def _compile_pvariable_expression(self, expr: Expression,
scope: Dict[str, TensorFluent],
**kwargs) -> TensorFluent:
'''Compile a pvariable expression `expr` into a TensorFluent
in the given `scope`. The resulting TensorFluent will have
batch dimension given by `batch_size`.
Args:
expr (:obj:`rddl2tf.expr.Expression`): A RDDL pvariable expression.
scope (Dict[str, :obj:`rddl2tf.core.fluent.TensorFluent`]): A fluent scope.
kwargs: Additional keyword arguments.
Returns:
:obj:`rddl2tf.core.fluent.TensorFluent`: The compiled expression as a TensorFluent.
'''
etype = expr.etype
args = expr.args
name = expr._pvar_to_name(args)
if name not in scope:
raise ValueError('Variable {} not in scope.'.format(name))
fluent = scope[name]
scope = args[1] if args[1] is not None else []
if isinstance(fluent, TensorFluent):
fluent = TensorFluent(fluent.tensor, scope, batch=fluent.batch)
elif isinstance(fluent, tf.Tensor):
fluent = TensorFluent(fluent, scope, batch=True)
else:
raise ValueError(
'Variable in scope must be TensorFluent-like: {}'.format(
fluent))
return fluent
def _get_reparameterization(
self, expr: Expression, scope: ShapeScope, noise: NoiseList
) -> TensorFluentShape:
etype = expr.etype
args = expr.args
if etype[0] == "constant":
return TensorFluentShape([1], batch=False)
elif etype[0] == "pvar":
name = expr._pvar_to_name(args)
if name not in scope:
raise ValueError("Variable {} not in scope.".format(name))
shape = scope[name]
return shape
elif etype[0] == "randomvar":
if etype[1] == "Normal":
mean_shape = self._get_reparameterization(args[0], scope, noise)
var_shape = self._get_reparameterization(args[1], scope, noise)
shape = ReparameterizationCompiler._broadcast(mean_shape, var_shape)
dist = tf.distributions.Normal(loc=0.0, scale=1.0)
noise.append((dist, shape.as_list()))
return shape
elif etype[1] == "Exponential":
rate_shape = self._get_reparameterization(args[0], scope, noise)
dist = tf.distributions.Uniform(low=0.0, high=1.0)
noise.append((dist, rate_shape.as_list()))
return rate_shape
elif etype[1] == "Gamma":
for fluent in self.rddl.get_dependencies(expr):
if fluent.is_state_fluent() or fluent.is_action_fluent():
raise ValueError(
f"Expression is not an exogenous event: {expr}"
)
shape = []
with self.graph.as_default():
scope = self._scope.non_fluents(self.non_fluents)
shape_fluent = self._compile_expression(args[0], scope, noise=None)
scale_fluent = self._compile_expression(args[1], scope, noise=None)
concentration = shape_fluent.tensor
rate = 1 / scale_fluent.tensor
dist = tf.distributions.Gamma(concentration, rate)
noise.append((dist, shape))
return shape
elif etype[1] == "Uniform":
low_shape = self._get_reparameterization(args[0], scope, noise)
high_shape = self._get_reparameterization(args[1], scope, noise)
shape = ReparameterizationCompiler._broadcast(low_shape, high_shape)
dist = tf.distributions.Uniform(low=0.0, high=1.0)
noise.append((dist, shape.as_list()))
return shape
elif etype[0] in ["arithmetic", "boolean", "relational"]:
op1_shape = self._get_reparameterization(args[0], scope, noise)
shape = op1_shape
if len(args) > 1:
op2_shape = self._get_reparameterization(args[1], scope, noise)
shape = ReparameterizationCompiler._broadcast(op1_shape, op2_shape)
return shape
elif etype[0] == "func":
op1_shape = self._get_reparameterization(args[0], scope, noise)
shape = op1_shape
if len(args) > 1:
if len(args) == 2:
op2_shape = self._get_reparameterization(args[1], scope, noise)
shape = ReparameterizationCompiler._broadcast(op1_shape, op2_shape)
else:
raise ValueError("Invalid function:\n{}".format(expr))
return shape
elif etype[0] == "control":
if etype[1] == "if":
condition_shape = self._get_reparameterization(args[0], scope, noise)
true_case_shape = self._get_reparameterization(args[1], scope, noise)
false_case_shape = self._get_reparameterization(args[2], scope, noise)
shape = ReparameterizationCompiler._broadcast(
condition_shape, true_case_shape
)
shape = ReparameterizationCompiler._broadcast(shape, false_case_shape)
return shape
else:
raise ValueError("Invalid control flow expression:\n{}".format(expr))
elif etype[0] == "aggregation":
return self._get_reparameterization(args[-1], scope, noise)
raise ValueError("Expression type unknown: {}".format(etype))
def name(self) -> str:
'''Returns the CPF's pvariable name.'''
return Expression._pvar_to_name(self.pvar[1])
def _get_reparameterization(expr: Expression, scope: ShapeScope,
noise: Noise) -> TensorFluentShape:
etype = expr.etype
args = expr.args
if etype[0] == 'constant':
return TensorFluentShape([1], batch=False)
elif etype[0] == 'pvar':
name = expr._pvar_to_name(args)
if name not in scope:
raise ValueError('Variable {} not in scope.'.format(name))
shape = scope[name]
return shape
elif etype[0] == 'randomvar':
if etype[1] == 'Normal':
mean_shape = _get_reparameterization(args[0], scope, noise)
var_shape = _get_reparameterization(args[1], scope, noise)
shape = _broadcast(mean_shape, var_shape)
dist = tf.distributions.Normal(loc=0.0, scale=1.0)
noise.append((dist, shape.as_list()))
return shape
elif etype[1] == 'Exponential':
rate_shape = _get_reparameterization(args[0], scope, noise)
dist = tf.distributions.Uniform(low=0.0, high=1.0)
noise.append((dist, rate_shape.as_list()))
return rate_shape
elif etype[1] == 'Gamma':
raise NotImplementedError
elif etype[1] == 'Uniform':
low_shape = _get_reparameterization(args[0], scope, noise)
high_shape = _get_reparameterization(args[1], scope, noise)
shape = _broadcast(low_shape, high_shape)
dist = tf.distributions.Uniform(low=0.0, high=1.0)
noise.append((dist, shape.as_list()))
return shape
elif etype[0] in ['arithmetic', 'boolean', 'relational']:
op1_shape = _get_reparameterization(args[0], scope, noise)
shape = op1_shape
if len(args) > 1:
op2_shape = _get_reparameterization(args[1], scope, noise)
shape = _broadcast(op1_shape, op2_shape)
return shape
elif etype[0] == 'func':
op1_shape = _get_reparameterization(args[0], scope, noise)
shape = op1_shape
if len(args) > 1:
if len(args) == 2:
op2_shape = _get_reparameterization(args[1], scope, noise)
shape = _broadcast(op1_shape, op2_shape)
else:
raise ValueError('Invalid function:\n{}'.format(expr))
return shape
elif etype[0] == 'control':
if etype[1] == 'if':
condition_shape = _get_reparameterization(args[0], scope, noise)
true_case_shape = _get_reparameterization(args[1], scope, noise)
false_case_shape = _get_reparameterization(args[2], scope, noise)
shape = _broadcast(condition_shape, true_case_shape)
shape = _broadcast(shape, false_case_shape)
return shape
else:
raise ValueError(
'Invalid control flow expression:\n{}'.format(expr))
elif etype[0] == 'aggregation':
return _get_reparameterization(args[-1], scope, noise)
raise ValueError('Expression type unknown: {}'.format(etype))