| | |
- Boost.Python.instance(__builtin__.object)
-
- _nemo.Configuration
- _nemo.Network
- _nemo.Simulation
class Configuration(Boost.Python.instance) |
| |
Global configuration
|
| |
- Method resolution order:
- Configuration
- Boost.Python.instance
- __builtin__.object
Methods defined here:
- __init__(...)
- __init__( (object)arg1) -> None :
C++ signature :
void __init__(_object*)
- __reduce__ = (...)
- backend_description(...)
- backend_description( (Configuration)arg1) -> str :
Description of the currently selected simulation backend
Returns Textual description of the currently selected backend
The backend can be changed using setCudaBackend or setCpuBackend
C++ signature :
char const* backend_description(nemo::Configuration {lvalue})
- set_cpu_backend(...)
- set_cpu_backend( (Configuration)arg1) -> None :
specify that the CPU backend should be used
C++ signature :
void set_cpu_backend(nemo::Configuration {lvalue})
- set_cuda_backend(...)
- set_cuda_backend( (Configuration)arg1, (int)arg2) -> None :
specify that the CUDA backend should be used
Inputs:
deviceNumber
Specify that the CUDA backend should be used and optionally specify a
desired device. If the (default) device value of -1 is used the backend
will choose the best available device. If the cuda backend (and the chosen
device) cannot be used for whatever reason, an exception is raised. The
device numbering is the numbering used internally by nemo (see
cudaDeviceCount and cudaDeviceDescription). This device numbering may
differ from the one provided by the CUDA driver directly, since NeMo
ignores any devices it cannot use.
C++ signature :
void set_cuda_backend(nemo::Configuration {lvalue},int)
- set_stdp_function(...)
- set_stdp_function( (Configuration)arg1, (std_vector_float)arg2, (std_vector_float)arg3, (float)arg4, (float)arg5) -> None :
enable STDP and set the global STDP function
Inputs:
prefire -- STDP function values for spikes arrival times before the postsynaptic firing, starting closest to the postsynaptic firing
postfire -- STDP function values for spikes arrival times after the postsynaptic firing, starting closest to the postsynaptic firing
minWeight -- Lowest (negative) weight beyond which inhibitory synapses are not potentiated
maxWeight -- Highest (positive) weight beyond which excitatory synapses are not potentiated
The STDP function is specified by providing the values sampled at integer
cycles within the STDP window.
C++ signature :
void set_stdp_function(nemo::Configuration {lvalue},std::vector<float, std::allocator<float> >,std::vector<float, std::allocator<float> >,float,float)
Data and other attributes defined here:
- __instance_size__ = 24
Data descriptors inherited from Boost.Python.instance:
- __dict__
- __weakref__
Data and other attributes inherited from Boost.Python.instance:
- __new__ = <built-in method __new__ of Boost.Python.class object>
- T.__new__(S, ...) -> a new object with type S, a subtype of T
|
class Network(Boost.Python.instance) |
| |
A Network is constructed by adding individual neurons and synapses to the
network. Neurons are given indices (from 0) which should be unique for each
neuron. When adding synapses the source or target neurons need not
necessarily exist yet, but should be defined before the network is
finalised.
|
| |
- Method resolution order:
- Network
- Boost.Python.instance
- __builtin__.object
Methods defined here:
- __init__(...)
- __init__( (object)arg1) -> None :
C++ signature :
void __init__(_object*)
- __reduce__ = (...)
- add_neuron(...)
- object add_neuron(tuple args, dict kwds) :
add one or more neurons to the network
Inputs:
type -- Neuron type
idx -- Neuron index
parameters... -- all neuron parameters
state... -- all state variables
The meaning of the parameters and state variables varies depending on the
neuron typeThis function may be called either in a scalar or vector form.
In the scalar form all inputs are scalars. In the vector form, the neuron
index argument plus any number of the other arguments are lists of the same
length. In this second form scalar inputs are replicated the appropriate
number of times
C++ signature :
object add_neuron(tuple args, dict kwds)
- add_neuron_type(...)
