pygenn.genn_model.GeNNModel Class Reference

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

Inheritance diagram for pygenn.genn_model.GeNNModel:

Public Member Functions
def	__init__ (self, precision="float", model_name="GeNNModel", backend=None, time_precision=None, genn_log_level=genn_wrapper.warning, code_gen_log_level=genn_wrapper.warning, backend_log_level=genn_wrapper.warning, preference_kwargs)
	Init GeNNModel. More...
def	use_backend (self)
def	use_backend (self, backend)
def	timing_enabled (self)
def	timing_enabled (self, timing)
def	batch_size (self)
def	batch_size (self, batch_size)
def	default_var_location (self)
	Default variable location - defines where state variables are initialised. More...
def	default_var_location (self, location)
def	default_sparse_connectivity_location (location)
	Default sparse connectivity mode - where connectivity is initialised. More...
def	default_sparse_connectivity_location (self, location)
def	model_name (self)
	Name of the model. More...
def	model_name (self, model_name)
def	t (self)
	Simulation time in ms. More...
def	t (self, t)
def	timestep (self)
	Simulation time step. More...
def	timestep (self, timestep)
def	free_device_mem_bytes (self)
def	dT (self)
	Step size. More...
def	dT (self, dt)
def	neuron_update_time (self)
def	init_time (self)
def	presynaptic_update_time (self)
def	postsynaptic_update_time (self)
def	synapse_dynamics_time (self)
def	init_sparse_time (self)
def	get_custom_update_time (self, name)
def	get_custom_update_transpose_time (self, name)
def	add_neuron_population (self, pop_name, num_neurons, neuron, param_space, var_space)
	Add a neuron population to the GeNN model. More...
def	add_synapse_population (self, pop_name, matrix_type, delay_steps, source, target, w_update_model, wu_param_space, wu_var_space, wu_pre_var_space, wu_post_var_space, postsyn_model, ps_param_space, ps_var_space, connectivity_initialiser=None)
	Add a synapse population to the GeNN model. More...
def	add_slave_synapse_population (self, pop_name, master_pop, delay_steps, source, target, postsyn_model, ps_param_space, ps_var_space)
	Add a 'slave' population to the GeNN model which shares weights and connectivity with a 'master' population. More...
def	add_current_source (self, cs_name, current_source_model, pop, param_space, var_space)
	Add a current source to the GeNN model. More...
def	add_custom_update (self, cu_name, group_name, custom_update_model, param_space, var_space, var_ref_space, egp_ref_space={})
	Add a current source to the GeNN model. More...
def	build (self, path_to_model="./", force_rebuild=False)
	Finalize and build a GeNN model. More...
def	load (self, path_to_model="./", num_recording_timesteps=None)
	import the model as shared library and initialize it More...
def	unload (self)
def	reinitialise (self)
	reinitialise model to its original state without re-loading More...
def	step_time (self)
	Make one simulation step. More...
def	custom_update (self, name)
	Perform custom update. More...
def	pull_state_from_device (self, pop_name)
	Pull state from the device for a given population. More...
def	pull_spikes_from_device (self, pop_name)
	Pull spikes from the device for a given population. More...
def	pull_spike_events_from_device (self, pop_name)
	Pull spike events from the device for a given population. More...
def	pull_current_spikes_from_device (self, pop_name)
	Pull spikes from the device for a given population. More...
def	pull_current_spike_events_from_device (self, pop_name)
	Pull spike events from the device for a given population. More...
def	pull_connectivity_from_device (self, pop_name)
	Pull connectivity from the device for a given population. More...
def	pull_var_from_device (self, pop_name, var_name)
	Pull variable from the device for a given population. More...
def	pull_extra_global_param_from_device (self, pop_name, egp_name, size=None)
	Pull extra global parameter from the device for a given population. More...
def	push_state_to_device (self, pop_name)
	Push state to the device for a given population. More...
def	push_spikes_to_device (self, pop_name)
	Push spikes to the device for a given population. More...
def	push_spike_events_to_device (self, pop_name)
	Push spike events to the device for a given population. More...
def	push_current_spikes_to_device (self, pop_name)
	Push current spikes to the device for a given population. More...
def	push_current_spike_events_to_device (self, pop_name)
	Push current spike events to the device for a given population. More...
def	push_connectivity_to_device (self, pop_name)
	Push connectivity to the device for a given population. More...
def	push_var_to_device (self, pop_name, var_name)
	Push variable to the device for a given population. More...
def	push_extra_global_param_to_device (self, pop_name, egp_name, size=None)
	Push extra global parameter to the device for a given population. More...
def	pull_recording_buffers_from_device (self)
	Pull recording buffers from device. More...
def	end (self)
	Free memory. More...

