def test_compile_probabilistic_normal_random_variable(self):
mean = Expression(('number', 0.0))
var = Expression(('number', 1.0))
normal = Expression(('randomvar', ('Normal', (mean, var))))
expressions = [normal]
self._test_random_variable_expressions(expressions)
def test_compile_probabilistic_gamma_random_variable(self):
shape = Expression(('number', 5.0))
scale = Expression(('number', 1.0))
gamma = Expression(('randomvar', ('Gamma', (shape, scale))))
expressions = [gamma]
self._test_random_variable_expressions(expressions)
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 test_gamma(compiler):
if compiler.rddl.domain.name == "reservoir":
shape_scope = compiler._get_reparameterization_shape_scope()
shape = Expression(("pvar_expr", ("RAIN_SHAPE", ["?res"])))
scale = Expression(("pvar_expr", ("RAIN_SCALE", ["?res"])))
gamma = Expression(("randomvar", ("Gamma", (shape, scale))))
noise_lst = compiler._get_expression_reparameterization(
gamma, shape_scope)
assert isinstance(noise_lst, list)
assert len(noise_lst) == 1
dist, shape = noise_lst[0]
assert isinstance(dist, GAMMA)
assert dist.batch_shape == (8, )
assert dist.event_shape == ()
assert shape == []
def p_expr(self, p):
'''expr : pvar_expr
| group_expr
| function_expr
| relational_expr
| boolean_expr
| quantifier_expr
| numerical_expr
| aggregation_expr
| control_expr
| randomvar_expr'''
p[0] = Expression(p[1])
import pytest
import tensorflow as tf
from pyrddl.expr import Expression
import rddlgym
from rddl2tf.compilers import ReparameterizationCompiler
from rddl2tf.core.fluent import TensorFluent
from rddl2tf.core.fluentshape import TensorFluentShape
NORMAL = tf.compat.v1.distributions.Normal
GAMMA = tf.compat.v1.distributions.Gamma
UNIFORM = tf.compat.v1.distributions.Uniform
ZERO = Expression(("number", 0.0))
ONE = Expression(("number", 1.0))
TWO = Expression(("+", (ONE, ONE)))
Z = Expression(("randomvar", ("Normal", (ZERO, ONE))))
MU = Expression(("pvar_expr", ("mu", ["?x"])))
SIGMA = Expression(("pvar_expr", ("sigma", ["?x"])))
X1 = Expression(("randomvar", ("Normal", (MU, ONE))))
X2 = Expression(("randomvar", ("Normal", (ZERO, SIGMA))))
X3 = Expression(("randomvar", ("Normal", (MU, SIGMA))))
X4 = Expression(("randomvar", ("Normal", (Z, ONE))))
X5 = Expression(("randomvar", ("Normal", (X1, ONE))))
X6 = Expression(("randomvar", ("Normal", (X1, SIGMA))))
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 test_get_dependencies(rddl):
if rddl.domain.name == 'reservoir':
shape = Expression(('pvar_expr', ('RAIN_SHAPE', ['?res'])))
scale = Expression(('pvar_expr', ('RAIN_SCALE', ['?res'])))
gamma = Expression(('randomvar', ('Gamma', (shape, scale))))
deps = rddl.get_dependencies(gamma)
assert isinstance(deps, set)
assert len(deps) == 2
deps = set(map(lambda fluent: str(fluent), deps))
assert deps == {'RAIN_SCALE/1', 'RAIN_SHAPE/1'}
elif rddl.domain.name == 'hvac_vav_fix':
mean = Expression(('pvar_expr', ('TEMP_OUTSIDE_MEAN', ['?s'])))
variance = Expression(('pvar_expr', ('TEMP_OUTSIDE_VARIANCE', ['?s'])))
normal = Expression(('randomvar', ('Normal', (mean, variance))))
deps = rddl.get_dependencies(normal)
assert isinstance(deps, set)
assert len(deps) == 2
deps = set(map(lambda fluent: str(fluent), deps))
