def __init__(self, data={}, burst_size=1):
"""
Constructor of the OutputParameters class. Either copies the parameters
from the given "data" dictionary or uses the given arguments.
:param data: dictionary the arguments are preferably copied from.
:param burst_size: number of expected output spikes in a neuron.
"""
utils.init_key(self, data, "burst_size", burst_size, _type=int)
def __init__(self, data={}, burst_size=1):
"""
Constructor of the OutputParameters class. Either copies the parameters
from the given "data" dictionary or uses the given arguments.
:param data: dictionary the arguments are preferably copied from.
:param burst_size: number of expected output spikes in a neuron.
"""
utils.init_key(self, data, "burst_size", burst_size, _type=int)
def __init__(self, data={}, data_params={}, topology_params={},
input_params={}, output_params={}, experiments=[]):
utils.init_key(self, data, "data", data_params)
utils.init_key(self, data, "topology", topology_params)
utils.init_key(self, data, "input", input_params)
utils.init_key(self, data, "output", output_params)
utils.init_key(self, data, "experiments", experiments)
self["data"] = DataParameters(self["data"])
self["topology"] = TopologyParameters(self["topology"])
self["input"] = InputParameters(self["input"])
self["output"] = OutputParameters(self["output"])
if not isinstance(self["experiments"], list):
self["experiments"] = [self["experiments"]]
for i in xrange(len(self["experiments"])):
self["experiments"][i] = ExperimentDescriptor(
self["experiments"][i])
def __init__(self, data={}, name="", repeat=1, sweeps={}):
"""
Creates an experiment descriptor, which consists of a repeat count and
a number of sweep descriptors for the dimensions that should be varied.
"""
utils.init_key(self, data, "name", name)
utils.init_key(self, data, "repeat", repeat)
utils.init_key(self, data, "sweeps", sweeps)
def __init__(self, data={}, burst_size=1, time_window=100.0, isi=1.0,
sigma_t=0.0, sigma_t_offs=0.0, p0=0.0, p1=0.0):
"""
Constructor of the InputParameters class. If a dictionary or another
InputParameters class is passed as first parameter, elements from this
instance are copied with a precedence over the specified parameters.
:param data: dictionary from which parameters are copied preferably.
:param burst_size: number of spikes representing a "one".
:param time_window: time between the presentation of two input samples.
:param isi: inter spike interval between spike of a burst.
:param sigma_t: spike jitter
:param sigma_t_offs: spike offset jitter
:param p0: probability with which individual spikes are omitted.
:param p1: probability with which spikes from a "one" are randomly
introduced for a sample containing zeros.
"""
utils.init_key(self, data, "burst_size", burst_size, _type=int)
utils.init_key(self, data, "time_window", time_window)
utils.init_key(self, data, "isi", isi)
utils.init_key(self, data, "sigma_t", sigma_t)
utils.init_key(self, data, "sigma_t_offs", sigma_t_offs)
utils.init_key(self, data, "p0", p0)
utils.init_key(self, data, "p1", p1)
def __init__(self, data={}, input_times=[], input_indices=[],
output_times=[], output_indices=[], input_params={},
data_params={},
topology_params={}, meta_data={}, mat_in=[], mat_out=[]):
utils.init_key(self, data, "input_times", input_times)
utils.init_key(self, data, "input_indices", input_indices)
utils.init_key(self, data, "output_times", output_times)
utils.init_key(self, data, "output_indices", output_indices)
utils.init_key(self, data, "input_params", input_params)
utils.init_key(self, data, "data_params", data_params)
utils.init_key(self, data, "topology_params", topology_params)
utils.init_key(self, data, "meta_data", meta_data)
utils.init_key(self, data, "mat_in", mat_in)
utils.init_key(self, data, "mat_out", mat_out)
self["input_params"] = InputParameters(self["input_params"])
self["data_params"] = DataParameters(self["data_params"])
self["topology_params"] = TopologyParameters(self["topology_params"])
