topo.learningfn Package

---

learningfn Package

A family of function objects for changing a set of weights over time.

Learning functions come in two varieties: LearningFunction, and CFPLearningFunction. A LearningFunction (e.g. Hebbian) applies to one set of weights, typically from one ConnectionField. To apply learning to an entire CFProjection, a LearningFunction can be plugged in to CFPLF_Plugin. CFPLF_Plugin is one example of a CFPLearningFunction, which is a function that works with the entire Projection at once. Some optimizations and algorithms can only be applied at the full CFPLearningFn level, so there are other CFPLearningFns beyond CFPLF_Plugin.

Any new learning functions added to this directory will automatically become available for any model.

class topo.learningfn.AntiHebbian(**params)

Bases: topo.base.functionfamily.LearningFn

Opposite of the basic Hebbian rule.

Same as Dayan and Abbott, 2001, equation 8.3, except that the weight change is negative. I.e., each weight decreases in proportion to the product of this neuron’s activity and the input activity.

Requires some form of output_fn normalization for stability.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07ab462a48>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07ab4627e0>

inspect_value = <functools.partial object at 0x2b07ab462aa0>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07ab4629f0>

set_param = <functools.partial object at 0x2b07ab462af8>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

class topo.learningfn.IdentityLF(**params)

Bases: topo.base.functionfamily.LearningFn

Identity function; does not modify the weights.

For speed, calling this function object is sometimes optimized away entirely. To make this feasible, it is not allowable to derive other classes from this object, modify it to have different behavior, add side effects, or anything of that nature.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07ab462890>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07ab462838>

inspect_value = <functools.partial object at 0x2b07ab4628e8>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07ab462a48>

set_param = <functools.partial object at 0x2b07ab462940>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

class topo.learningfn.BCMFixed(**params)

Bases: topo.base.functionfamily.LearningFn

Bienenstock, Cooper, and Munro (1982) learning rule with a fixed threshold.

(See Dayan and Abbott, 2001, equation 8.12) In the BCM rule, activities change only when there is both pre- and post-synaptic activity. The full BCM rule requires a sliding threshold (see CFPBCM), but this version is simpler and easier to analyze.

Requires some form of output_fn normalization for stability.

param Number unit_threshold (allow_None=False, bounds=(0, None), constant=False, default=0.5, inclusive_bounds=(True, True), instantiate=False, pickle_default_value=True, precedence=None, readonly=False, time_dependent=True, time_fn=<Time Time00001>)

Threshold between LTD and LTP.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07ab462ba8>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07ab462aa0>

inspect_value = <functools.partial object at 0x2b07ab462890>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07ab4627e0>

set_param = <functools.partial object at 0x2b07ab462c00>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

class topo.learningfn.Covariance(**params)

Bases: topo.base.functionfamily.LearningFn

Covariance learning rule supporting either input or unit thresholds.

As presented by Dayan and Abbott (2001), covariance rules allow either potentiation or depression of the same synapse, depending on an activity level. By default, this implementation follows Dayan and Abbott equation 8.8, with the unit_threshold determining the level of postsynaptic activity (activity of the target unit), below which LTD (depression) will occur.

If you wish to use an input threshold as in Dayan and Abbott equation 8.9 instead, set unit_threshold to zero and change input_threshold to some positive value instead. When both thresholds are zero this rule degenerates to the standard Hebbian rule.

Requires some form of output_fn normalization for stability.

param Number input_threshold (allow_None=False, bounds=(0, None), constant=False, default=0.0, inclusive_bounds=(True, True), instantiate=False, pickle_default_value=True, precedence=None, readonly=False, time_dependent=True, time_fn=<Time Time00001>)

Threshold between LTD and LTP, applied to the input activity.

param Number unit_threshold (allow_None=False, bounds=(0, None), constant=False, default=0.5, inclusive_bounds=(True, True), instantiate=False, pickle_default_value=True, precedence=None, readonly=False, time_dependent=True, time_fn=<Time Time00001>)

Threshold between LTD and LTP, applied to the activity of this unit.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07ab4628e8>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07ab462838>

inspect_value = <functools.partial object at 0x2b07ab462ba8>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07ab462998>

set_param = <functools.partial object at 0x2b07ab462a48>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

class topo.learningfn.LearningFn(**params)

Bases: param.parameterized.Parameterized

Abstract base class for learning functions that plug into CFPLF_Plugin.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07ab462aa0>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07ab4628e8>

inspect_value = <functools.partial object at 0x2b07ab462af8>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07ab462890>

set_param = <functools.partial object at 0x2b07ab4627e0>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

class topo.learningfn.Oja(**params)

Bases: topo.base.functionfamily.LearningFn

Oja’s rule (Oja, 1982; Dayan and Abbott, 2001, equation 8.16.)

