_unnormalized = p_y_given_model(mu_x_1)
p2_expected_unnormalized = p_y_given_model(mu_x_2)
p1_expected = p1_expected_unnormalized / (p1_expected_unnormalized + p2_expected_unnormalized)
p2_expected = p2_expected_unnormalized / (p1_expected_unnormalized + p2_expected_unnormalized)
assert history.max_t == nr_populations-1
assert abs(mp.p[0] - p1_expected) + abs(mp.p[1] - p2_expected) < .07
def test_gaussian_multiple_populations_crossval_kde(db_path, sampler):
sigma_x = 1
sigma_y = .5
y_observed = 2
def model(args):
return {"y": st.norm(args['x'], sigma_y).rvs()}
models = [model]
models = list(map(SimpleModel, models))
nr_populations = 4
population_size = ConstantPopulationSize(600)
parameter_given_model_prior_distribution = [
Distribution(x=st.norm(0, sigma_x))
]
parameter_perturbation_kernels = [
GridSearchCV(MultivariateNormalTransition(),
{"scaling": sp.logspace(-1, 1.5, 5)})
]
abc = ABCSMC(models,
parameter_given_model_prior_distribution,
MinMaxDistanceFunction(measures_to_use=["y"]),
population_size,
transitions=parameter_perturbation_kernels,
eps=MedianEpsilon(.2),
sampler=sampler)
abc.new(db_path, {"y": y_observed})
minimum_epsilon = -1
abc.do_not_stop_when_only_single_model_alive()
history = abc.run(minimum_epsilon, max_nr_populations=nr_populations)
posterior_x, posterior_weight = history.get_distribution(0, None)
posterior_x = posterior_x["x"].as_matrix()
sort_indices = sp.argsort(posterior_x)
f_empirical = sp.interpolate.interp1d(
sp.hstack((-200, posterior_x[sort_indices], 200)),
sp.hstack((0, sp.cumsum(posterior_weight[sort_indices]), 1)))
sigma_x_given_y = 1 / sp.sqrt(1 / sigma_x**2 + 1 / sigma_y**2)
mu_x_given_y = sigma_x_given_y**2 * y_observed / sigma_y**2
expected_posterior_x = st.norm(mu_x_given_y, sigma_x_given_y)
x = sp.linspace(-8, 8)
max_distribution_difference = sp.absolute(
f_empirical(x) - expected_posterior_x.cdf(x)).max()
assert max_distribution_difference < 0.052
assert history.max_t == nr_populations - 1
mean_emp, std_emp = mean_and_std(posterior_x, posterior_weight)
assert abs(mean_emp - mu_x_given_y) < .07
assert abs(std_emp - sigma_x_given_y) < .12
def test_all_in_one_model(db_path, sampler):
models = [AllInOneModel() for _ in range(2)]
population_size = ConstantPopulationSize(800)
parameter_given_model_prior_distribution = [Distribution(theta=RV("beta",
1, 1))
for _ in range(2)]
abc = ABCSMC(models, parameter_given_model_prior_distribution,
MinMaxDistanceFunction(measures_to_use=["result"]),
population_size,
eps=MedianEpsilon(.1),
sampler=sampler)
abc.new(db_path, {"result": 2})
minimum_epsilon = .2
history = abc.run(minimum_epsilon, max_nr_populations=3)
mp = history.get_model_probabilities(history.max_t)
assert abs(mp.p[0] - .5) + abs(mp.p[1] - .5) < .08
def test_gaussian_single_population(db_path, sampler):
sigma_prior = 1
sigma_ground_truth = 1
observed_data = 1
def model(args):
return {"y": st.norm(args['x'], sigma_ground_truth).rvs()}
models = [model]
models = list(map(SimpleModel, models))
nr_populations = 1
population_size = ConstantPopulationSize(600)
parameter_given_model_prior_distribution = [
Distribution(x=RV("norm", 0, sigma_prior))
]
abc = ABCSMC(models,
parameter_given_model_prior_distribution,
