def _initialize_kernel(self):
"""Create variables and logp function necessary to run kernel
This method should not be overwritten. If needed, use `setup_kernel`
instead.
"""
# Create dictionary that stores original variables shape and size
initial_point = self.model.recompute_initial_point(
seed=self.rng.integers(2**30))
for v in self.variables:
self.var_info[v.name] = (initial_point[v.name].shape,
initial_point[v.name].size)
# Create particles bijection map
if self.start:
init_rnd = self.start
else:
init_rnd = self.initialize_population()
population = []
for i in range(self.draws):
point = Point(
{v.name: init_rnd[v.name][i]
for v in self.variables},
model=self.model)
population.append(DictToArrayBijection.map(point).data)
self.tempered_posterior = np.array(floatX(population))
# Initialize prior and likelihood log probabilities
shared = make_shared_replacements(initial_point, self.variables,
self.model)
self.prior_logp_func = _logp_forw(initial_point, [self.model.varlogpt],
self.variables, shared)
self.likelihood_logp_func = _logp_forw(initial_point,
[self.model.datalogpt],
self.variables, shared)
priors = [
self.prior_logp_func(sample) for sample in self.tempered_posterior
]
likelihoods = [
self.likelihood_logp_func(sample)
for sample in self.tempered_posterior
]
self.prior_logp = np.array(priors).squeeze()
self.likelihood_logp = np.array(likelihoods).squeeze()
def __init__(
self,
vars=None,
batch_size=None,
total_size=None,
step_size=1.0,
model=None,
random_seed=None,
minibatches=None,
minibatch_tensors=None,
**kwargs
):
warnings.warn(EXPERIMENTAL_WARNING)
model = modelcontext(model)
if vars is None:
vars = model.value_vars
else:
vars = [model.rvs_to_values.get(var, var) for var in vars]
vars = inputvars(vars)
self.model = model
self.vars = vars
self.batch_size = batch_size
self.total_size = total_size
_value_error(
total_size != None or batch_size != None,
"total_size and batch_size of training data have to be specified",
)
self.expected_iter = int(total_size / batch_size)
# set random stream
self.random = None
if random_seed is None:
self.random = at_rng()
else:
self.random = at_rng(random_seed)
self.step_size = step_size
shared = make_shared_replacements(vars, model)
self.updates = OrderedDict()
# XXX: This needs to be refactored
self.q_size = None # int(sum(v.dsize for v in self.vars))
# This seems to be the only place that `Model.flatten` is used.
# TODO: Why not _actually_ flatten the variables?
# E.g. `flat_vars = at.concatenate([var.ravel() for var in vars])`
# or `set_subtensor` the `vars` into a `at.vector`?
flat_view = model.flatten(vars)
self.inarray = [flat_view.input]
self.dlog_prior = prior_dlogp(vars, model, flat_view)
self.dlogp_elemwise = elemwise_dlogL(vars, model, flat_view)
# XXX: This needs to be refactored
self.q_size = None # int(sum(v.dsize for v in self.vars))
if minibatch_tensors is not None:
_check_minibatches(minibatch_tensors, minibatches)
self.minibatches = minibatches
# Replace input shared variables with tensors
def is_shared(t):
return isinstance(t, aesara.compile.sharedvalue.SharedVariable)
tensors = [(t.type() if is_shared(t) else t) for t in minibatch_tensors]
updates = OrderedDict(
{t: t_ for t, t_ in zip(minibatch_tensors, tensors) if is_shared(t)}
)
self.minibatch_tensors = tensors
self.inarray += self.minibatch_tensors
self.updates.update(updates)
self._initialize_values()
super().__init__(vars, shared)
def __init__(self, vars=None, num_particles=40, max_stages=100, batch="auto", model=None):
_log.warning("BART is experimental. Use with caution.")
model = modelcontext(model)
initial_values = model.recompute_initial_point()
value_bart = inputvars(vars)[0]
self.bart = model.values_to_rvs[value_bart].owner.op
self.X = self.bart.X
self.Y = self.bart.Y
self.missing_data = np.any(np.isnan(self.X))
self.m = self.bart.m
self.alpha = self.bart.alpha
self.k = self.bart.k
self.alpha_vec = self.bart.split_prior
if self.alpha_vec is None:
self.alpha_vec = np.ones(self.X.shape[1])
self.init_mean = self.Y.mean()
# if data is binary
Y_unique = np.unique(self.Y)
if Y_unique.size == 2 and np.all(Y_unique == [0, 1]):
self.mu_std = 6 / (self.k * self.m ** 0.5)
# maybe we need to check for count data
else:
self.mu_std = (2 * self.Y.std()) / (self.k * self.m ** 0.5)
self.num_observations = self.X.shape[0]
self.num_variates = self.X.shape[1]
self.available_predictors = list(range(self.num_variates))
self.sum_trees = np.full_like(self.Y, self.init_mean).astype(aesara.config.floatX)
self.a_tree = Tree.init_tree(
leaf_node_value=self.init_mean / self.m,
idx_data_points=np.arange(self.num_observations, dtype="int32"),
)
self.mean = fast_mean()
self.normal = NormalSampler()
self.prior_prob_leaf_node = compute_prior_probability(self.alpha)
self.ssv = SampleSplittingVariable(self.alpha_vec)
self.tune = True
if batch == "auto":
batch = max(1, int(self.m * 0.1))
self.batch = (batch, batch)
else:
if isinstance(batch, (tuple, list)):
self.batch = batch
else:
self.batch = (batch, batch)
self.log_num_particles = np.log(num_particles)
self.indices = list(range(2, num_particles))
self.len_indices = len(self.indices)
self.max_stages = max_stages
shared = make_shared_replacements(initial_values, vars, model)
self.likelihood_logp = logp(initial_values, [model.datalogpt], vars, shared)
self.all_particles = []
for i in range(self.m):
self.a_tree.leaf_node_value = self.init_mean / self.m
p = ParticleTree(self.a_tree)
self.all_particles.append(p)
self.all_trees = np.array([p.tree for p in self.all_particles])
super().__init__(vars, shared)