Public Attributes
	use_backend
	backend_log_level
	default_var_location
	model_name
	neuron_populations
	synapse_populations
	current_sources
	custom_updates
	dT
	genn_types

Detailed Description

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

Constructor & Destructor Documentation

__init__()

def pygenn.genn_model.GeNNModel.__init__	(		self,
			precision = "float",
			model_name = "GeNNModel",
			backend = None,
			time_precision = None,
			genn_log_level = genn_wrapper.warning,
			code_gen_log_level = genn_wrapper.warning,
			backend_log_level = genn_wrapper.warning,
			preference_kwargs
	)

Init GeNNModel.

Parameters

precision	string precision as string ("float", "double" or "long double"). defaults to float.
model_name	string name of the model. Defaults to "GeNNModel".
backend	string specifying name of backend module to use Defaults to None to pick 'best' backend for your system
time_precision	string time precision as string ("float", "double" or "long double"). defaults to float.
genn_log_level	Log level for GeNN
code_gen_log_level	Log level for GeNN code-generator
backend_log_level	Log level for backend
preference_kwargs	Additional keyword arguments to set in backend preferences structure

Member Function Documentation

add_current_source()

def pygenn.genn_model.GeNNModel.add_current_source	(		self,
			cs_name,
			current_source_model,
			pop,
			param_space,
			var_space
	)

Add a current source to the GeNN model.

Parameters

cs_name	name of the new current source
current_source_model	type of the CurrentSourceModels class as string or instance of CurrentSourceModels class derived from pygenn.genn_wrapper.CurrentSourceModels.Custom (see also pygenn.genn_model.create_custom_current_source_class)
pop	population into which the current source should be injected (either name or NeuronGroup object)
param_space	dict with param values for the CurrentSourceModels class
var_space	dict with initial variable values for the CurrentSourceModels class

add_custom_update()

def pygenn.genn_model.GeNNModel.add_custom_update	(		self,
			cu_name,
			group_name,
			custom_update_model,
			param_space,
			var_space,
			var_ref_space,
			egp_ref_space = {}
	)

Add a current source to the GeNN model.

Parameters

cu_name	name of the new current source
group_name	name of
custom_update_model	type of the CustomUpdateModel class as string or instance of CustomUpdateModel class derived from pygenn.genn_wrapper.CustomUpdateModel.Custom (see also pygenn.genn_model.create_custom_custom_update_class)
param_space	dict with param values for the CustomUpdateModel class
var_space	dict with initial variable values for the CustomUpdateModel class
var_ref_space	dict with variable references for the CustomUpdateModel class
egp_ref_space	dict with extra global parameter references for the CustomUpdateModel class

add_neuron_population()

def pygenn.genn_model.GeNNModel.add_neuron_population	(		self,
			pop_name,
			num_neurons,
			neuron,
			param_space,
			var_space
	)

Add a neuron population to the GeNN model.

Parameters

pop_name	name of the new population
num_neurons	number of neurons in the new population
neuron	type of the NeuronModels class as string or instance of neuron class derived from pygenn.genn_wrapper.NeuronModels.Custom (see also pygenn.genn_model.create_custom_neuron_class)
param_space	dict with param values for the NeuronModels class
var_space	dict with initial variable values for the NeuronModels class

add_slave_synapse_population()

def pygenn.genn_model.GeNNModel.add_slave_synapse_population	(		self,
			pop_name,
			master_pop,
			delay_steps,
			source,
			target,
			postsyn_model,
			ps_param_space,
			ps_var_space
	)

Add a 'slave' population to the GeNN model which shares weights and connectivity with a 'master' population.

Parameters

pop_name	name of the new population
master_pop	master synapse group to share weights with (either name or SynapseGroup object)
delay_steps	delay in number of steps
source	source neuron group (either name or NeuronGroup object)
target	target neuron group (either name or NeuronGroup object)
postsyn_model	type of the PostsynapticModels class as string or instance of postsynaptic model class derived from pygenn.genn_wrapper.PostsynapticModels.Custom (see also pygenn.genn_model.create_custom_postsynaptic_class)
ps_param_space	dict with param values for the PostsynapticModels class
ps_var_space	dict with initial variable values for the PostsynapticModels class

add_synapse_population()

def pygenn.genn_model.GeNNModel.add_synapse_population	(		self,
			pop_name,
			matrix_type,
			delay_steps,
			source,
			target,
			w_update_model,
			wu_param_space,
			wu_var_space,
			wu_pre_var_space,
			wu_post_var_space,
			postsyn_model,
			ps_param_space,
			ps_var_space,
			connectivity_initialiser = None
	)

Add a synapse population to the GeNN model.

