cypress Namespace Reference

Namespaces
	filesystem
	internal
	nef
	sli
	spikes
	transformations

Classes
struct	Accessor
class	AllToAllConnector
class	Backend
class	BinnfDecodeException
class	BS_Lib
	Singleton class for runtime loading of the BrainScaleS Backend, which wraps all BrainScaleS libraries and will be provided pre-compiled. More...
struct	Comperator
class	ConnectableMixin
class	ConnectionDescriptor
class	Connector
class	ConstantSizeNeuronParametersBase
class	CypressException
class	DataMixin
class	EifCondExpIsfaIsta
class	EifCondExpIsfaIstaParameters
class	EifCondExpIsfaIstaSignals
class	energenie
class	ExecutionError
class	FixedFanConnectorBase
class	FixedFanInConnector
class	FixedFanOutConnector
class	FixedProbabilityConnector
class	FixedProbabilityConnectorBase
class	FromListConnector
class	FunctorConnector
class	FunctorConnectorBase
class	GeNN
class	GENN_Lib
	Singleton class for runtime loading of the GeNN Backend, which wraps all GeNN/GPU libraries. More...
class	HomogeneousPopulationRequiredException
class	IfCondExp
class	IfCondExpParameters
class	IfCondExpSignals
class	IfCurrExp
class	IfCurrExpParameters
class	IfCurrExpSignals
class	IfFacetsHardware1
class	IfFacetsHardware1Parameters
class	IfFacetsHardware1Signals
class	InvalidConnectionException
class	InvalidParameter
class	InvalidParameterArraySize
class	InvalidSignal
class	IterableMixin
struct	IterableRange
struct	LocalConnection
class	LogBackend
class	LogFileBackend
class	Logger
class	LogStreamBackend
class	Matrix
class	MissingColumnException
class	NEST
class	NetIO4
class	Network
class	NetworkBase
struct	NetworkRuntime
class	Neuron
class	NeuronBase
class	NeuronMixin
class	NeuronParameter
class	NeuronParameters
class	NeuronParametersBase
class	NeuronSignals
class	NeuronSignalsBase
class	NeuronType
class	NeuronTypeBase
class	NeuronTypeTransformation
class	NMPI
class	NoSuchPopulationException
class	NotSupportedException
class	NullNeuron
class	NullNeuronParameters
class	NullNeuronSignals
struct	NumericPolicy
class	OneToOneConnector
class	Optional
class	Population
class	PopulationBase
class	PopulationData
class	PopulationDataView
class	PopulationMixin
class	PopulationView
class	PopulationViewBase
class	PowerDevice
class	PowerManagementBackend
class	Process
class	PyNN
class	PythonInstance
class	RandomConnector
struct	Range
class	Resource
struct	Resources
class	RNG
	Singleton class for. More...
class	SignalNotRecordedException
class	Slurm
class	SpikePairRuleAdditive
class	SpikePairRuleMultiplicative
class	SpikeSourceArray
class	SpikeSourceArrayParameters
class	SpikeSourceArraySignals
class	SpikeSourceConstFreq
class	SpikeSourceConstFreqParameters
class	SpikeSourceConstFreqSignals
class	SpikeSourceConstInterval
class	SpikeSourceConstIntervalParameters
class	SpikeSourceConstIntervalSignals
class	SpikeSourcePoisson
class	SpikeSourcePoissonParameters
class	SpikeSourcePoissonSignals
class	SpikingUtils
class	StaticSynapse
struct	Synapse
class	SynapseBase
class	Terminal
class	ToJson
class	Transformation
struct	TransformationAuxData
class	TransformationException
struct	TransformationProperties
class	Transformations
class	TsodyksMarkramMechanism
class	UniformConnector
class	UniformFunctorConnector
class	UniformFunctorConnectorBase
class	Vector
class	ViewableMixin

Typedefs
using	LIF = IfCondExp
using	LIFParameters = IfCondExpParameters
using	LIFSignals = IfCondExpSignals
using	AdEx = EifCondExpIsfaIsta
using	AdExParameters = EifCondExpIsfaIstaParameters
using	AdExSignals = EifCondExpIsfaIstaSignals
using	TransformationCtor = std::function< std::unique_ptr< Transformation >()>
using	TransformationTest = std::function< bool(const Backend &, const NetworkBase &)>
using	RegisteredTransformation = size_t
using	sx_double_t = long double
using	sx_int_t = unsigned long long int
using	Real = double
using	NeuronIndex = int32_t
using	PopulationIndex = int32_t
using	Json = nlohmann::json

Enumerations
enum	LogSeverity : int32_t {   DEBUG = 10, INFO = 20, WARNING = 30, ERROR = 40,   FATAL_ERROR = 50 }
enum	MatrixFlags { MatrixFlags::NONE = 0, MatrixFlags::ZEROS = 1 }