- add_neuron_type( (Network)arg1, (str)arg2) -> int :
register a new neuron type with the network
Inputs:
name -- canonical name of the neuron type. The neuron type data is loaded from a plugin configuration file of the same name.
Returns index of the the neuron type, to be used when adding neurons
This function must be called before neurons of the specified type can be
added to the network.
C++ signature :
unsigned int add_neuron_type(nemo::Network {lvalue},std::string)
- add_synapse(...)
- add_synapse( (Network)arg1, (object)arg2, (object)arg3, (object)arg4, (object)arg5, (object)arg6) -> object :
add a single synapse to the network
Inputs:
source -- Index of source neuron
target -- Index of target neuron
delay -- Synapse conductance delay in milliseconds
weight -- Synapse weights
plastic -- Boolean specifying whether or not this synapse is plastic
Returns Unique synapse ID
The input arguments can be any combination of lists of equal length and
scalars. If the input arguments contain a mix of scalars and lists the
scalar arguments are replicated the required number of times.
C++ signature :
_object* add_synapse(nemo::Network {lvalue},_object*,_object*,_object*,_object*,_object*)
- get_neuron_parameter(...)
- get_neuron_parameter( (Network)arg1, (object)arg2, (int)arg3) -> object :
get neuron parameter
Inputs:
idx -- neuron index
varno -- variable index
Returns value of the neuron parameter
The neuron parameters do not change during simulation. For the Izhikevich
model: 0=a, 1=b, 2=c, 3=d. The neuron index may be either scalar or a list.
The output has the same length as the neuron input
C++ signature :
_object* get_neuron_parameter(nemo::Network {lvalue},_object*,unsigned int)
- get_neuron_state(...)
- get_neuron_state( (Network)arg1, (object)arg2, (int)arg3) -> object :
get neuron state variable
Inputs:
idx -- neuron index
varno -- variable index
Returns value of the relevant variable
For the Izhikevich model: 0=u, 1=v. The neuron index may be either scalar
or a list. The output has the same length as the neuron input
C++ signature :
_object* get_neuron_state(nemo::Network {lvalue},_object*,unsigned int)
- get_synapse_delay(...)
- get_synapse_delay( (Network)arg1, (object)arg2) -> object :
return the conduction delay for the specified synapse
Inputs:
synapse -- synapse id (as returned by addSynapse)
Returns conduction delay of the specified synapse
The input synapse indices may be either a scalar or a list. The return
value has the same form
C++ signature :
_object* get_synapse_delay(nemo::Network,_object*)
- get_synapse_plastic(...)
- get_synapse_plastic( (Network)arg1, (object)arg2) -> object :
return the boolean plasticity status for the specified synapse
Inputs:
synapse -- synapse id (as returned by addSynapse)
Returns plasticity status of the specified synapse
The input synapse indices may be either a scalar or a list. The return
value has the same form
C++ signature :
_object* get_synapse_plastic(nemo::Network,_object*)
- get_synapse_source(...)
- get_synapse_source( (Network)arg1, (int)arg2) -> int :
C++ signature :
unsigned int get_synapse_source(nemo::Network {lvalue},unsigned long)
get_synapse_source( (Network)arg1, (object)arg2) -> object :
return the source neuron of the specified synapse
Inputs:
synapse -- synapse id (as returned by addSynapse)
Returns source neuron index
The input synapse indices may be either a scalar or a list. The return
value has the same form
C++ signature :
_object* get_synapse_source(nemo::Network,_object*)
- get_synapse_target(...)
- get_synapse_target( (Network)arg1, (object)arg2) -> object :
return the target of the specified synapse
Inputs:
synapse -- synapse id (as returned by addSynapse)
Returns target neuron index
The input synapse indices may be either a scalar or a list. The return
value has the same form
C++ signature :
_object* get_synapse_target(nemo::Network,_object*)
- get_synapse_weight(...)