def __init__(self, data={}, name="", populations=[], connections=[],
input_times=[], input_indices=[], input_split=[],
spatial_split=[],
input_params=[], data_params=[], topology_params=[],
meta_data=[],
mat_in=[], mat_out=[]):
utils.init_key(self, data, "name", name)
utils.init_key(self, data, "populations", populations)
utils.init_key(self, data, "connections", connections)
utils.init_key(self, data, "input_times", input_times)
utils.init_key(self, data, "input_indices", input_indices)
utils.init_key(self, data, "input_split", input_split)
utils.init_key(self, data, "spatial_split", spatial_split)
utils.init_key(self, data, "input_params", input_params)
utils.init_key(self, data, "meta_data", meta_data)
utils.init_key(self, data, "data_params", data_params)
utils.init_key(self, data, "topology_params", topology_params)
utils.init_key(self, data, "mat_in", mat_in)
utils.init_key(self, data, "mat_out", mat_out)
# Fix things up in case a NetworkInstance was passed to the constructor
if (len(self["spatial_split"]) == 0 and len(self["populations"]) > 0):
self["input_split"] = [self["input_split"]]
self["input_params"] = [self["input_params"]]
self["data_params"] = [self["data_params"]]
self["topology_params"] = [self["topology_params"]]
self["meta_data"] = [self["meta_data"]]
self["mat_in"] = [self["mat_in"]]
self["mat_out"] = [self["mat_out"]]
self["spatial_split"].append({
"population": len(self["populations"]),
"input": len(self["input_times"])
});
def __init__(self, data={}, populations=[], connections=[],
input_times=[], input_indices=[], input_split=[],
input_params={},
data_params={}, topology_params={}, meta_data={}, mat_in=[],
mat_out=[]):
utils.init_key(self, data, "populations", populations)
utils.init_key(self, data, "connections", connections)
utils.init_key(self, data, "input_times", input_times)
utils.init_key(self, data, "input_indices", input_indices)
utils.init_key(self, data, "input_split", input_split)
utils.init_key(self, data, "input_params", input_params)
utils.init_key(self, data, "data_params", data_params)
utils.init_key(self, data, "topology_params", topology_params)
utils.init_key(self, data, "meta_data", meta_data)
utils.init_key(self, data, "mat_in", mat_in)
utils.init_key(self, data, "mat_out", mat_out)
if not isinstance(self["input_params"], list):
self["input_params"] = [self["input_params"]]
for i in xrange(len(self["input_params"])):
self["input_params"][i] = InputParameters(self["input_params"][i])
self["data_params"] = DataParameters(self["data_params"])
self["topology_params"] = TopologyParameters(self["topology_params"])
def __init__(self, data={}, n_bits_in=16, n_bits_out=16, n_ones_in=3,
n_ones_out=3, n_samples=-1, algorithm="balanced"):
"""
Fills the network structure with the given parameters.
:param n_bits_in: number of input bits.
:param n_bits_out: number of output bits.
:param n_ones_in: number of ones in the input per sample.
:param n_ones_out: number of ones in the output per sample.
:param n_samples: number of samples. If negative, the number of samples
with the maximum information are chosen.
"""
# If "n_bits" is specified, set both n_bits_in and n_bits_out
if ("n_bits" in data) and (data["n_bits"] > 0):
self["n_bits_in"] = int(data["n_bits"])
self["n_bits_out"] = int(data["n_bits"])
self["n_bits"] = int(data["n_bits"])
else:
utils.init_key(self, data, "n_bits_in", n_bits_in, _type=int)
utils.init_key(self, data, "n_bits_out", n_bits_out, _type=int)
utils.init_key(self, data, "n_bits",
self["n_bits_in"] if self["n_bits_in"] == self[
"n_bits_out"]
else -1, _type=int)
utils.init_key(self, data, "n_ones_in", n_ones_in, _type=int)
utils.init_key(self, data, "n_ones_out", n_ones_out, _type=int)
utils.init_key(self, data, "n_samples", n_samples, _type=int)
utils.init_key(self, data, "algorithm", algorithm, _type=str)
if not self["algorithm"] in ["random", "balanced", "unique"]:
raise Exception("Invalid data generation algorithm \""
+ self["algorithm"]
+ "\", must be one of {\"random\", \"balanced\", \"unique\"}!")