Functions
void	to_json (Json &j, const Network &netw)
	Automatic conversion of a Network object to JSON. Now you can uses Json(network) as you would do with STL containers. More...
void	from_json (const Json &j, Network &netw)
	With the method you can just use Network netw = json.get<Network>() as you can do with STL containers. More...
void	from_json (const Json &json, NetworkRuntime &runtime)
	Read in NetworkRuntime from json: json["sim"] == runtime.sim;. More...
void	to_json (Json &json, const NetworkRuntime &runtime)
	NetworkRuntime to json conversion: json["sim"] == runtime.sim;. More...
template<typename T >
void	to_json (Json &j, const Matrix< T > &mat)
	Automatic conversion of the internal matrix type to json. Use Json(mat). More...
template<typename T >
void	from_json (const Json &j, Matrix< T > &mat)
	Conversion from json to internal matrix type. Use json.get<Matrix<T>> as you would do with a STL container. More...
std::vector< std::vector< LocalConnection > >	instantiate_connections (const std::vector< ConnectionDescriptor > &descrs)
bool	instantiate_connections_to_file (std::string filename, const std::vector< ConnectionDescriptor > &descrs)
template<typename Iterator >
IterableRange< Iterator >	make_iterable_range (Iterator begin, Iterator end)
Json &	join (Json &tar, const Json &src)
Logger &	global_logger ()
template<typename T , size_t Rows, size_t Cols>
Matrix< T >	make_matrix (const std::array< std::array< T, Cols >, Rows > &init)
template<typename T >
Matrix< T >	make_matrix (std::initializer_list< T > list)
template<typename T >
std::map< std::string, T >	json_to_map (const cypress::Json &obj)
template<typename T >
std::vector< T >	read_check (std::map< std::string, T > &input, const std::vector< std::string > &names, const std::vector< T > &defaults)
constexpr Range< float >	linspace (float x0, float x1, ptrdiff_t num)
constexpr Range< double >	linspace (double x0, double x1, ptrdiff_t num)
constexpr Range< size_t >	range (ptrdiff_t i)
constexpr Range< ptrdiff_t >	range (ptrdiff_t i0, ptrdiff_t i1)
constexpr Range< float >	range (float x0, float x1, float step)
constexpr Range< double >	range (double x0, double x1, double step)
void	create_dot (const NetworkBase &netw, const std::string graph_label="Network Architecture", const std::string filename="graph.dot", const bool call_dot=true)
	Generates a dot file for the given network. Calls "dot" to generate a pdf with a visualization of the network and list connectors. More...

Detailed Description

íle brainscales_wrap.hpp

Defines a Singleton for dynamically loading the brainscales backend

Author

Christoph Ostrau

íle genn.hpp

This is the backend implementation for GeNN, allowing simulations on GPUs.

Author

Christoph Ostrau

íle genn_lib.hpp

Defines a Singleton for dynamically loading the genn backend

Author

Christoph Ostrau

íle rng.hpp

Defines a Singleton for the random number generator

Author

Christoph Ostrau

---

Class Documentation

struct cypress::NetworkRuntime

Structure storing information about the network runtime.

Class Members
Real	duration	biological runtime in ms
Real	finalize	Finalization time. Time required to retrieve the results from the driver.
Real	initialize	Initialization time. Time required to construct the network.
Real	sim	Raw simulation time. This may include setup and finalization time required by the hardware timer.
Real	sim_pure	Theoretical runtime, where direct measurement is not possible. E.g.: simulation time * time_scale_factor
Real	total	Total runtime – sum of the three other times.

struct cypress::TransformationAuxData

Auxiliary data which may be affected by the transformation.

Class Members
Real	duration	Contains the network simulation duration. Certain transformations may choose to extend the actual simulation duration.

struct cypress::TransformationProperties

Structure describing the

Class Members
size_t	cost	A non-zero cost value which can be used to indicate the cost associated with executing this transformation.
bool	lossy	If true indicates that the transformation cannot be transformed without loosing informatoion. Usage of these transformations can be disabled globally.

Typedef Documentation

using cypress::AdEx = typedef EifCondExpIsfaIsta

using cypress::AdExParameters = typedef EifCondExpIsfaIstaParameters

using cypress::AdExSignals = typedef EifCondExpIsfaIstaSignals

using cypress::Json = typedef nlohmann::json

using cypress::LIF = typedef IfCondExp

using cypress::LIFParameters = typedef IfCondExpParameters

using cypress::LIFSignals = typedef IfCondExpSignals

using cypress::NeuronIndex = typedef int32_t

using cypress::PopulationIndex = typedef int32_t

using cypress::Real = typedef double

The Real type is a floating point type. Its width can be chosen by the user by setting the CYPRESS_REAL_WIDTH macro.

using cypress::RegisteredTransformation = typedef size_t

Stub type which is used to allow the registration of new transformations as part of a static value initialization.

using cypress::sx_double_t = typedef long double

Double type to be used in suffix declarations.

using cypress::sx_int_t = typedef unsigned long long int

Integer type to be used in suffix declarations.

using cypress::TransformationCtor = typedef std::function<std::unique_ptr<Transformation>()>

Constructor function type which constructs a new Transformation instance.

using cypress::TransformationTest = typedef std::function<bool(const Backend &, const NetworkBase &)>

Test function type which is used to test whether a certain transformation should be applied to a Network/Backend combination.