- get_synapse_weight( (Network)arg1, (object)arg2) -> object :
return the weight for the specified synapse
Inputs:
synapse -- synapse id (as returned by addSynapse)
Returns weight of the specified synapse
The input synapse indices may be either a scalar or a list. The return
value has the same form
C++ signature :
_object* get_synapse_weight(nemo::Network,_object*)
- get_synapses_from(...)
- get_synapses_from( (Network)arg1, (int)arg2) -> std_vector_uint64 :
return the synapse ids for all synapses with the given source neuron
Inputs:
source -- source neuron index
Returns synapse ids
The input synapse indices may be either a scalar or a list. The return
value has the same form
C++ signature :
std::vector<unsigned long, std::allocator<unsigned long> > get_synapses_from(nemo::Network {lvalue},unsigned int)
- neuron_count(...)
- neuron_count( (Network)arg1) -> int :
Returns number of neurons in the network
C++ signature :
unsigned int neuron_count(nemo::Network {lvalue})
- set_neuron(...)
- object set_neuron(tuple args, dict kwds) :
modify one or more existing neurons
Inputs:
idx -- Neuron index
parameters... -- all neuron parameters
state... -- all state variables
The meaning of the parameters and state variables varies depending on the
neuron type (specified when the neuron was created)This function may be
called either in a scalar or vector form. In the scalar form all inputs are
scalars. In the vector form, the neuron index argument plus any number of
the other arguments are lists of the same length. In this second form
scalar inputs are replicated the appropriate number of times
C++ signature :
object set_neuron(tuple args, dict kwds)
- set_neuron_parameter(...)
- set_neuron_parameter( (Network)arg1, (object)arg2, (int)arg3, (object)arg4) -> None :
set neuron parameter
Inputs:
idx -- neuron index
varno -- variable index
val -- value of the neuron parameter
The neuron parameters do not change during simulation. For the Izhikevich
model: 0=a, 1=b, 2=c, 3=d. The neuron and value parameters can be either
both scalar or both lists of the same length
C++ signature :
void set_neuron_parameter(nemo::Network {lvalue},_object*,unsigned int,_object*)
- set_neuron_state(...)
- set_neuron_state( (Network)arg1, (object)arg2, (int)arg3, (object)arg4) -> None :
set neuron state variable
Inputs:
idx -- neuron index
varno -- variable index
val -- value of the relevant variable
For the Izhikevich model: 0=u, 1=v. The neuron and value parameters can be
either both scalar or both lists of the same length
C++ signature :
void set_neuron_state(nemo::Network {lvalue},_object*,unsigned int,_object*)
Data and other attributes defined here:
- __instance_size__ = 40
Data descriptors inherited from Boost.Python.instance:
- __dict__
- __weakref__
Data and other attributes inherited from Boost.Python.instance:
- __new__ = <built-in method __new__ of Boost.Python.class object>
- T.__new__(S, ...) -> a new object with type S, a subtype of T
|
class Simulation(Boost.Python.instance) |
| |
A simulation is created from a network and a configuration object. The
simulation is run by stepping through it, providing stimulus as
appropriate. It is possible to read back synapse data at run time. The
simulation also maintains a timer for both simulated time and wallclock
time.
|
| |
- Method resolution order:
- Simulation
- Boost.Python.instance
- __builtin__.object
Methods defined here:
- __init__(...)
- __init__( (object)arg1, (Network)arg2, (Configuration)arg3) -> object :
C++ signature :
void* __init__(boost::python::api::object,nemo::Network,nemo::Configuration)
- __reduce__ = (...)
- apply_stdp(...)
- apply_stdp( (Simulation)arg1, (float)arg2) -> None :
update synapse weights using the accumulated STDP statistics
Inputs:
reward -- Multiplier for the accumulated weight change
C++ signature :
void apply_stdp(nemo::Simulation {lvalue},float)
- elapsed_simulation(...)
- elapsed_simulation( (Simulation)arg1) -> int :
Returns number of milliseconds of simulation time elapsed since first simulation step (or last timer reset)
C++ signature :
unsigned long elapsed_simulation(nemo::Simulation {lvalue})
- elapsed_wallclock(...)
