class Parameters(ModulatoryProjection_Base.Parameters):
"""
Attributes
----------
value
see `value <LearningProjection.value>`
:default value: numpy.array([0])
:type: numpy.ndarray
:read only: True
error_function
see `error_function <LearningProjection.error_function>`
:default value: `LinearCombination`(weights=numpy.array([[-1], [ 1]]))
:type: `Function`
function
see `function <LearningProjection.function>`
:default value: `Linear`
:type: `Function`
learning_function
see `learning_function <LearningProjection.learning_function>`
:default value: `BackPropagation`
:type: `Function`
learning_rate
see `learning_rate <LearningProjection.learning_rate>`
:default value: None
:type:
learning_signal
see `learning_signal <LearningProjection.learning_signal>`
:default value: None
:type:
:read only: True
"""
value = Parameter(np.array([0]),
read_only=True,
aliases=['weight_change_matrix'])
function = Parameter(Linear, stateful=False, loggable=False)
error_function = Parameter(LinearCombination(weights=[[-1], [1]]),
stateful=False,
loggable=False)
learning_function = Parameter(BackPropagation,
stateful=False,
loggable=False)
learning_rate = Parameter(None, modulable=True)
learning_signal = Parameter(None,
read_only=True,
getter=_learning_signal_getter,
setter=_learning_signal_setter)
def __init__(
self,
sender: tc.optional(tc.any(LearningSignal,
LearningMechanism)) = None,
receiver: tc.optional(tc.any(ParameterPort,
MappingProjection)) = None,
error_function: tc.optional(is_function_type) = LinearCombination(
weights=[[-1], [1]]),
learning_function: tc.optional(is_function_type) = BackPropagation,
# FIX: 10/3/17 - TEST IF THIS OK AND REINSTATE IF SO
# learning_signal_params:tc.optional(dict)=None,
learning_rate: tc.optional(tc.any(parameter_spec)) = None,
learning_enabled: tc.optional(tc.any(bool, tc.enum(ONLINE,
AFTER))) = None,
weight=None,
exponent=None,
params: tc.optional(dict) = None,
name=None,
prefs: is_pref_set = None,
**kwargs):
# IMPLEMENTATION NOTE:
# the error_function and learning_function arguments are implemented to preserve the ability to pass
# error function and learning function specifications from the specification of a LearningProjection (used
# to implement learning for a MappingProjection, e.g., in a tuple) to the LearningMechanism responsible
# for implementing the function; and for specifying the default LearningProjection for a Process.
# If receiver has not been assigned, defer init to Port.instantiate_projection_to_state()
if sender is None or receiver is None:
# Flag for deferred initialization
self.initialization_status = ContextFlags.DEFERRED_INIT
# parameters should be passed through methods like
# instantiate_sender instead of grabbed from attributes like this
self._learning_function = learning_function
self._learning_rate = learning_rate
self._error_function = error_function
# replaces similar code in _instantiate_sender
try:
if sender.owner.learning_rate is not None:
learning_rate = sender.owner.learning_rate
except AttributeError:
pass
super().__init__(sender=sender,
receiver=receiver,
weight=weight,
exponent=exponent,
params=params,
name=name,
prefs=prefs,
error_function=error_function,
learning_function=learning_function,
learning_rate=learning_rate,
learning_enabled=learning_enabled,
**kwargs)
class Parameters(ObjectiveMechanism.Parameters):
"""
Attributes
----------
variable
see `variable <ComparatorMechanism.variable>`
:default value: numpy.array([[0], [0]])
:type: ``numpy.ndarray``
:read only: True
function
see `function <ComparatorMechanism.function>`
:default value: `LinearCombination`(weights=numpy.array([[-1], [ 1]]))
:type: `Function`
output_ports
see `output_ports <ComparatorMechanism.output_ports>`
:default value: [`OUTCOME`]
:type: ``list``
:read only: True
sample
see `sample <ComparatorMechanism.sample>`
:default value: None
:type:
target
see `target <ComparatorMechanism.target>`
:default value: None
:type:
"""
# By default, ComparatorMechanism compares two 1D np.array input_ports
variable = Parameter(np.array([[0], [0]]),
read_only=True,
pnl_internal=True,
constructor_argument='default_variable')
function = Parameter(LinearCombination(weights=[[-1], [1]]),
stateful=False,
loggable=False)
sample = None
target = None
output_ports = Parameter(
[OUTCOME],
stateful=False,
loggable=False,
read_only=True,
structural=True,
)