Enumeration Type Documentation

enum cypress::LogSeverity : int32_t

The LogSeverity enum holds the LogSeverity of a log message. Higher severities are associated with higher values.

Enumerator
DEBUG
INFO
WARNING
ERROR
FATAL_ERROR

enum cypress::MatrixFlags

strong

Enumerator
NONE
ZEROS

Function Documentation

void cypress::create_dot	(	const NetworkBase &	netw,
		const std::string	graph_label = "Network Architecture",
		const std::string	filename = "graph.dot",
		const bool	call_dot = true
	)

Generates a dot file for the given network. Calls "dot" to generate a pdf with a visualization of the network and list connectors.

Parameters

netw	network instance
graph_label	label of the graph, Defaults to "Network Architecture".
filename	filename to store dot file Defaults to "graph.dot".
call_dot	call dot? output will be filename".pdf" Defaults to true.

void cypress::from_json	(	const Json &	j,
		Network &	netw
	)

With the method you can just use Network netw = json.get<Network>() as you can do with STL containers.

Parameters

j	source json object
netw	target network, in which the network will be constructed. NOTE: network will not be deleted!

void cypress::from_json	(	const Json &	json,
		NetworkRuntime &	runtime
	)

Read in NetworkRuntime from json: json["sim"] == runtime.sim;.

Parameters

json	source json
runtime	target runtime.

template<typename T >

void cypress::from_json	(	const Json &	j,
		Matrix< T > &	mat
	)

Conversion from json to internal matrix type. Use json.get<Matrix<T>> as you would do with a STL container.

Parameters

T	matrix type
j	source json
mat	target matrix

Logger& cypress::global_logger

(

)

std::vector<std::vector<LocalConnection> > cypress::instantiate_connections

(

const std::vector< ConnectionDescriptor > &

descrs

)

Instantiates the connection descriptors into a flat list of actual connections.

Parameters

descrs

is a list of connection descriptors which should be turned into a list of connections.

Returns

a vector of vectors containing the instantiated connections for each Descriptor

bool cypress::instantiate_connections_to_file	(	std::string	filename,
		const std::vector< ConnectionDescriptor > &	descrs
	)

Json& cypress::join	(	Json &	tar,
		const Json &	src
	)

Recursively joins one JSON object into another, overriding already existing keys in the target.

Parameters

tar	is the target JSON object. Keys in this object may be overriden.
src	is the source JSON object. All keys from this object will be available in tar.

Returns

a reference at tar.

template<typename T >

std::map<std::string, T> cypress::json_to_map

(

const cypress::Json &

obj

)

Store input form json in a map to check everything!

constexpr Range<float> cypress::linspace	(	float	x0,
		float	x1,
		ptrdiff_t	num
	)

constexpr Range<double> cypress::linspace	(	double	x0,
		double	x1,
		ptrdiff_t	num
	)

template<typename Iterator >

IterableRange<Iterator> cypress::make_iterable_range	(	Iterator	begin,
		Iterator	end
	)

Allows to create an instance of IterableRange while automatically deducing the iterator type.

Parameters

begin	is an iterator pointing at the first element.
end	is an iterator pointing at the last-plus-one element.

template<typename T , size_t Rows, size_t Cols>

Matrix<T> cypress::make_matrix

(

const std::array< std::array< T, Cols >, Rows > &

init

)

template<typename T >

Matrix<T> cypress::make_matrix

(

std::initializer_list< T >

list

)

constexpr Range<size_t> cypress::range

(

ptrdiff_t

i

)

constexpr Range<ptrdiff_t> cypress::range	(	ptrdiff_t	i0,
		ptrdiff_t	i1
	)

constexpr Range<float> cypress::range	(	float	x0,
		float	x1,
		float	step
	)

constexpr Range<double> cypress::range	(	double	x0,
		double	x1,
		double	step
	)

template<typename T >

std::vector<T> cypress::read_check	(	std::map< std::string, T > &	input,
		const std::vector< std::string > &	names,
		const std::vector< T > &	defaults
	)

void cypress::to_json	(	Json &	j,
		const Network &	netw
	)

Automatic conversion of a Network object to JSON. Now you can uses Json(network) as you would do with STL containers.

Parameters

j	target JSon object
netw	source network object

void cypress::to_json	(	Json &	json,
		const NetworkRuntime &	runtime
	)

NetworkRuntime to json conversion: json["sim"] == runtime.sim;.

Parameters

json	target json object
runtime	source NetworkRuntime object

template<typename T >

void cypress::to_json	(	Json &	j,
		const Matrix< T > &	mat
	)

Automatic conversion of the internal matrix type to json. Use Json(mat).

Parameters

T	Template parameter of the matrix
j	target json
mat	source matrix