Hebbian rule with soft multiplicative normalization, tending the weights toward a constant sum-squared value over time. Thus this function does not normally need a separate output_fn for normalization.

param Number alpha (allow_None=False, bounds=(0, None), constant=False, default=0.1, inclusive_bounds=(True, True), instantiate=False, pickle_default_value=True, precedence=None, readonly=False, time_dependent=True, time_fn=<Time Time00001>)

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07ab462940>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07ab462838>

inspect_value = <functools.partial object at 0x2b07ab462ba8>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07ab462aa0>

set_param = <functools.partial object at 0x2b07ab462b50>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

class topo.learningfn.Hebbian(**params)

Bases: topo.base.functionfamily.LearningFn

Basic Hebbian rule; Dayan and Abbott, 2001, equation 8.3.

Increases each weight in proportion to the product of this neuron’s activity and the input activity.

Requires some form of output_fn normalization for stability.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07ab462af8>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07ab462c00>

inspect_value = <functools.partial object at 0x2b07ab462890>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07ab462940>

set_param = <functools.partial object at 0x2b07ab462a48>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

class topo.learningfn.CPCA(**params)

Bases: topo.base.functionfamily.LearningFn

CPCA (Conditional Principal Component Analysis) rule.

(See O’Reilly and Munakata, Computational Explorations in Cognitive Neuroscience, 2000, equation 4.12.)

Increases each weight in proportion to the product of this neuron’s activity, input activity, and connection weights.

Has built-in normalization, and so does not require output_fn normalization for stability. Intended to be a more biologically plausible version of the Oja rule.

Submitted by Veldri Kurniawan and Lewis Ng.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07ab462998>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07ab462aa0>

inspect_value = <functools.partial object at 0x2b07ab462ba8>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07ab462c00>

set_param = <functools.partial object at 0x2b07ab4629f0>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

---

optimized Module

Learning functions and projection-level learning functions (see projfn.py) written in C to optimize performance.

Requires the weave package; without it unoptimized versions are used.

class topo.learningfn.optimized.CFPLF_BCMFixed(**params)

Bases: topo.base.cf.CFPLF_Plugin

Same as CFPLF_Plugin(single_cf_fn=BCMFixed()); just for non-optimized fallback.

param ClassSelector single_cf_fn (allow_None=False, constant=True, default=<BCMFixed BCMFixed01328>, instantiate=True, is_instance=True, pickle_default_value=True, precedence=None, readonly=True)

Accepts a LearningFn that will be applied to each CF individually.

constant_sum_connection_rate(n_units, learning_rate)

Return the learning rate for a single connection assuming that the total rate is to be divided evenly among all the units in the connection field.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07aafda890>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07aafda8e8>

inspect_value = <functools.partial object at 0x2b07aafda940>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07aafda5d0>

set_param = <functools.partial object at 0x2b07aafdaa48>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

class topo.learningfn.optimized.CFPLF_BCMFixed_opt(**params)

Bases: topo.base.cf.CFPLearningFn

CF-aware BCM learning rule.

Implemented in C for speed. Should be equivalent to BCMFixed for CF sheets, except faster.

As a side effect, sets the norm_total attribute on any cf whose weights are updated during learning, to speed up later operations that might depend on it.

May return without modifying anything if the learning rate turns out to be zero.

param Number unit_threshold (allow_None=False, bounds=(0, None), constant=False, default=0.5, inclusive_bounds=(True, True), instantiate=False, pickle_default_value=True, precedence=None, readonly=False, time_dependent=True, time_fn=<Time Time00001>)

Threshold between LTD and LTP.

constant_sum_connection_rate(n_units, learning_rate)

Return the learning rate for a single connection assuming that the total rate is to be divided evenly among all the units in the connection field.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07aafda628>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07aafda6d8>

inspect_value = <functools.partial object at 0x2b07aafda730>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07aafda7e0>

set_param = <functools.partial object at 0x2b07aafdaaa0>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

class topo.learningfn.optimized.CFPLF_Hebbian(**params)

Bases: topo.base.cf.CFPLF_Plugin

Same as CFPLF_Plugin(single_cf_fn=Hebbian()); just for non-optimized fallback.

