pygenn.genn_model Namespace Reference

This module provides the GeNNModel class to simplify working with pygenn module and helper functions to derive custom model classes. More...

Classes
class	GeNNModel
	GeNNModel class This class helps to define, build and run a GeNN model from python. More...

Functions
def	init_var (init_var_snippet, param_space)
	This helper function creates a VarInit object to easily initialise a variable using a snippet. More...
def	init_connectivity (init_sparse_connect_snippet, param_space)
	This helper function creates a InitSparseConnectivitySnippet::Init object to easily initialise connectivity using a snippet. More...
def	init_toeplitz_connectivity (init_toeplitz_connect_snippet, param_space)
	This helper function creates a InitToeplitzConnectivitySnippet::Init object to easily initialise connectivity using a snippet. More...
def	create_var_ref (pop, var_name)
	This helper function creates a Models::VarReference pointing to a neuron or current source variable for initialising variable references. More...
def	create_psm_var_ref (sg, var_name)
	This helper function creates a Models::VarReference pointing to a postsynaptic model variable for initialising variable references. More...
def	create_wu_pre_var_ref (sg, var_name)
	This helper function creates a Models::VarReference pointing to a presynaptic weight update model variable for initialising variable references. More...
def	create_wu_post_var_ref (sg, var_name)
	This helper function creates a Models::VarReference pointing to a postsynaptic weight update model variable for initialising variable references. More...
def	create_wu_var_ref (g, var_name, tp_sg=None, tp_var_name=None)
	This helper function creates a Models::WUVarReference pointing to a weight update model variable for initialising variable references. More...
def	create_egp_ref (pop, egp_name)
	This helper function creates a Models::EGPReference pointing to a neuron, current source or custom update extra global parameter for initialising references. More...
def	create_psm_egp_ref (sg, egp_name)
	This helper function creates a Models::EGPReference pointing to a postsynaptic model extra global parameter for initialising references. More...
def	create_wu_egp_ref (sg, egp_name)
	This helper function creates a Models::EGPReference pointing to a weight update model extra global parameter for initialising references. More...
def	create_custom_neuron_class (class_name, param_names=None, var_name_types=None, derived_params=None, sim_code=None, threshold_condition_code=None, reset_code=None, support_code=None, extra_global_params=None, additional_input_vars=None, is_auto_refractory_required=None, custom_body=None)
	This helper function creates a custom NeuronModel class. More...
def	create_custom_postsynaptic_class (class_name, param_names=None, var_name_types=None, derived_params=None, decay_code=None, apply_input_code=None, support_code=None, extra_global_params=None, custom_body=None)
	This helper function creates a custom PostsynapticModel class. More...
def	create_custom_weight_update_class (class_name, param_names=None, var_name_types=None, pre_var_name_types=None, post_var_name_types=None, derived_params=None, sim_code=None, event_code=None, learn_post_code=None, synapse_dynamics_code=None, event_threshold_condition_code=None, pre_spike_code=None, post_spike_code=None, pre_dynamics_code=None, post_dynamics_code=None, sim_support_code=None, learn_post_support_code=None, synapse_dynamics_suppport_code=None, extra_global_params=None, is_pre_spike_time_required=None, is_post_spike_time_required=None, is_pre_spike_event_time_required=None, is_prev_pre_spike_time_required=None, is_prev_post_spike_time_required=None, is_prev_pre_spike_event_time_required=None, custom_body=None)
	This helper function creates a custom WeightUpdateModel class. More...
def	create_custom_current_source_class (class_name, param_names=None, var_name_types=None, derived_params=None, injection_code=None, extra_global_params=None, custom_body=None)
	This helper function creates a custom NeuronModel class. More...
def	create_custom_custom_update_class (class_name, param_names=None, var_name_types=None, derived_params=None, var_refs=None, update_code=None, extra_global_params=None, extra_global_param_refs=None, custom_body=None)
	This helper function creates a custom CustomUpdate class. More...
def	create_custom_model_class (class_name, base, param_names, var_name_types, derived_params, extra_global_params, custom_body)
	This helper function completes a custom model class creation. More...
def	create_dpf_class (dp_func)
	Helper function to create derived parameter function class. More...
def	create_cmlf_class (cml_func)
	Helper function to create function class for calculating sizes of matrices initialised with sparse connectivity initialisation snippet. More...
def	create_cksf_class (cks_func)
	Helper function to create function class for calculating sizes of kernels from connectivity initialiser parameters. More...
def	create_custom_init_var_snippet_class (class_name, param_names=None, derived_params=None, var_init_code=None, extra_global_params=None, custom_body=None)
	This helper function creates a custom InitVarSnippet class. More...
def	create_custom_sparse_connect_init_snippet_class (class_name, param_names=None, derived_params=None, row_build_code=None, row_build_state_vars=None, col_build_code=None, col_build_state_vars=None, calc_max_row_len_func=None, calc_max_col_len_func=None, calc_kernel_size_func=None, extra_global_params=None, custom_body=None)
	This helper function creates a custom InitSparseConnectivitySnippet class. More...
def	create_custom_toeplitz_connect_init_snippet_class (class_name, param_names=None, derived_params=None, diagonal_build_code=None, diagonal_build_state_vars=None, calc_max_row_len_func=None, calc_kernel_size_func=None, extra_global_params=None, custom_body=None)
	This helper function creates a custom InitToeplitzConnectivitySnippet class. More...