- elapsed_wallclock( (Simulation)arg1) -> int :
Returns number of milliseconds of wall-clock time elapsed since first simulation step (or last timer reset)
C++ signature :
unsigned long elapsed_wallclock(nemo::Simulation {lvalue})
- get_delays(*args, **kwargs)
- get_membrane_potential(...)
- get_membrane_potential( (Simulation)arg1, (object)arg2) -> object :
get neuron membane potential
Inputs:
idx -- neuron index
Returns membrane potential
The neuron index may be either scalar or a list. The output has the same
length as the neuron input
C++ signature :
_object* get_membrane_potential(nemo::Simulation {lvalue},_object*)
- get_neuron_parameter(...)
- get_neuron_parameter( (Simulation)arg1, (object)arg2, (int)arg3) -> object :
get neuron parameter
Inputs:
idx -- neuron index
varno -- variable index
Returns value of the neuron parameter
The neuron parameters do not change during simulation. For the Izhikevich
model: 0=a, 1=b, 2=c, 3=d. The neuron index may be either scalar or a list.
The output has the same length as the neuron input
C++ signature :
_object* get_neuron_parameter(nemo::Simulation {lvalue},_object*,unsigned int)
- get_neuron_state(...)
- get_neuron_state( (Simulation)arg1, (object)arg2, (int)arg3) -> object :
get neuron state variable
Inputs:
idx -- neuron index
varno -- variable index
Returns value of the relevant variable
For the Izhikevich model: 0=u, 1=v. The neuron index may be either scalar
or a list. The output has the same length as the neuron input
C++ signature :
_object* get_neuron_state(nemo::Simulation {lvalue},_object*,unsigned int)
- get_plastic(*args, **kwargs)
- get_synapse_delay(...)
- get_synapse_delay( (Simulation)arg1, (object)arg2) -> object :
return the conduction delay for the specified synapse
Inputs:
synapse -- synapse id (as returned by addSynapse)
Returns conduction delay of the specified synapse
The input synapse indices may be either a scalar or a list. The return
value has the same form
C++ signature :
_object* get_synapse_delay(nemo::Simulation,_object*)
- get_synapse_plastic(...)
- get_synapse_plastic( (Simulation)arg1, (object)arg2) -> object :
return the boolean plasticity status for the specified synapse
Inputs:
synapse -- synapse id (as returned by addSynapse)
Returns plasticity status of the specified synapse
The input synapse indices may be either a scalar or a list. The return
value has the same form
C++ signature :
_object* get_synapse_plastic(nemo::Simulation,_object*)
- get_synapse_source(...)
- get_synapse_source( (Simulation)arg1, (object)arg2) -> object :
return the source neuron of the specified synapse
Inputs:
synapse -- synapse id (as returned by addSynapse)
Returns source neuron index
The input synapse indices may be either a scalar or a list. The return
value has the same form
C++ signature :
_object* get_synapse_source(nemo::Simulation,_object*)
- get_synapse_target(...)
- get_synapse_target( (Simulation)arg1, (object)arg2) -> object :
return the target of the specified synapse
Inputs:
synapse -- synapse id (as returned by addSynapse)
Returns target neuron index
The input synapse indices may be either a scalar or a list. The return
value has the same form
C++ signature :
_object* get_synapse_target(nemo::Simulation,_object*)
- get_synapse_weight(...)
- get_synapse_weight( (Simulation)arg1, (object)arg2) -> object :
return the weight for the specified synapse
Inputs:
synapse -- synapse id (as returned by addSynapse)
Returns weight of the specified synapse
The input synapse indices may be either a scalar or a list. The return
value has the same form
C++ signature :
_object* get_synapse_weight(nemo::Simulation,_object*)
- get_synapses_from(...)