# If n_ones_in and n_ones_out is not given, automatically calculate
# n_ones_in, n_ones_out and n_samples
if (self["n_ones_in"] <= 0) and (self["n_ones_out"] <= 0):
params = entropy.optimal_parameters(n_samples=self["n_samples"],
n_bits_in=self["n_bits_in"],
n_bits_out=self["n_bits_out"])
self["n_samples"] = params["n_samples"]
self["n_ones_in"] = params["n_ones_in"]
self["n_ones_out"] = params["n_ones_out"]
# Automatically choose the optimal number of samples
if self["n_samples"] <= 0:
self["n_samples"] = entropy.optimal_sample_count(
n_bits_out=self["n_bits_out"],
n_bits_in=self["n_bits_in"],
n_ones_out=self["n_ones_out"],
n_ones_in=self["n_ones_in"])
def __init__(
self,
data={},
input_times=[],
input_indices=[],
output_times=[],
output_indices=[],
input_params={},
data_params={},
topology_params={},
meta_data={},
mat_in=[],
mat_out=[],
):
utils.init_key(self, data, "input_times", input_times)
utils.init_key(self, data, "input_indices", input_indices)
utils.init_key(self, data, "output_times", output_times)
utils.init_key(self, data, "output_indices", output_indices)
utils.init_key(self, data, "input_params", input_params)
utils.init_key(self, data, "data_params", data_params)
utils.init_key(self, data, "topology_params", topology_params)
utils.init_key(self, data, "meta_data", meta_data)
utils.init_key(self, data, "mat_in", mat_in)
utils.init_key(self, data, "mat_out", mat_out)
self["input_params"] = InputParameters(self["input_params"])
self["data_params"] = DataParameters(self["data_params"])
self["topology_params"] = TopologyParameters(self["topology_params"])
def __init__(
self,
data={},
name="",
populations=[],
connections=[],
input_times=[],
input_indices=[],
input_split=[],
spatial_split=[],
input_params=[],
data_params=[],
topology_params=[],
meta_data=[],
mat_in=[],
mat_out=[],
):
utils.init_key(self, data, "name", name)
utils.init_key(self, data, "populations", populations)
utils.init_key(self, data, "connections", connections)
utils.init_key(self, data, "input_times", input_times)
utils.init_key(self, data, "input_indices", input_indices)
utils.init_key(self, data, "input_split", input_split)
utils.init_key(self, data, "spatial_split", spatial_split)
utils.init_key(self, data, "input_params", input_params)
utils.init_key(self, data, "meta_data", meta_data)
utils.init_key(self, data, "data_params", data_params)
utils.init_key(self, data, "topology_params", topology_params)
utils.init_key(self, data, "mat_in", mat_in)
utils.init_key(self, data, "mat_out", mat_out)
# Fix things up in case a NetworkInstance was passed to the constructor
if len(self["spatial_split"]) == 0 and len(self["populations"]) > 0:
self["input_split"] = [self["input_split"]]
self["input_params"] = [self["input_params"]]
self["data_params"] = [self["data_params"]]
self["topology_params"] = [self["topology_params"]]
self["meta_data"] = [self["meta_data"]]
self["mat_in"] = [self["mat_in"]]
self["mat_out"] = [self["mat_out"]]
self["spatial_split"].append({"population": len(self["populations"]), "input": len(self["input_times"])})
def __init__(
self,
data={},
populations=[],
connections=[],
input_times=[],
input_indices=[],
input_split=[],
input_params={},
data_params={},
topology_params={},
meta_data={},
mat_in=[],
mat_out=[],
):
utils.init_key(self, data, "populations", populations)
utils.init_key(self, data, "connections", connections)
utils.init_key(self, data, "input_times", input_times)
utils.init_key(self, data, "input_indices", input_indices)
utils.init_key(self, data, "input_split", input_split)
utils.init_key(self, data, "input_params", input_params)
utils.init_key(self, data, "data_params", data_params)
utils.init_key(self, data, "topology_params", topology_params)
utils.init_key(self, data, "meta_data", meta_data)
utils.init_key(self, data, "mat_in", mat_in)
utils.init_key(self, data, "mat_out", mat_out)
if not isinstance(self["input_params"], list):
self["input_params"] = [self["input_params"]]
for i in xrange(len(self["input_params"])):
self["input_params"][i] = InputParameters(self["input_params"][i])
self["data_params"] = DataParameters(self["data_params"])
self["topology_params"] = TopologyParameters(self["topology_params"])
def __init__(
self, data={}, n_bits_in=16, n_bits_out=16, n_ones_in=3, n_ones_out=3, n_samples=-1, algorithm="balanced"
):
"""
Fills the network structure with the given parameters.
:param n_bits_in: number of input bits.
:param n_bits_out: number of output bits.
:param n_ones_in: number of ones in the input per sample.
:param n_ones_out: number of ones in the output per sample.
:param n_samples: number of samples. If negative, the number of samples
with the maximum information are chosen.
"""
# If "n_bits" is specified, set both n_bits_in and n_bits_out
if ("n_bits" in data) and (data["n_bits"] > 0):
self["n_bits_in"] = int(data["n_bits"])
self["n_bits_out"] = int(data["n_bits"])
self["n_bits"] = int(data["n_bits"])
else:
utils.init_key(self, data, "n_bits_in", n_bits_in, _type=int)
utils.init_key(self, data, "n_bits_out", n_bits_out, _type=int)
utils.init_key(
self, data, "n_bits", self["n_bits_in"] if self["n_bits_in"] == self["n_bits_out"] else -1, _type=int
)
utils.init_key(self, data, "n_ones_in", n_ones_in, _type=int)
utils.init_key(self, data, "n_ones_out", n_ones_out, _type=int)
utils.init_key(self, data, "n_samples", n_samples, _type=int)
utils.init_key(self, data, "algorithm", algorithm, _type=str)
if not self["algorithm"] in ["random", "balanced", "unique"]:
raise Exception(
'Invalid data generation algorithm "'
+ self["algorithm"]
+ '", must be one of {"random", "balanced", "unique"}!'