Variables
	backend_modules = OrderedDict()
	m = import_module(".genn_wrapper." + b + "Backend", "pygenn")
	GeNNType = namedtuple("GeNNType", ["np_dtype", "assign_ext_ptr_array", "assign_ext_ptr_single"])

Detailed Description

This module provides the GeNNModel class to simplify working with pygenn module and helper functions to derive custom model classes.

GeNNModel should be used to configure a model, build, load and finally run it. Recording is done manually by pulling from the population of interest and then copying the values from Variable.view attribute. Each simulation step must be triggered manually by calling step_time function.

Examples

The following example shows in a (very) simplified manner how to build and
run a simulation using GeNNModel::

    from pygenn.genn_model import GeNNModel
    gm = GeNNModel("float", "test")

    # add populations
    neuron_pop = gm.add_neuron_population(_parameters_truncated_)
    syn_pop = gm.add_synapse_population(_parameters_truncated_)

    # build and load model
    gm.build()
    gm.load()

    Vs = numpy.empty((simulation_length, population_size))
    # Variable.view provides a view into a raw C array
    # here a Variable call V (voltage) will be recorded
    v_view = neuron_pop.vars["V"].view

    # run a simulation for 1000 steps
    for i in range 1000:
        # manually trigger one simulation step
        gm.step_time()
        # when you pull state from device, views of all variables
        # are updated and show current simulated values
        neuron_pop.pull_state_from_device()
        # finally, record voltage by copying form view into array.
        Vs[i,:] = v_view

Function Documentation

create_cksf_class()

def pygenn.genn_model.create_cksf_class

(

cks_func

)

Helper function to create function class for calculating sizes of kernels from connectivity initialiser parameters.

Parameters

cks_func

a function which computes the kernel size and takes one arg "pars" (vector of double)

create_cmlf_class()

def pygenn.genn_model.create_cmlf_class

(

cml_func

)

Helper function to create function class for calculating sizes of matrices initialised with sparse connectivity initialisation snippet.

Parameters

cml_func

a function which computes the length and takes three args "num_pre" (unsigned int), "num_post" (unsigned int) and "pars" (vector of double)

create_custom_current_source_class()

def pygenn.genn_model.create_custom_current_source_class	(		class_name,
			param_names = None,
			var_name_types = None,
			derived_params = None,
			injection_code = None,
			extra_global_params = None,
			custom_body = None
	)

This helper function creates a custom NeuronModel class.