Parameters

pop_name	name of the new population
matrix_type	type of the matrix as string
delay_steps	delay in number of steps
source	source neuron group (either name or NeuronGroup object)
target	target neuron group (either name or NeuronGroup object)
w_update_model	type of the WeightUpdateModels class as string or instance of weight update model class derived from pygenn.genn_wrapper.WeightUpdateModels.Custom (see also pygenn.genn_model.create_custom_weight_update_class)
wu_param_space	dict with param values for the WeightUpdateModels class
wu_var_space	dict with initial values for WeightUpdateModels state variables
wu_pre_var_space	dict with initial values for WeightUpdateModels presynaptic variables
wu_post_var_space	dict with initial values for WeightUpdateModels postsynaptic variables
postsyn_model	type of the PostsynapticModels class as string or instance of postsynaptic model class derived from pygenn.genn_wrapper.PostsynapticModels.Custom (see also pygenn.genn_model.create_custom_postsynaptic_class)
ps_param_space	dict with param values for the PostsynapticModels class
ps_var_space	dict with initial variable values for the PostsynapticModels class
connectivity_initialiser	InitSparseConnectivitySnippet::Init for connectivity

batch_size() [1/2]

def pygenn.genn_model.GeNNModel.batch_size

(

self

)

batch_size() [2/2]

def pygenn.genn_model.GeNNModel.batch_size	(		self,
			batch_size
	)

build()

def pygenn.genn_model.GeNNModel.build	(		self,
			path_to_model = "./",
			force_rebuild = False
	)

Finalize and build a GeNN model.

Parameters

path_to_model	path where to place the generated model code. Defaults to the local directory.
force_rebuild	should model be rebuilt even if it doesn't appear to be required

custom_update()

def pygenn.genn_model.GeNNModel.custom_update	(		self,
			name
	)

Perform custom update.

default_sparse_connectivity_location() [1/2]

def pygenn.genn_model.GeNNModel.default_sparse_connectivity_location

(

location

)

Default sparse connectivity mode - where connectivity is initialised.

default_sparse_connectivity_location() [2/2]

def pygenn.genn_model.GeNNModel.default_sparse_connectivity_location	(		self,
			location
	)

default_var_location() [1/2]

def pygenn.genn_model.GeNNModel.default_var_location

(

self

)

Default variable location - defines where state variables are initialised.

default_var_location() [2/2]

def pygenn.genn_model.GeNNModel.default_var_location	(		self,
			location
	)

dT() [1/2]

def pygenn.genn_model.GeNNModel.dT

(

self

)

Step size.

dT() [2/2]

def pygenn.genn_model.GeNNModel.dT	(		self,
			dt
	)

end()

def pygenn.genn_model.GeNNModel.end

(

self

)

Free memory.

free_device_mem_bytes()

def pygenn.genn_model.GeNNModel.free_device_mem_bytes

(

self

)

get_custom_update_time()

def pygenn.genn_model.GeNNModel.get_custom_update_time	(		self,
			name
	)

get_custom_update_transpose_time()

def pygenn.genn_model.GeNNModel.get_custom_update_transpose_time	(		self,
			name
	)

init_sparse_time()

def pygenn.genn_model.GeNNModel.init_sparse_time

(

self

)

init_time()

def pygenn.genn_model.GeNNModel.init_time

(

self

)

load()

def pygenn.genn_model.GeNNModel.load	(		self,
			path_to_model = "./",
			num_recording_timesteps = None
	)

import the model as shared library and initialize it

model_name() [1/2]

def pygenn.genn_model.GeNNModel.model_name

(

self

)

Name of the model.

model_name() [2/2]

def pygenn.genn_model.GeNNModel.model_name	(		self,
			model_name
	)

neuron_update_time()

def pygenn.genn_model.GeNNModel.neuron_update_time

(

self

)

postsynaptic_update_time()

def pygenn.genn_model.GeNNModel.postsynaptic_update_time

(

self

)

presynaptic_update_time()

def pygenn.genn_model.GeNNModel.presynaptic_update_time

(

self

)

pull_connectivity_from_device()

def pygenn.genn_model.GeNNModel.pull_connectivity_from_device	(		self,
			pop_name
	)

Pull connectivity from the device for a given population.

pull_current_spike_events_from_device()

def pygenn.genn_model.GeNNModel.pull_current_spike_events_from_device	(		self,
			pop_name
	)

Pull spike events from the device for a given population.

pull_current_spikes_from_device()

def pygenn.genn_model.GeNNModel.pull_current_spikes_from_device	(		self,
			pop_name
	)

Pull spikes from the device for a given population.