param ClassSelector single_cf_fn (allow_None=False, constant=True, default=<Hebbian Hebbian01327>, instantiate=True, is_instance=True, pickle_default_value=True, precedence=None, readonly=True)

Accepts a LearningFn that will be applied to each CF individually.

constant_sum_connection_rate(n_units, learning_rate)

Return the learning rate for a single connection assuming that the total rate is to be divided evenly among all the units in the connection field.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07aafda890>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07aafda940>

inspect_value = <functools.partial object at 0x2b07aafda838>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07aafdaaa0>

set_param = <functools.partial object at 0x2b07aafda998>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

class topo.learningfn.optimized.CFPLF_Hebbian_opt(**params)

Bases: topo.base.cf.CFPLearningFn

CF-aware Hebbian learning rule.

Implemented in C for speed. Should be equivalent to CFPLF_Plugin(single_cf_fn=Hebbian), except faster.

As a side effect, sets the norm_total attribute on any cf whose weights are updated during learning, to speed up later operations that might depend on it.

May return without modifying anything if the learning rate turns out to be zero.

param ClassSelector single_cf_fn (allow_None=False, constant=True, default=<Hebbian Hebbian01326>, instantiate=False, is_instance=True, pickle_default_value=True, precedence=None, readonly=True)

constant_sum_connection_rate(n_units, learning_rate)

Return the learning rate for a single connection assuming that the total rate is to be divided evenly among all the units in the connection field.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07aafda6d8>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07aafda730>

inspect_value = <functools.partial object at 0x2b07aafdaa48>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07aafda680>

set_param = <functools.partial object at 0x2b07aafda9f0>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

class topo.learningfn.optimized.CFPLF_Scaled(**params)

Bases: topo.learningfn.projfn.CFPLF_PluginScaled

Same as CFPLF_PluginScaled(single_cf_fn=Hebbian()); just for non-optimized fallback.

param ClassSelector single_cf_fn (allow_None=False, constant=True, default=<Hebbian Hebbian01330>, instantiate=True, is_instance=True, pickle_default_value=True, precedence=None, readonly=True)

Accepts a LearningFn that will be applied to each CF individually.

param Parameter learning_rate_scaling_factor (allow_None=True, constant=False, default=None, instantiate=False, pickle_default_value=True, precedence=None, readonly=False)

Matrix of scaling factors for scaling the learning rate of each CF individually.

constant_sum_connection_rate(n_units, learning_rate)

Return the learning rate for a single connection assuming that the total rate is to be divided evenly among all the units in the connection field.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07aafda890>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07aafda838>

inspect_value = <functools.partial object at 0x2b07aafda788>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07aafda998>

set_param = <functools.partial object at 0x2b07aafda8e8>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

update_scaling_factor(new_scaling_factor)

Update the single-connection learning rate scaling factor.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

class topo.learningfn.optimized.CFPLF_Scaled_opt(**params)

Bases: topo.learningfn.projfn.CFPLF_PluginScaled

CF-aware Scaled Hebbian learning rule.

Implemented in C for speed. Should be equivalent to CFPLF_PluginScaled(single_cf_fn=Hebbian), except faster.

As a side effect, sets the norm_total attribute on any cf whose weights are updated during learning, to speed up later operations that might depend on it.

param ClassSelector single_cf_fn (allow_None=False, constant=True, default=<Hebbian Hebbian01329>, instantiate=True, is_instance=True, pickle_default_value=True, precedence=None, readonly=True)

Accepts a LearningFn that will be applied to each CF individually.

param Parameter learning_rate_scaling_factor (allow_None=True, constant=False, default=None, instantiate=False, pickle_default_value=True, precedence=None, readonly=False)

Matrix of scaling factors for scaling the learning rate of each CF individually.

constant_sum_connection_rate(n_units, learning_rate)

Return the learning rate for a single connection assuming that the total rate is to be divided evenly among all the units in the connection field.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07aafda6d8>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07aafda730>

inspect_value = <functools.partial object at 0x2b07aafdaa48>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07aafda9f0>

set_param = <functools.partial object at 0x2b07aafda5d0>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

update_scaling_factor(new_scaling_factor)

Update the single-connection learning rate scaling factor.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

class topo.learningfn.optimized.CFPLF_Trace_opt(**params)

Bases: topo.base.cf.CFPLearningFn

Optimized version of CFPLF_Trace; see projfn.py for more info

param ClassSelector single_cf_fn (allow_None=False, constant=True, default=<Hebbian Hebbian01331>, instantiate=False, is_instance=True, pickle_default_value=True, precedence=None, readonly=True)