See also

create_custom_neuron_class

create_custom_weight_update_class

create_custom_current_source_class

create_custom_init_var_snippet_class

create_custom_sparse_connect_init_snippet_class

Parameters

class_name	name of the new class
param_names	list of strings with param names of the model
var_name_types	list of pairs of strings with varible names and types of the model
derived_params	list of pairs, where the first member is string with name of the derived parameter and the second should be a functor returned by create_dpf_class
injection_code	string with the current injection code
extra_global_params	list of pairs of strings with names and types of additional parameters
custom_body	dictionary with additional attributes and methods of the new class

create_custom_custom_update_class()

def pygenn.genn_model.create_custom_custom_update_class	(		class_name,
			param_names = None,
			var_name_types = None,
			derived_params = None,
			var_refs = None,
			update_code = None,
			extra_global_params = None,
			extra_global_param_refs = None,
			custom_body = None
	)

This helper function creates a custom CustomUpdate class.

See also

create_custom_neuron_class

create_custom_weight_update_class

create_custom_current_source_class

create_custom_init_var_snippet_class

create_custom_sparse_connect_init_snippet_class

Parameters

class_name	name of the new class
param_names	list of strings with param names of the model
var_name_types	list of tuples of strings with varible names and types of the variable
derived_params	list of tuples, where the first member is string with name of the derived parameter and the second should be a functor returned by create_dpf_class
var_refs	list of tuples of strings with varible names and types of variabled variable
update_code	string with the current injection code
extra_global_params	list of pairs of strings with names and types of additional parameters
extra_global_param_refs	list of pairs of strings with names and types of extra global parameter references
custom_body	dictionary with additional attributes and methods of the new class

create_custom_init_var_snippet_class()

def pygenn.genn_model.create_custom_init_var_snippet_class	(		class_name,
			param_names = None,
			derived_params = None,
			var_init_code = None,
			extra_global_params = None,
			custom_body = None
	)

This helper function creates a custom InitVarSnippet class.

See also

create_custom_neuron_class

create_custom_weight_update_class

create_custom_postsynaptic_class

create_custom_current_source_class

create_custom_sparse_connect_init_snippet_class

Parameters

class_name	name of the new class
param_names	list of strings with param names of the model
derived_params	list of pairs, where the first member is string with name of the derived parameter and the second MUST be an instance of the pygenn.genn_wrapper.DerivedParamFunc class
var_init_code	string with the variable initialization code
extra_global_params	list of pairs of strings with names and types of additional parameters
custom_body	dictionary with additional attributes and methods of the new class

create_custom_model_class()

def pygenn.genn_model.create_custom_model_class	(		class_name,
			base,
			param_names,
			var_name_types,
			derived_params,
			extra_global_params,
			custom_body
	)

This helper function completes a custom model class creation.

This part is common for all model classes and is nearly useless on its own unless you specify custom_body.

See also

create_custom_neuron_class

create_custom_weight_update_class

create_custom_postsynaptic_class

create_custom_current_source_class

create_custom_init_var_snippet_class

create_custom_sparse_connect_init_snippet_class

Parameters

class_name	name of the new class
base	base class
param_names	list of strings with param names of the model
var_name_types	list of pairs of strings with varible names and types of the model
derived_params	list of pairs, where the first member is string with name of the derived parameter and the second should be a functor returned by create_dpf_class
extra_global_params	list of pairs of strings with names and types of additional parameters
custom_body	dictionary with attributes and methods of the new class

create_custom_neuron_class()

def pygenn.genn_model.create_custom_neuron_class	(		class_name,
			param_names = None,
			var_name_types = None,
			derived_params = None,
			sim_code = None,
			threshold_condition_code = None,
			reset_code = None,
			support_code = None,
			extra_global_params = None,
			additional_input_vars = None,
			is_auto_refractory_required = None,
			custom_body = None
	)

This helper function creates a custom NeuronModel class.