)
# If n_ones_in and n_ones_out is not given, automatically calculate
# n_ones_in, n_ones_out and n_samples
if (self["n_ones_in"] <= 0) and (self["n_ones_out"] <= 0):
params = entropy.optimal_parameters(
n_samples=self["n_samples"], n_bits_in=self["n_bits_in"], n_bits_out=self["n_bits_out"]
)
self["n_samples"] = params["n_samples"]
self["n_ones_in"] = params["n_ones_in"]
self["n_ones_out"] = params["n_ones_out"]
# Automatically choose the optimal number of samples
if self["n_samples"] <= 0:
self["n_samples"] = entropy.optimal_sample_count(
n_bits_out=self["n_bits_out"],
n_bits_in=self["n_bits_in"],
n_ones_out=self["n_ones_out"],
n_ones_in=self["n_ones_in"],
)
def __init__(
self, data={}, params={}, param_noise={}, multiplicity=1, neuron_type=pynl.TYPE_IF_COND_EXP, w=0.03, sigma_w=0.0
):
"""
Constructor of the TopologyParameters class.
:param data: is a dictionary from which the arguments are copied
preferably.
:param params: dictionary containing the neuron parameters.
:param param_noise: dictionary potentially containing a standard
deviation for each parameter.
:param multiplicity: number of neurons and signals each component is
represented with.
:param neuron_type: PyNNLess neuron type name.
:param w: synapse weight
:param sigma_w: synapse weight standard deviation.
"""
utils.init_key(self, data, "params", params)
utils.init_key(self, data, "param_noise", param_noise)
utils.init_key(self, data, "multiplicity", multiplicity)
utils.init_key(self, data, "neuron_type", neuron_type)
utils.init_key(self, data, "w", w)
utils.init_key(self, data, "sigma_w", sigma_w)
self["params"] = self._restore_alternatives(
pynl.PyNNLess.merge_default_parameters(self["params"], self["neuron_type"]), self["params"]
)
def __init__(self, data={}, vmin=0, vmax=1, count=10):
utils.init_key(self, data, "min", vmin)
utils.init_key(self, data, "max", vmax)
utils.init_key(self, data, "count", count)
def __init__(self, data={}, params={}, param_noise={}, multiplicity=1,
neuron_type=pynl.TYPE_IF_COND_EXP, w=0.03, sigma_w=0.0):
"""
Constructor of the TopologyParameters class.
:param data: is a dictionary from which the arguments are copied
preferably.
:param params: dictionary containing the neuron parameters.
:param param_noise: dictionary potentially containing a standard
deviation for each parameter.
:param multiplicity: number of neurons and signals each component is
represented with.
:param neuron_type: PyNNLess neuron type name.
:param w: synapse weight
:param sigma_w: synapse weight standard deviation.
"""
utils.init_key(self, data, "params", params)
utils.init_key(self, data, "param_noise", param_noise)
utils.init_key(self, data, "multiplicity", multiplicity)
utils.init_key(self, data, "neuron_type", neuron_type)
utils.init_key(self, data, "w", w)
utils.init_key(self, data, "sigma_w", sigma_w)
self["params"] = self._restore_alternatives(
pynl.PyNNLess.merge_default_parameters(self["params"],
self["neuron_type"]),
self["params"])
def __init__(
self, data={}, burst_size=1, time_window=100.0, isi=1.0, sigma_t=0.0, sigma_t_offs=0.0, p0=0.0, p1=0.0
):
"""
Constructor of the InputParameters class. If a dictionary or another
InputParameters class is passed as first parameter, elements from this
instance are copied with a precedence over the specified parameters.
:param data: dictionary from which parameters are copied preferably.
:param burst_size: number of spikes representing a "one".
:param time_window: time between the presentation of two input samples.
:param isi: inter spike interval between spike of a burst.
:param sigma_t: spike jitter
:param sigma_t_offs: spike offset jitter
:param p0: probability with which individual spikes are omitted.
:param p1: probability with which spikes from a "one" are randomly
introduced for a sample containing zeros.
"""
utils.init_key(self, data, "burst_size", burst_size, _type=int)
utils.init_key(self, data, "time_window", time_window)
utils.init_key(self, data, "isi", isi)
utils.init_key(self, data, "sigma_t", sigma_t)
utils.init_key(self, data, "sigma_t_offs", sigma_t_offs)
utils.init_key(self, data, "p0", p0)
utils.init_key(self, data, "p1", p1)