LearningFn that will be applied to each CF individually.

param Number trace_strength (allow_None=False, bounds=(0.0, 1.0), constant=False, default=0.5, inclusive_bounds=(True, True), instantiate=False, pickle_default_value=True, precedence=None, readonly=False, time_dependent=True, time_fn=<Time Time00001>)

How much the learning is dominated by the activity trace, relative to the current value.

constant_sum_connection_rate(n_units, learning_rate)

Return the learning rate for a single connection assuming that the total rate is to be divided evenly among all the units in the connection field.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07aafda628>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07aafda838>

inspect_value = <functools.partial object at 0x2b07aafda788>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07aafda890>

set_param = <functools.partial object at 0x2b07aafda940>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

---

projfn Module

Learning functions for Projections.

For example, CFProjectionLearningFunctions compute a new set of ConnectionFields when given an input and output pattern and a set of ConnectionField objects.

class topo.learningfn.projfn.CFPLF_Identity(**params)

Bases: topo.base.cf.CFPLearningFn

CFLearningFunction performing no learning.

param ClassSelector single_cf_fn (allow_None=False, constant=True, default=<IdentityLF IdentityLF01275>, instantiate=True, is_instance=True, pickle_default_value=True, precedence=None, readonly=False)

constant_sum_connection_rate(n_units, learning_rate)

Return the learning rate for a single connection assuming that the total rate is to be divided evenly among all the units in the connection field.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07ab940db8>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07ab940998>

inspect_value = <functools.partial object at 0x2b07ab9406d8>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07ab940418>

set_param = <functools.partial object at 0x2b07ab940940>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

class topo.learningfn.projfn.CFPLF_Plugin(**params)

Bases: topo.base.cf.CFPLearningFn

CFPLearningFunction applying the specified single_cf_fn to each CF.

param ClassSelector single_cf_fn (allow_None=False, constant=False, default=<Hebbian Hebbian01276>, instantiate=True, is_instance=True, pickle_default_value=True, precedence=None, readonly=False)

Accepts a LearningFn that will be applied to each CF individually.

constant_sum_connection_rate(n_units, learning_rate)

Return the learning rate for a single connection assuming that the total rate is to be divided evenly among all the units in the connection field.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07ab940940>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07ab9408e8>

inspect_value = <functools.partial object at 0x2b07ab940c58>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07ab940d60>

set_param = <functools.partial object at 0x2b07ab940c00>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

class topo.learningfn.projfn.CFPLF_EuclideanHebbian(**params)

Bases: topo.base.cf.CFPLearningFn

Hebbian CFProjection learning rule based on Euclidean distance.

Learning is driven by the distance from the input pattern to the weights, scaled by the current activity. To implement a Kohonen SOM algorithm, the activity should be the neighborhood kernel centered around the winning unit, as implemented by KernelMax.

constant_sum_connection_rate(n_units, learning_rate)

Return the learning rate for a single connection assuming that the total rate is to be divided evenly among all the units in the connection field.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07ab940998>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07ab9406d8>

inspect_value = <functools.partial object at 0x2b07ab940cb0>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07ab940af8>

set_param = <functools.partial object at 0x2b07ab9409f0>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

class topo.learningfn.projfn.CFPLF_Trace(**params)

Bases: topo.base.cf.CFPLearningFn

LearningFn that incorporates a trace of recent activity, not just the current activity.

Based on P. Foldiak (1991), “Learning Invariance from Transformation Sequences”, Neural Computation 3:194-200. Also see Sutton and Barto (1981) and Wallis and Rolls (1997).

Incorporates a decay term to keep the weight vector bounded, and so it does not normally require any output_fn normalization for stability.

NOT YET TESTED.

param ClassSelector single_cf_fn (allow_None=False, constant=False, default=<Hebbian Hebbian01322>, instantiate=True, is_instance=True, pickle_default_value=True, precedence=None, readonly=False)

LearningFn that will be applied to each CF individually.

param Number trace_strength (allow_None=False, bounds=(0.0, 1.0), constant=False, default=0.5, inclusive_bounds=(True, True), instantiate=False, pickle_default_value=True, precedence=None, readonly=False, time_dependent=True, time_fn=<Time Time00001>)

How much the learning is dominated by the activity trace, relative to the current value.

constant_sum_connection_rate(n_units, learning_rate)

Return the learning rate for a single connection assuming that the total rate is to be divided evenly among all the units in the connection field.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07ab9408e8>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07ab940c58>

inspect_value = <functools.partial object at 0x2b07ab940d08>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07ab940db8>

set_param = <functools.partial object at 0x2b07ab940730>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

class topo.learningfn.projfn.CFPLF_OutstarHebbian(**params)

Bases: topo.base.cf.CFPLearningFn

CFPLearningFunction applying the specified (default is Hebbian) single_cf_fn to each CF, where normalization is done in an outstar-manner.