See also

create_custom_postsynaptic_class

create_custom_weight_update_class

create_custom_current_source_class

create_custom_init_var_snippet_class

create_custom_sparse_connect_init_snippet_class

Parameters

class_name	name of the new class
param_names	list of strings with param names of the model
var_name_types	list of pairs of strings with varible names and types of the model
derived_params	list of pairs, where the first member is string with name of the derived parameter and the second should be a functor returned by create_dpf_class
sim_code	string with the simulation code
threshold_condition_code	string with the threshold condition code
reset_code	string with the reset code
support_code	string with the support code
extra_global_params	list of pairs of strings with names and types of additional parameters
additional_input_vars	list of tuples with names and types as strings and initial values of additional local input variables
is_auto_refractory_required	does this model require auto-refractory logic to be generated?
custom_body	dictionary with additional attributes and methods of the new class

create_custom_postsynaptic_class()

def pygenn.genn_model.create_custom_postsynaptic_class	(		class_name,
			param_names = None,
			var_name_types = None,
			derived_params = None,
			decay_code = None,
			apply_input_code = None,
			support_code = None,
			extra_global_params = None,
			custom_body = None
	)

This helper function creates a custom PostsynapticModel class.

See also

create_custom_neuron_class

create_custom_weight_update_class

create_custom_current_source_class

create_custom_init_var_snippet_class

create_custom_sparse_connect_init_snippet_class

Parameters

class_name	name of the new class
param_names	list of strings with param names of the model
var_name_types	list of pairs of strings with varible names and types of the model
derived_params	list of pairs, where the first member is string with name of the derived parameter and the second should be a functor returned by create_dpf_class
decay_code	string with the decay code
apply_input_code	string with the apply input code
support_code	string with the support code
extra_global_params	list of pairs of strings with names and types of additional parameters
custom_body	dictionary with additional attributes and methods of the new class

create_custom_sparse_connect_init_snippet_class()

def pygenn.genn_model.create_custom_sparse_connect_init_snippet_class	(		class_name,
			param_names = None,
			derived_params = None,
			row_build_code = None,
			row_build_state_vars = None,
			col_build_code = None,
			col_build_state_vars = None,
			calc_max_row_len_func = None,
			calc_max_col_len_func = None,
			calc_kernel_size_func = None,
			extra_global_params = None,
			custom_body = None
	)

This helper function creates a custom InitSparseConnectivitySnippet class.

See also

create_custom_neuron_class

create_custom_weight_update_class

create_custom_postsynaptic_class

create_custom_current_source_class

create_custom_init_var_snippet_class

Parameters

class_name	name of the new class
param_names	list of strings with param names of the model
derived_params	list of pairs, where the first member is string with name of the derived parameter and the second MUST be an instance of the class which inherits from pygenn.genn_wrapper.DerivedParamFunc
row_build_code	string with row building initialization code
row_build_state_vars	list of tuples of state variables, their types and their initial values to use across row building loop
col_build_code	string with column building initialization code
col_build_state_vars	list of tuples of state variables, their types and their initial values to use across column building loop
calc_max_row_len_func	instance of class inheriting from CalcMaxLengthFunc used to calculate maximum row length of synaptic matrix
calc_max_col_len_func	instance of class inheriting from CalcMaxLengthFunc used to calculate maximum col length of synaptic matrix
calc_kernel_size_func	instance of class inheriting from CalcKernelSizeFunc used to calculate kernel dimensions
extra_global_params	list of pairs of strings with names and types of additional parameters
custom_body	dictionary with additional attributes and methods of the new class

create_custom_toeplitz_connect_init_snippet_class()

def pygenn.genn_model.create_custom_toeplitz_connect_init_snippet_class	(		class_name,
			param_names = None,
			derived_params = None,
			diagonal_build_code = None,
			diagonal_build_state_vars = None,
			calc_max_row_len_func = None,
			calc_kernel_size_func = None,
			extra_global_params = None,
			custom_body = None
	)

This helper function creates a custom InitToeplitzConnectivitySnippet class.