MinMaxDistanceFunction(measures_to_use=["y"]),
population_size,
eps=MedianEpsilon(.1),
sampler=sampler)
abc.new(db_path, {"y": observed_data})
minimum_epsilon = -1
abc.do_not_stop_when_only_single_model_alive()
history = abc.run(minimum_epsilon, max_nr_populations=nr_populations)
posterior_x, posterior_weight = history.get_distribution(0, None)
posterior_x = posterior_x["x"].as_matrix()
sort_indices = sp.argsort(posterior_x)
f_empirical = sp.interpolate.interp1d(
sp.hstack((-200, posterior_x[sort_indices], 200)),
sp.hstack((0, sp.cumsum(posterior_weight[sort_indices]), 1)))
sigma_x_given_y = 1 / sp.sqrt(1 / sigma_prior**2 +
1 / sigma_ground_truth**2)
mu_x_given_y = sigma_x_given_y**2 * observed_data / sigma_ground_truth**2
expected_posterior_x = st.norm(mu_x_given_y, sigma_x_given_y)
x = sp.linspace(-8, 8)
max_distribution_difference = sp.absolute(
f_empirical(x) - expected_posterior_x.cdf(x)).max()
assert max_distribution_difference < 0.12
assert history.max_t == nr_populations - 1
mean_emp, std_emp = mean_and_std(posterior_x, posterior_weight)
assert abs(mean_emp - mu_x_given_y) < .07
assert abs(std_emp - sigma_x_given_y) < .1
def test_two_competing_gaussians_single_population(db_path, sampler,
transition):
sigma_x = .5
sigma_y = .5
y_observed = 1
def model(args):
return {"y": st.norm(args['x'], sigma_y).rvs()}
models = [model, model]
models = list(map(SimpleModel, models))
population_size = ConstantPopulationSize(500)
mu_x_1, mu_x_2 = 0, 1
parameter_given_model_prior_distribution = [
Distribution(x=st.norm(mu_x_1, sigma_x)),
Distribution(x=st.norm(mu_x_2, sigma_x))
]
abc = ABCSMC(models,
parameter_given_model_prior_distribution,
MinMaxDistanceFunction(measures_to_use=["y"]),
population_size,
transitions=[transition(), transition()],
eps=MedianEpsilon(.02),
sampler=sampler)
abc.new(db_path, {"y": y_observed})
minimum_epsilon = -1
nr_populations = 1
abc.do_not_stop_when_only_single_model_alive()
history = abc.run(minimum_epsilon, max_nr_populations=1)
mp = history.get_model_probabilities(history.max_t)
def p_y_given_model(mu_x_model):
return st.norm(mu_x_model,
sp.sqrt(sigma_y**2 + sigma_x**2)).pdf(y_observed)
p1_expected_unnormalized = p_y_given_model(mu_x_1)
p2_expected_unnormalized = p_y_given_model(mu_x_2)
p1_expected = p1_expected_unnormalized / (p1_expected_unnormalized +
p2_expected_unnormalized)
p2_expected = p2_expected_unnormalized / (p1_expected_unnormalized +
p2_expected_unnormalized)
assert history.max_t == nr_populations - 1
assert abs(mp.p[0] - p1_expected) + abs(mp.p[1] - p2_expected) < .07
def test_beta_binomial_different_priors_initial_epsilon_from_sample(
db_path, sampler):
binomial_n = 5
def model(args):
return {"result": st.binom(binomial_n, args.theta).rvs()}
models = [model for _ in range(2)]
models = list(map(SimpleModel, models))
population_size = ConstantPopulationSize(800)
a1, b1 = 1, 1
a2, b2 = 10, 1
parameter_given_model_prior_distribution = [
Distribution(theta=RV("beta", a1, b1)),
Distribution(theta=RV("beta", a2, b2))
]
abc = ABCSMC(models,
parameter_given_model_prior_distribution,
MinMaxDistanceFunction(measures_to_use=["result"]),
population_size,
eps=MedianEpsilon(median_multiplier=.9),