Presumably does not need a separate output_fn for normalization.

NOT YET TESTED.

param ClassSelector single_cf_fn (allow_None=False, constant=False, default=<Hebbian Hebbian01323>, instantiate=True, is_instance=True, pickle_default_value=True, precedence=None, readonly=False)

LearningFn that will be applied to each CF individually.

constant_sum_connection_rate(n_units, learning_rate)

Return the learning rate for a single connection assuming that the total rate is to be divided evenly among all the units in the connection field.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07ab940418>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07ab940578>

inspect_value = <functools.partial object at 0x2b07ab940520>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07ab940470>

set_param = <functools.partial object at 0x2b07ab940b50>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

class topo.learningfn.projfn.HomeoSynaptic(**params)

Bases: topo.base.cf.CFPLearningFn

Learning function using homeostatic synaptic scaling from Sullivan & de Sa, “Homeostatic Synaptic Scaling in Self-Organizing Maps”, Neural Networks (2006), 19(6-7):734-43.

Does not necessarily require output_fn normalization for stability.

param Number activity_target (allow_None=False, bounds=(0, None), constant=False, default=0.1, inclusive_bounds=(True, True), instantiate=False, pickle_default_value=True, precedence=None, readonly=False, time_dependent=True, time_fn=<Time Time00001>)

Target average activity

param Number beta_c (allow_None=False, bounds=(0, None), constant=False, default=0.005, inclusive_bounds=(True, True), instantiate=False, pickle_default_value=True, precedence=None, readonly=False, time_dependent=True, time_fn=<Time Time00001>)

time window over which the neuron’s firing rate is averaged

param ClassSelector single_cf_fn (allow_None=False, constant=False, default=<Hebbian Hebbian01324>, instantiate=True, is_instance=True, pickle_default_value=True, precedence=None, readonly=False)

LearningFn that will be applied to each CF individually

param Number beta_n (allow_None=False, bounds=(0, None), constant=False, default=0.01, inclusive_bounds=(True, True), instantiate=False, pickle_default_value=True, precedence=None, readonly=False, time_dependent=True, time_fn=<Time Time00001>)

homeostatic learning rate

constant_sum_connection_rate(n_units, learning_rate)

Return the learning rate for a single connection assuming that the total rate is to be divided evenly among all the units in the connection field.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07ab940cb0>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07ab940f18>

inspect_value = <functools.partial object at 0x2b07ab940520>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07ab9404c8>

set_param = <functools.partial object at 0x2b07ab940e10>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

class topo.learningfn.projfn.CFPLF_PluginScaled(**params)

Bases: topo.base.cf.CFPLearningFn

CFPLearningFunction applying the specified single_cf_fn to each CF. Scales the single-connection learning rate by a scaling factor that is different for each individual unit. Thus each individual connection field uses a different learning rate.

param ClassSelector single_cf_fn (allow_None=False, constant=False, default=<Hebbian Hebbian01325>, instantiate=True, is_instance=True, pickle_default_value=True, precedence=None, readonly=False)

Accepts a LearningFn that will be applied to each CF individually.

param Parameter learning_rate_scaling_factor (allow_None=True, constant=False, default=None, instantiate=False, pickle_default_value=True, precedence=None, readonly=False)

Matrix of scaling factors for scaling the learning rate of each CF individually.

constant_sum_connection_rate(n_units, learning_rate)

Return the learning rate for a single connection assuming that the total rate is to be divided evenly among all the units in the connection field.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07ab940e10>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07ab940c00>

inspect_value = <functools.partial object at 0x2b07ab940af8>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07ab940d60>

set_param = <functools.partial object at 0x2b07ab940578>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

update_scaling_factor(new_scaling_factor)

Update the single-connection learning rate scaling factor.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

---

som Module

SOM-based learning functions for CFProjections.

class topo.learningfn.som.CFPLF_HebbianSOM(**params)

Bases: topo.learningfn.som.CFPLF_SOM

Hebbian learning rule for CFProjections to Self-Organizing Maps.

