Cypress
1.0
C++ Spiking Neural Network Simulation Framework
|
#include <delta_sigma.hpp>
Classes | |
class | DiscreteWindow |
struct | ExponentialWindow |
struct | GaussWindow |
Static Public Member Functions | |
static std::vector< Real > | encode (const std::vector< Real > &values, const DiscreteWindow &window, Real t0, Real min_val=-1.0, Real max_val=1.0, Real min_spike_interval=DEFAULT_MIN_SPIKE_INTERVAL) |
template<typename Fun > | |
static std::vector< Real > | encode (const Fun &f, const DiscreteWindow &window, Real t0, Real t1, Real min_val=-1.0, Real max_val=1.0, Real min_spike_interval=DEFAULT_MIN_SPIKE_INTERVAL) |
static std::vector< Real > | decode (const std::vector< Real > &spikes, const DiscreteWindow &window, Real t0, Real t1, Real min_val=-1.0, Real max_val=1.0) |
Static Public Attributes | |
static constexpr Real | DEFAULT_RESPONSE_TIME = 50e-3 |
static constexpr Real | DEFAULT_STEP = 1e-4 |
static constexpr Real | DEFAULT_EPS = 1e-6 |
static constexpr Real | DEFAULT_MIN_SPIKE_INTERVAL = 1e-3 |
Structure providing static methods for encoding continuous-valued functions as a series of spikes and vice versa.
|
static |
Converts the given input spike train into a continuous-valued function by convoluting the spikes with the given window function. Note that the result sampling interval is equivalent to that specified in the window function.
spikes | is an array containing the input spike times. |
window | is the discretised window function that should be used. |
t0 | is the timestamp the first element in the result vector corresponds to. |
t1 | is the time the last timestamp in the result vector should correspond to. |
|
static |
Encodes a time-series of continuous-valued list of values to a time-series of spikes, which can be decoded by convoluting the spike train with the given window function. Implemented as a delta-sigma encoder.
values | is a list containing the function values. The sampling interval is taken from the given discretised window function. |
t0 | is the time offset that should be added to the resulting spikes. |
window | is the discretised window function that should be used for encoding. |
min_spike_interval | is the minimum delay between two spikes should be greater or equal to the window sampling interval. |
min_val | is the minimum value occuring in the list of values. |
max_val | is the maximum value occuring in the list of values. |
|
inlinestatic |
Encodes a time-series of continuous-valued function as a time-series of spikes, which can be decoded by convoluting the spike train with the given window function. Implemented as a delta-sigma encoder.
values | is a list containing the function values. The sampling interval is taken from the given discretised window function. |
window | is the discretised window function that should be used for encoding. |
min_spike_interval | is the minimum delay between two spikes should be greater or equal to the window sampling interval. |
t0 | is the time offset that should be added to the resulting spikes. |
min_val | is the minimum value occuring in the list of values. |
max_val | is the maximum value occuring in the list of values. |
|
static |
|
static |
|
static |
|
static |