sampler=sampler)
n1 = 2
abc.new(db_path, {"result": n1})
minimum_epsilon = -1
history = abc.run(minimum_epsilon, max_nr_populations=5)
mp = history.get_model_probabilities(history.max_t)
def B(a, b):
return gamma(a) * gamma(b) / gamma(a + b)
def expected_p(a, b, n1):
return binom(binomial_n, n1) * B(a + n1, b + binomial_n - n1) / B(a, b)
p1_expected_unnormalized = expected_p(a1, b1, n1)
p2_expected_unnormalized = expected_p(a2, b2, n1)
p1_expected = p1_expected_unnormalized / (p1_expected_unnormalized +
p2_expected_unnormalized)
p2_expected = p2_expected_unnormalized / (p1_expected_unnormalized +
p2_expected_unnormalized)
assert abs(mp.p[0] - p1_expected) + abs(mp.p[1] - p2_expected) < .08
def test_continuous_non_gaussian(db_path, sampler):
def model(args):
return {"result": sp.rand() * args['u']}
models = [model]
models = list(map(SimpleModel, models))
population_size = ConstantPopulationSize(250)
parameter_given_model_prior_distribution = [Distribution(u=RV("uniform", 0,
1))]
abc = ABCSMC(models, parameter_given_model_prior_distribution,
MinMaxDistanceFunction(measures_to_use=["result"]),
population_size,
eps=MedianEpsilon(.2),
sampler=sampler)
d_observed = .5
abc.new(db_path, {"result": d_observed})
abc.do_not_stop_when_only_single_model_alive()
minimum_epsilon = -1
history = abc.run(minimum_epsilon, max_nr_populations=2)
posterior_x, posterior_weight = history.get_distribution(0, None)
posterior_x = posterior_x["u"].values
sort_indices = sp.argsort(posterior_x)
f_empirical = sp.interpolate.interp1d(sp.hstack((-200,
posterior_x[sort_indices],
200)),
sp.hstack((0,
sp.cumsum(
posterior_weight[
sort_indices]),
1)))
@sp.vectorize
def f_expected(u):
return (sp.log(u)-sp.log(d_observed)) / (- sp.log(d_observed)) * \
(u > d_observed)
x = sp.linspace(0.1, 1)
max_distribution_difference = sp.absolute(f_empirical(x) -
f_expected(x)).max()
assert max_distribution_difference < 0.12
def test_beta_binomial_two_identical_models_adaptive(db_path, sampler):
binomial_n = 5
def model_fun(args):
return {"result": st.binom(binomial_n, args.theta).rvs()}
models = [model_fun for _ in range(2)]
models = list(map(SimpleModel, models))
population_size = AdaptivePopulationSize(800)
parameter_given_model_prior_distribution = [
Distribution(theta=st.beta(1, 1)) for _ in range(2)]
abc = ABCSMC(models, parameter_given_model_prior_distribution,
MinMaxDistanceFunction(measures_to_use=["result"]),
population_size,
eps=MedianEpsilon(.1),
sampler=sampler)
abc.new(db_path, {"result": 2})
minimum_epsilon = .2
history = abc.run(minimum_epsilon, max_nr_populations=3)
mp = history.get_model_probabilities(history.max_t)
assert abs(mp.p[0] - .5) + abs(mp.p[1] - .5) < .08
def test_two_parameters_and_two_used():
dist_f = MinMaxDistanceFunction(measures_to_use=["a", "b"])
abc = MockABC([{"a": -3, "b": 2}, {"a": 3, "b": 3}, {"a": 10, "b": 4}])
dist_f.initialize(abc.sample_from_prior())
d = dist_f({"a": 1, "b": 10}, {"a": 2, "b": 12})
assert 1 / 13 + 2 / 2 == d
def test_single_parameter():
dist_f = MinMaxDistanceFunction(measures_to_use=["a"])
abc = MockABC([{"a": -3}, {"a": 3}, {"a": 10}])
dist_f.initialize(abc.sample_from_prior())
d = dist_f({"a": 1}, {"a": 2})
assert 1 / 13 == d