See also

create_custom_neuron_class

create_custom_weight_update_class

create_custom_postsynaptic_class

create_custom_current_source_class

create_custom_init_var_snippet_class

Parameters

class_name	name of the new class
param_names	list of strings with param names of the model
derived_params	list of pairs, where the first member is string with name of the derived parameter and the second MUST be an instance of the class which inherits from pygenn.genn_wrapper.DerivedParamFunc
diagonal_build_code	string with diagonal building initialization code
diagonal_build_state_vars	list of tuples of state variables, their types and their initial values to use across diagonal building loop
calc_max_row_len_func	instance of class inheriting from CalcMaxLengthFunc used to calculate maximum row length of synaptic matrix
calc_kernel_size_func	instance of class inheriting from CalcKernelSizeFunc used to calculate kernel dimensions
extra_global_params	list of pairs of strings with names and types of additional parameters
custom_body	dictionary with additional attributes and methods of the new class

create_custom_weight_update_class()

def pygenn.genn_model.create_custom_weight_update_class	(		class_name,
			param_names = None,
			var_name_types = None,
			pre_var_name_types = None,
			post_var_name_types = None,
			derived_params = None,
			sim_code = None,
			event_code = None,
			learn_post_code = None,
			synapse_dynamics_code = None,
			event_threshold_condition_code = None,
			pre_spike_code = None,
			post_spike_code = None,
			pre_dynamics_code = None,
			post_dynamics_code = None,
			sim_support_code = None,
			learn_post_support_code = None,
			synapse_dynamics_suppport_code = None,
			extra_global_params = None,
			is_pre_spike_time_required = None,
			is_post_spike_time_required = None,
			is_pre_spike_event_time_required = None,
			is_prev_pre_spike_time_required = None,
			is_prev_post_spike_time_required = None,
			is_prev_pre_spike_event_time_required = None,
			custom_body = None
	)

This helper function creates a custom WeightUpdateModel class.

See also

create_custom_neuron_class

create_custom_postsynaptic_class

create_custom_current_source_class

create_custom_init_var_snippet_class

create_custom_sparse_connect_init_snippet_class

Parameters

class_name	name of the new class
param_names	list of strings with param names of the model
var_name_types	list of pairs of strings with variable names and types of the model
pre_var_name_types	list of pairs of strings with presynaptic variable names and types of the model
post_var_name_types	list of pairs of strings with postsynaptic variable names and types of the model
derived_params	list of pairs, where the first member is string with name of the derived parameter and the second should be a functor returned by create_dpf_class
sim_code	string with the simulation code
event_code	string with the event code
learn_post_code	string with the code to include in learn_synapse_post kernel/function
synapse_dynamics_code	string with the synapse dynamics code
event_threshold_condition_code	string with the event threshold condition code
pre_spike_code	string with the code run once per spiking presynaptic neuron
post_spike_code	string with the code run once per spiking postsynaptic neuron
pre_dynamics_code	string with the code run every timestep on presynaptic neuron
post_dynamics_code	string with the code run every timestep on postsynaptic neuron
sim_support_code	string with simulation support code
learn_post_support_code	string with support code for learn_synapse_post kernel/function
synapse_dynamics_suppport_code	string with synapse dynamics support code
extra_global_params	list of pairs of strings with names and types of additional parameters
is_pre_spike_time_required	boolean, is presynaptic spike time required in any weight update kernels?
is_post_spike_time_required	boolean, is postsynaptic spike time required in any weight update kernels?
is_pre_spike_event_time_required	boolean, is presynaptic spike-like-event time required in any weight update kernels?
is_prev_pre_spike_time_required	boolean, is previous presynaptic spike time required in any weight update kernels?
is_prev_post_spike_time_required	boolean, is previous postsynaptic spike time required in any weight update kernels?
is_prev_pre_spike_event_time_required	boolean, is previous presynaptic spike-like-event time required in any weight update kernels?
custom_body	dictionary with additional attributes and methods of the new class

create_dpf_class()

def pygenn.genn_model.create_dpf_class

(

dp_func

)

Helper function to create derived parameter function class.

Parameters

dp_func

a function which computes the derived parameter and takes two args "pars" (vector of double) and "dt" (double)

create_egp_ref()

def pygenn.genn_model.create_egp_ref	(		pop,
			egp_name
	)

This helper function creates a Models::EGPReference pointing to a neuron, current source or custom update extra global parameter for initialising references.

Parameters

pop	population, either a NeuronGroup or CurrentSource object
egp_name	name of extra global parameter in population to reference

create_psm_egp_ref()

def pygenn.genn_model.create_psm_egp_ref	(		sg,
			egp_name
	)

This helper function creates a Models::EGPReference pointing to a postsynaptic model extra global parameter for initialising references.