assert deps == {'TEMP_OUTSIDE_MEAN/1', 'TEMP_OUTSIDE_VARIANCE/1'}
elif rddl.domain.name == 'Navigation':
location = Expression(('pvar_expr', ('location', ['?l'])))
move = Expression(('pvar_expr', ('move', ['?l'])))
deceleration = Expression(
('prod', (('typed_var', ('?z', 'zone')),
Expression(('pvar_expr', ('deceleration', ['?z']))))))
mean = Expression(
('+', (location, Expression(('*', (deceleration, move))))))
variance = Expression(('*', (Expression(
('pvar_expr', ('MOVE_VARIANCE_MULT', ['?l']))), move)))
normal = Expression(('randomvar', ('Normal', (mean, variance))))
deps = rddl.get_dependencies(normal)
assert isinstance(deps, set)
assert len(deps) == 5
deps = set(map(lambda fluent: str(fluent), deps))
assert deps == {
'location/1', 'move/1', 'DECELERATION_ZONE_CENTER/2',
'DECELERATION_ZONE_DECAY/1', 'MOVE_VARIANCE_MULT/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))
def p_case_expr(self, p):
'''case_expr : penum_expr
| pvar_expr
| numerical_expr'''
p[0] = Expression(p[1])
def p_term(self, p):
'''term : VAR
| ENUM_VAL
| pvar_expr'''
p[0] = Expression(p[1]) if isinstance(
p[1], tuple) and p[1][0] == 'pvar_expr' else p[1]
def setUpClass(cls):
cls.zero = Expression(('number', 0.0))
cls.one = Expression(('number', 1.0))
cls.two = Expression(('+', (cls.one, cls.one)))
cls.z = Expression(('randomvar', ('Normal', (cls.zero, cls.one))))
cls.mu = Expression(('pvar_expr', ('mu', ['?x'])))
cls.sigma = Expression(('pvar_expr', ('sigma', ['?x'])))
cls.x1 = Expression(('randomvar', ('Normal', (cls.mu, cls.one))))
cls.x2 = Expression(('randomvar', ('Normal', (cls.zero, cls.sigma))))
cls.x3 = Expression(('randomvar', ('Normal', (cls.mu, cls.sigma))))
cls.x4 = Expression(('randomvar', ('Normal', (cls.z, cls.one))))
cls.x5 = Expression(('randomvar', ('Normal', (cls.x1, cls.one))))
cls.x6 = Expression(('randomvar', ('Normal', (cls.x1, cls.sigma))))
cls.mu_plus_z = Expression(('+', (cls.mu, cls.z)))
cls.z_plus_mu = Expression(('+', (cls.z, cls.mu)))
cls.mu_plus_x2 = Expression(('+', (cls.mu, cls.x2)))
cls.x2_plus_mu = Expression(('+', (cls.x2, cls.mu)))
cls.x1_plus_z = Expression(('+', (cls.x1, cls.z)))
cls.z_plus_x1 = Expression(('+', (cls.z, cls.x1)))
cls.z_times_z = Expression(('*', (cls.z, cls.z)))
cls.x2_times_x2 = Expression(('*', (cls.x2, cls.x2)))
cls.x7 = Expression(('randomvar', ('Normal', (cls.one, cls.z_times_z))))
cls.x8 = Expression(('randomvar', ('Normal', (cls.mu, cls.z_times_z))))
cls.x9 = Expression(('randomvar', ('Normal', (cls.x3, cls.z_times_z))))
cls.exp_2 = Expression(('func', ('exp', [cls.two])))
cls.exp_z = Expression(('func', ('exp', [cls.z])))
cls.exp_x1 = Expression(('func', ('exp', [cls.x1])))
cls.y1 = Expression(('randomvar', ('Normal', (cls.one, cls.exp_z))))
cls.y2 = Expression(('randomvar', ('Normal', (cls.mu, cls.exp_z))))
cls.y3 = Expression(('randomvar', ('Normal', (cls.mu, cls.exp_x1))))
cls.gamma_shape = Expression(('pvar_expr', ('shape', ['?r'])))
cls.gamma_scale = Expression(('pvar_expr', ('scale', ['?r'])))
cls.gamma1 = Expression(('randomvar', ('Gamma', (cls.gamma_shape, cls.gamma_scale))))
cls.exp_rate = Expression(('pvar_expr', ('rate', ['?r'])))
cls.exp1 = Expression(('randomvar', ('Exponential', (cls.exp_rate,))))