- get_synapses_from( (Simulation)arg1, (int)arg2) -> std_vector_uint64 :
return the synapse ids for all synapses with the given source neuron
Inputs:
source -- source neuron index
Returns synapse ids
The input synapse indices may be either a scalar or a list. The return
value has the same form
C++ signature :
std::vector<unsigned long, std::allocator<unsigned long> > get_synapses_from(nemo::Simulation {lvalue},unsigned int)
- get_targets(*args, **kwargs)
- get_weights(*args, **kwargs)
- reset_timer(...)
- reset_timer( (Simulation)arg1) -> None :
reset both wall-clock and simulation timer
C++ signature :
void reset_timer(nemo::Simulation {lvalue})
- set_neuron(...)
- object set_neuron(tuple args, dict kwds) :
modify one or more existing neurons
Inputs:
idx -- Neuron index
parameters... -- all neuron parameters
state... -- all state variables
The meaning of the parameters and state variables varies depending on the
neuron type (specified when the neuron was created)This function may be
called either in a scalar or vector form. In the scalar form all inputs are
scalars. In the vector form, the neuron index argument plus any number of
the other arguments are lists of the same length. In this second form
scalar inputs are replicated the appropriate number of times
C++ signature :
object set_neuron(tuple args, dict kwds)
- set_neuron_parameter(...)
- set_neuron_parameter( (Simulation)arg1, (object)arg2, (int)arg3, (object)arg4) -> None :
set neuron parameter
Inputs:
idx -- neuron index
varno -- variable index
val -- value of the neuron parameter
The neuron parameters do not change during simulation. For the Izhikevich
model: 0=a, 1=b, 2=c, 3=d. The neuron and value parameters can be either
both scalar or both lists of the same length
C++ signature :
void set_neuron_parameter(nemo::Simulation {lvalue},_object*,unsigned int,_object*)
- set_neuron_state(...)
- set_neuron_state( (Simulation)arg1, (object)arg2, (int)arg3, (object)arg4) -> None :
set neuron state variable
Inputs:
idx -- neuron index
varno -- variable index
val -- value of the relevant variable
For the Izhikevich model: 0=u, 1=v. The neuron and value parameters can be
either both scalar or both lists of the same length
C++ signature :
void set_neuron_state(nemo::Simulation {lvalue},_object*,unsigned int,_object*)
- step(self, fstim=None, istim=None)
- run simulation for a single cycle (1ms)
Inputs:
fstim -- An optional list of neurons which will be forced to fire this cycle
istim -- An optional list of neuron index/current pairs for external stimulus of the network
- step_f(...)
- step_f( (Simulation)arg1, (std_vector_unsigned)arg2) -> std_vector_unsigned :
C++ signature :
std::vector<unsigned int, std::allocator<unsigned int> > step_f(nemo::Simulation {lvalue},std::vector<unsigned int, std::allocator<unsigned int> >)
- step_fi(...)
- step_fi( (Simulation)arg1, (std_vector_unsigned)arg2, (object)arg3) -> std_vector_unsigned :
C++ signature :
std::vector<unsigned int, std::allocator<unsigned int> > step_fi(nemo::Simulation {lvalue},std::vector<unsigned int, std::allocator<unsigned int> >,std::vector<std::pair<unsigned int, float>, std::allocator<std::pair<unsigned int, float> > >)
- step_i(...)
- step_i( (Simulation)arg1, (object)arg2) -> std_vector_unsigned :
C++ signature :
std::vector<unsigned int, std::allocator<unsigned int> > step_i(nemo::Simulation {lvalue},std::vector<std::pair<unsigned int, float>, std::allocator<std::pair<unsigned int, float> > >)
- step_noinput(...)
- step_noinput( (Simulation)arg1) -> std_vector_unsigned :
C++ signature :
std::vector<unsigned int, std::allocator<unsigned int> > step_noinput(nemo::Simulation {lvalue})
Data descriptors inherited from Boost.Python.instance:
- __dict__
- __weakref__
Data and other attributes inherited from Boost.Python.instance:
- __new__ = <built-in method __new__ of Boost.Python.class object>
- T.__new__(S, ...) -> a new object with type S, a subtype of T
| |