This implementation is obsolete and will be removed soon. Please see examples/cfsom_or.ty for current SOM support.

param Number learning_radius (allow_None=False, bounds=None, constant=False, default=0.0, inclusive_bounds=(True, True), instantiate=False, pickle_default_value=True, precedence=None, readonly=False, time_dependent=True, time_fn=<Time Time00001>)

The radius of the neighborhood function to be used for learning. Typically, this value will be set by the Sheet or Projection owning this CFPLearningFn, but it can also be set explicitly by the user.

param Number crop_radius_multiplier (allow_None=False, bounds=None, constant=False, default=3.0, inclusive_bounds=(True, True), instantiate=False, pickle_default_value=True, precedence=None, readonly=False, time_dependent=True, time_fn=<Time Time00001>)

Factor by which the radius should be multiplied, when deciding how far from the winner to keep updating the weights.

param ClassSelector neighborhood_kernel_generator (allow_None=False, constant=False, default=<Gaussian Gaussian01332>, instantiate=True, is_instance=True, pickle_default_value=True, precedence=None, readonly=False)

Neighborhood function

constant_sum_connection_rate(n_units, learning_rate)

Return the learning rate for a single connection assuming that the total rate is to be divided evenly among all the units in the connection field.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07ab1087e0>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07ab108838>

inspect_value = <functools.partial object at 0x2b07ab108890>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07ab108788>

set_param = <functools.partial object at 0x2b07ab108998>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

class topo.learningfn.som.CFPLF_SOM(**params)

Bases: topo.base.cf.CFPLearningFn

An abstract base class of learning functions for Self-Organizing Maps.

This implementation is obsolete and will be removed soon. Please see examples/cfsom_or.ty for current SOM support.

param Number learning_radius (allow_None=False, bounds=None, constant=False, default=0.0, inclusive_bounds=(True, True), instantiate=False, pickle_default_value=True, precedence=None, readonly=False, time_dependent=True, time_fn=<Time Time00001>)

The radius of the neighborhood function to be used for learning. Typically, this value will be set by the Sheet or Projection owning this CFPLearningFn, but it can also be set explicitly by the user.

constant_sum_connection_rate(n_units, learning_rate)

Return the learning rate for a single connection assuming that the total rate is to be divided evenly among all the units in the connection field.

debug(msg, *args, **kw)

Print msg merged with args as a debugging statement.

See Python’s logging module for details of message formatting.

defaults()

Return {parameter_name:parameter.default} for all non-constant Parameters.

Note that a Parameter for which instantiate==True has its default instantiated.

force_new_dynamic_value = <functools.partial object at 0x2b07ab1085d0>

get_param_values(onlychanged=False)

Return a list of name,value pairs for all Parameters of this object.

If onlychanged is True, will only return values that are not equal to the default value.

get_value_generator = <functools.partial object at 0x2b07ab108628>

inspect_value = <functools.partial object at 0x2b07ab108680>

message(msg, *args, **kw)

Print msg merged with args as a message.

See Python’s logging module for details of message formatting.

classmethod params(parameter_name=None)

Return the Parameters of this class as the dictionary {name: parameter_object}

Includes Parameters from this class and its superclasses.

pprint(imports=None, prefix=' ', unknown_value='<?>', qualify=False, separator='')

(Experimental) Pretty printed representation that may be evaluated with eval. See pprint() function for more details.

classmethod print_param_defaults()

Print the default values of all cls’s Parameters.

print_param_values()

Print the values of all this object’s Parameters.

script_repr(imports=, []prefix=' ')

Variant of __repr__ designed for generating a runnable script.

classmethod set_default(param_name, value)

Set the default value of param_name.

Equivalent to setting param_name on the class.

set_dynamic_time_fn = <functools.partial object at 0x2b07ab108998>

set_param = <functools.partial object at 0x2b07ab1088e8>

state_pop()

Restore the most recently saved state.

See state_push() for more details.

state_push()

Save this instance’s state.

For Parameterized instances, this includes the state of dynamically generated values.

Subclasses that maintain short-term state should additionally save and restore that state using state_push() and state_pop().

Generally, this method is used by operations that need to test something without permanently altering the objects’ state.

verbose(msg, *args, **kw)

Print msg merged with args as a verbose message.

See Python’s logging module for details of message formatting.

warning(msg, *args, **kw)

Print msg merged with args as a warning, unless module variable warnings_as_exceptions is True, then raise an Exception containing the arguments.

See Python’s logging module for details of message formatting.

---