def __init__(
self,
sender: tc.optional(tc.any(LearningSignal,
LearningMechanism)) = None,
receiver: tc.optional(tc.any(ParameterPort,
MappingProjection)) = None,
error_function: tc.optional(is_function_type) = LinearCombination(
weights=[[-1], [1]]),
learning_function: tc.optional(is_function_type) = BackPropagation,
# FIX: 10/3/17 - TEST IF THIS OK AND REINSTATE IF SO
# learning_signal_params:tc.optional(dict)=None,
learning_rate: tc.optional(tc.any(parameter_spec)) = None,
learning_enabled: tc.optional(tc.any(bool, tc.enum(ONLINE,
AFTER))) = None,
weight=None,
exponent=None,
params: tc.optional(dict) = None,
name=None,
prefs: is_pref_set = None,
**kwargs):
# IMPLEMENTATION NOTE:
# the error_function and learning_function arguments are implemented to preserve the ability to pass
# error function and learning function specifications from the specification of a LearningProjection (used
# to implement learning for a MappingProjection, e.g., in a tuple) to the LearningMechanism responsible
# for implementing the function; and for specifying the default LearningProjection for a Process.
# Assign args to params and functionParams dicts
params = self._assign_args_to_param_dicts(
error_function=error_function,
learning_function=learning_function,
learning_rate=learning_rate,
# FIX: 10/3/17 - TEST IF THIS OK AND REINSTATE IF SO
# learning_signal_params=learning_signal_params,
learning_enabled=learning_enabled,
weight=weight,
exponent=exponent,
params=params)
# If receiver has not been assigned, defer init to Port.instantiate_projection_to_state()
if sender is None or receiver is None:
# Flag for deferred initialization
self.initialization_status = ContextFlags.DEFERRED_INIT
super().__init__(sender=sender,
receiver=receiver,
weight=weight,
exponent=exponent,
params=params,
name=name,
prefs=prefs,
**kwargs)
def __init__(
self,
default_variable=None,
sample: tc.optional(
tc.any(OutputState, Mechanism_Base, dict, is_numeric, str)) = None,
target: tc.optional(
tc.any(OutputState, Mechanism_Base, dict, is_numeric, str)) = None,
function=LinearCombination(weights=[[-1], [1]]),
output_states: tc.optional(tc.any(str, Iterable)) = (OUTCOME, ),
params=None,
name=None,
prefs: is_pref_set = None,
**
input_states # IMPLEMENTATION NOTE: this is for backward compatibility
):
input_states = self._merge_legacy_constructor_args(
sample, target, default_variable, input_states)
# Default output_states is specified in constructor as a tuple rather than a list
# to avoid "gotcha" associated with mutable default arguments
# (see: bit.ly/2uID3s3 and http://docs.python-guide.org/en/latest/writing/gotchas/)
if isinstance(output_states, (str, tuple)):
output_states = list(output_states)
# IMPLEMENTATION NOTE: The following prevents the default from being updated by subsequent assignment
# (in this case, to [OUTCOME, {NAME= MSE}]), but fails to expose default in IDE
# output_states = output_states or [OUTCOME, MSE]
# Create a StandardOutputStates object from the list of stand_output_states specified for the class
if not isinstance(self.standard_output_states, StandardOutputStates):
self.standard_output_states = StandardOutputStates(
self, self.standard_output_states, indices=PRIMARY)
super().__init__( # monitor=[sample, target],
monitor=input_states,
function=function,
output_states=output_states.copy(
), # prevent default from getting overwritten by later assign
params=params,
name=name,
prefs=prefs,
context=ContextFlags.CONSTRUCTOR)
# Require Projection to TARGET InputState (already required for SAMPLE as primary InputState)
self.input_states[1].parameters.require_projection_in_composition.set(
True, override=True)
def __init__(self,
default_variable=None,
sample: tc.optional(
tc.any(OutputPort, Mechanism_Base, dict, is_numeric,
str)) = None,
target: tc.optional(
tc.any(OutputPort, Mechanism_Base, dict, is_numeric,
str)) = None,
function=LinearCombination(weights=[[-1], [1]]),
output_ports: tc.optional(tc.any(str, Iterable)) = None,
params=None,
name=None,
prefs: is_pref_set = None,
**kwargs):
input_ports = kwargs.pop(INPUT_PORTS, {})
if input_ports:
input_ports = {INPUT_PORTS: input_ports}
input_ports = self._merge_legacy_constructor_args(
sample, target, default_variable, input_ports)
# Default output_ports is specified in constructor as a tuple rather than a list
# to avoid "gotcha" associated with mutable default arguments
# (see: bit.ly/2uID3s3 and http://docs.python-guide.org/en/latest/writing/gotchas/)
if isinstance(output_ports, (str, tuple)):
output_ports = list(output_ports)
# IMPLEMENTATION NOTE: The following prevents the default from being updated by subsequent assignment
# (in this case, to [OUTCOME, {NAME= MSE}]), but fails to expose default in IDE
# output_ports = output_ports or [OUTCOME, MSE]
super().__init__(
monitor=input_ports,
function=function,
output_ports=
output_ports, # prevent default from getting overwritten by later assign
params=params,
name=name,
prefs=prefs,
**kwargs)
# Require Projection to TARGET InputPort (already required for SAMPLE as primary InputPort)
self.input_ports[1].parameters.require_projection_in_composition._set(
True, Context())
class Parameters(ObjectiveMechanism.Parameters):
"""
Attributes
----------
variable
see `variable <ComparatorMechanism.variable>`
:default value: numpy.array([[0], [0]])
:type: numpy.ndarray
:read only: True
function
see `function <ComparatorMechanism.function>`
:default value: `LinearCombination`(offset=0.0, operation=sum, scale=1.0, weights=numpy.array([[-1], [ 1]]))
:type: `Function`
sample
see `sample <ComparatorMechanism.sample>`
:default value: None
:type:
target
see `target <ComparatorMechanism.target>`
:default value: None
:type:
"""
# By default, ComparatorMechanism compares two 1D np.array input_states
variable = Parameter(np.array([[0], [0]]), read_only=True)
function = Parameter(LinearCombination(weights=[[-1], [1]]),
stateful=False,
loggable=False)
sample = None
target = None