pull_extra_global_param_from_device()

def pygenn.genn_model.GeNNModel.pull_extra_global_param_from_device	(		self,
			pop_name,
			egp_name,
			size = None
	)

Pull extra global parameter from the device for a given population.

pull_recording_buffers_from_device()

def pygenn.genn_model.GeNNModel.pull_recording_buffers_from_device

(

self

)

Pull recording buffers from device.

pull_spike_events_from_device()

def pygenn.genn_model.GeNNModel.pull_spike_events_from_device	(		self,
			pop_name
	)

Pull spike events from the device for a given population.

pull_spikes_from_device()

def pygenn.genn_model.GeNNModel.pull_spikes_from_device	(		self,
			pop_name
	)

Pull spikes from the device for a given population.

pull_state_from_device()

def pygenn.genn_model.GeNNModel.pull_state_from_device	(		self,
			pop_name
	)

Pull state from the device for a given population.

pull_var_from_device()

def pygenn.genn_model.GeNNModel.pull_var_from_device	(		self,
			pop_name,
			var_name
	)

Pull variable from the device for a given population.

push_connectivity_to_device()

def pygenn.genn_model.GeNNModel.push_connectivity_to_device	(		self,
			pop_name
	)

Push connectivity to the device for a given population.

push_current_spike_events_to_device()

def pygenn.genn_model.GeNNModel.push_current_spike_events_to_device	(		self,
			pop_name
	)

Push current spike events to the device for a given population.

push_current_spikes_to_device()

def pygenn.genn_model.GeNNModel.push_current_spikes_to_device	(		self,
			pop_name
	)

Push current spikes to the device for a given population.

push_extra_global_param_to_device()

def pygenn.genn_model.GeNNModel.push_extra_global_param_to_device	(		self,
			pop_name,
			egp_name,
			size = None
	)

Push extra global parameter to the device for a given population.

push_spike_events_to_device()

def pygenn.genn_model.GeNNModel.push_spike_events_to_device	(		self,
			pop_name
	)

Push spike events to the device for a given population.

push_spikes_to_device()

def pygenn.genn_model.GeNNModel.push_spikes_to_device	(		self,
			pop_name
	)

Push spikes to the device for a given population.

push_state_to_device()

def pygenn.genn_model.GeNNModel.push_state_to_device	(		self,
			pop_name
	)

Push state to the device for a given population.

push_var_to_device()

def pygenn.genn_model.GeNNModel.push_var_to_device	(		self,
			pop_name,
			var_name
	)

Push variable to the device for a given population.

reinitialise()

def pygenn.genn_model.GeNNModel.reinitialise

(

self

)

reinitialise model to its original state without re-loading

step_time()

def pygenn.genn_model.GeNNModel.step_time

(

self

)

Make one simulation step.

synapse_dynamics_time()

def pygenn.genn_model.GeNNModel.synapse_dynamics_time

(

self

)

t() [1/2]

def pygenn.genn_model.GeNNModel.t

(

self

)

Simulation time in ms.

t() [2/2]

def pygenn.genn_model.GeNNModel.t	(		self,
			t
	)

timestep() [1/2]

def pygenn.genn_model.GeNNModel.timestep

(

self

)

Simulation time step.

timestep() [2/2]

def pygenn.genn_model.GeNNModel.timestep	(		self,
			timestep
	)

timing_enabled() [1/2]

def pygenn.genn_model.GeNNModel.timing_enabled

(

self

)

timing_enabled() [2/2]

def pygenn.genn_model.GeNNModel.timing_enabled	(		self,
			timing
	)

unload()

def pygenn.genn_model.GeNNModel.unload

(

self

)

use_backend() [1/2]

def pygenn.genn_model.GeNNModel.use_backend

(

self

)

use_backend() [2/2]

def pygenn.genn_model.GeNNModel.use_backend	(		self,
			backend
	)

Member Data Documentation

backend_log_level

pygenn.genn_model.GeNNModel.backend_log_level

current_sources

pygenn.genn_model.GeNNModel.current_sources

custom_updates

pygenn.genn_model.GeNNModel.custom_updates

default_var_location

pygenn.genn_model.GeNNModel.default_var_location

dT

pygenn.genn_model.GeNNModel.dT

genn_types

pygenn.genn_model.GeNNModel.genn_types

model_name

pygenn.genn_model.GeNNModel.model_name

neuron_populations

pygenn.genn_model.GeNNModel.neuron_populations

synapse_populations

pygenn.genn_model.GeNNModel.synapse_populations

use_backend

pygenn.genn_model.GeNNModel.use_backend

---

The documentation for this class was generated from the following file:

• genn_model.py