Parameters

sg	SynapseGroup object
egp_name	name of postsynaptic model extra global parameter in synapse group to reference

create_psm_var_ref()

def pygenn.genn_model.create_psm_var_ref	(		sg,
			var_name
	)

This helper function creates a Models::VarReference pointing to a postsynaptic model variable for initialising variable references.

Parameters

sg	SynapseGroup object
var_name	name of postsynaptic model variable in synapse group to reference

create_var_ref()

def pygenn.genn_model.create_var_ref	(		pop,
			var_name
	)

This helper function creates a Models::VarReference pointing to a neuron or current source variable for initialising variable references.

Parameters

pop	population, either a NeuronGroup or CurrentSource object
var_name	name of variable in population to reference

create_wu_egp_ref()

def pygenn.genn_model.create_wu_egp_ref	(		sg,
			egp_name
	)

This helper function creates a Models::EGPReference pointing to a weight update model extra global parameter for initialising references.

Parameters

sg	SynapseGroup object
egp_name	name of weight update model extra global parameter in synapse group to reference

create_wu_post_var_ref()

def pygenn.genn_model.create_wu_post_var_ref	(		sg,
			var_name
	)

This helper function creates a Models::VarReference pointing to a postsynaptic weight update model variable for initialising variable references.

Parameters

sg	SynapseGroup object
var_name	name of postsynaptic weight update model variable in synapse group to reference

create_wu_pre_var_ref()

def pygenn.genn_model.create_wu_pre_var_ref	(		sg,
			var_name
	)

This helper function creates a Models::VarReference pointing to a presynaptic weight update model variable for initialising variable references.

Parameters

sg	SynapseGroup object
var_name	name of presynaptic weight update model variable in synapse group to reference

create_wu_var_ref()

def pygenn.genn_model.create_wu_var_ref	(		g,
			var_name,
			tp_sg = None,
			tp_var_name = None
	)

This helper function creates a Models::WUVarReference pointing to a weight update model variable for initialising variable references.

Parameters

g	SynapseGroup or CustomUpdate object
var_name	name of weight update model variable in synapse group to reference
tp_sg	(optional) SynapseGroup object to copy transpose of variable to
tp_var_name	(optional) name of weight update model variable in tranpose synapse group to copy transpose to

init_connectivity()

def pygenn.genn_model.init_connectivity	(		init_sparse_connect_snippet,
			param_space
	)

This helper function creates a InitSparseConnectivitySnippet::Init object to easily initialise connectivity using a snippet.

Parameters

init_sparse_connect_snippet	type of the InitSparseConnectivitySnippet class as string or instance of class derived from InitSparseConnectivitySnippet::Custom.
param_space	dict with param values for the InitSparseConnectivitySnippet class

init_toeplitz_connectivity()

def pygenn.genn_model.init_toeplitz_connectivity	(		init_toeplitz_connect_snippet,
			param_space
	)

This helper function creates a InitToeplitzConnectivitySnippet::Init object to easily initialise connectivity using a snippet.

Parameters

init_toeplitz_connect_snippet	type of the InitToeplitzConnectivitySnippet class as string or instance of class derived from InitSparseConnectivitySnippet::Custom.
param_space	dict with param values for the InitToeplitzConnectivitySnippet class

init_var()

def pygenn.genn_model.init_var	(		init_var_snippet,
			param_space
	)

This helper function creates a VarInit object to easily initialise a variable using a snippet.

Parameters

init_var_snippet	type of the InitVarSnippet class as string or instance of class derived from InitVarSnippet::Custom class.
param_space	dict with param values for the InitVarSnippet class

Variable Documentation

backend_modules

pygenn.genn_model.backend_modules = OrderedDict()

GeNNType

pygenn.genn_model.GeNNType = namedtuple("GeNNType", ["np_dtype", "assign_ext_ptr_array", "assign_ext_ptr_single"])

m

pygenn.genn_model.m = import_module(".genn_wrapper." + b + "Backend", "pygenn")