Graph classes#
NNGT provides four main graph classes that provide specific features to work as conveniently as possible with different object types: topological versus spaceembedded graphs or neuronal networks.

The basic graph class, which inherits from a library class such as 

The detailed class that inherits from 

The detailed class that inherits from 

Class that inherits from 
A summary of the methods provided by these classes as well as more detailed
descriptions are provided below.
Unless specified, child classes can use all methods from the parent class
(the only exception is set_types()
which is not available
to the Network
subclasses).
Summary of the class members and methods#
Graph#
The main class for topological graphs.

The basic graph class, which inherits from a library class such as 

Return the graph adjacency matrix. 
Remove all edges from the graph 

Returns a deepcopy of the current 


Remove a list of edges 

Remove nodes (and associated edges) from the graph. 
Access edge attributes. 


Return the ID a given edge or a list of edges in the graph. 
Number of edges in the graph 

Edges of the graph, sorted by order of creation, as an array of 2tuple. 


Import a saved graph from a file. 

Create a 

Creates a 

Return the type of an attribute (e.g. 

Returns the normalized betweenness centrality of the nodes and edges. 

Degree sequence of all the nodes. 

Returns the delays of all or a subset of the edges. 

Density of the graph. 

Attributes of the graph's edges. 

Return the type of all or a subset of the edges. 

Return the edges in the network fulfilling a given condition. 

Attributes of the graph's edges. 

Return the nodes in the network fulfilling a given condition. 
Return a coarsegrained version of the graph containing one node per 


Returns the weights of all or a subset of the edges. 
Returns the underlying library object. 

Unique 


Whether edge is present in the graph. 

Return whether the graph is connected. 
Whether the graph is directed or not 

Whether the graph is a subclass of 

Whether the graph is embedded in space (i.e. 

Whether the edges have weights 


Turn a 

Turn a 
Name of the graph. 


Return the neighbours of node. 

Adding a connection to the graph, with optional properties. 

Create a new attribute for the edges. 

Add a list of edges to the graph. 

Adding a node to the graph, with optional properties. 

Create a new attribute for the nodes. 
Access node attributes. 

Number of nodes in the graph 

Returns the number of alive instances. 


Set the delay for spike propagation between neurons. 

Set attributes to the connections between neurons. 

Set graph name 

Set attributes to the connections between neurons. 

Set the synaptic/connection types. 

Set the synaptic weights. 
Object structuring the graph into specific groups. 


Save graph to file; options detailed below. 

Convert the graph to its undirected variant. 
Type of the graph. 
SpatialGraph#
Subclass of Graph
providing additional tools to work with
spatial graphs. It works together with the Shape
object
from the geometry
module.

The detailed class that inherits from 

Returns a copy of the nodes' positions as a (N, 2) array. 

Set the nodes' positions as a (N, 2) array. 
The environment's spatial structure. 
Network#
Subclass of Graph
providing additional tools to work with
neuronal networks. It works together with the
NeuralPop
object.

The detailed class that inherits from 

Generate a network containing a population of two neural groups: inhibitory and excitatory neurons. 

Generate a network from gids. 

Return the type of the neurons (+1 for excitatory, 1 for inhibitory). 
Return the ids of the nodes in the 




Properties of a neuron in the graph. 
Returns the number of alive instances. 




Send the network to NEST. 

Generate a network containing only one type of neurons. 
SpatialNetwork#
Subclass of Graph
providing additional tools to work with
spatial neuronal networks. It works together with both
NeuralPop
and the
Shape
object from the geometry
module.

Class that inherits from 

Generate a network containing a population of two neural groups: inhibitory and excitatory neurons. 

Generate a network from gids. 

Return the type of the neurons (+1 for excitatory, 1 for inhibitory). 

Returns a copy of the nodes' positions as a (N, 2) array. 

Return the ids of the nodes in the 



Properties of a neuron in the graph. 

Returns the number of alive instances. 



Set the nodes' positions as a (N, 2) array. 

The environment's spatial structure. 

Send the network to NEST. 

Generate a network containing only one type of neurons. 
Details#
 class nngt.Graph(*args, **kwargs)[source]#
The basic graph class, which inherits from a library class such as
graph_tool.Graph
,networkx.DiGraph
, origraph.Graph
.The objects provides several functions to easily access some basic properties.
Initialize Graph instance
Changed in version 2.0: Renamed from_graph to copy_graph.
Changed in version 2.2: Added structure argument.
 Parameters:
nodes (int, optional (default: 0)) – Number of nodes in the graph.
name (string, optional (default: “Graph”)) – The name of this
Graph
instance.weighted (bool, optional (default: True)) – Whether the graph edges have weight properties.
directed (bool, optional (default: True)) – Whether the graph is directed or undirected.
copy_graph (
Graph
, optional) – An optionalGraph
that will be copied.structure (
Structure
, optional (default: None)) – A structure dividing the graph into specific groups, which can be used to generate specific connectivities and visualise the connections in a more coarsegrained manner.kwargs (optional keywords arguments) – Optional arguments that can be passed to the graph, e.g. a dict containing information on the synaptic weights (
weights={"distribution": "constant", "value": 2.3}
which is equivalent toweights=2.3
), the synaptic delays, or atype
information.
Note
When using copy_graph, only the topological properties are copied (nodes, edges, and attributes), spatial and biological properties are ignored. To copy a graph exactly, use
copy()
. Returns:
self (
Graph
)
 adjacency_matrix(types=False, weights=False, mformat='csr')[source]#
Return the graph adjacency matrix.
Note
Source nodes are represented by the rows, targets by the corresponding columns.
 Parameters:
types (bool, optional (default: False)) – Wether the edge types should be taken into account (negative values for inhibitory connections).
weights (bool or string, optional (default: False)) – Whether the adjacecy matrix should be weighted. If True, all connections are multiply bythe associated synaptic strength; if weight is a string, the connections are scaled bythe corresponding edge attribute.
mformat (str, optional (default: “csr”)) – Type of
scipy.sparse
matrix that will be returned, by defaultscipy.sparse.csr_matrix
.
 Returns:
mat (
scipy.sparse
matrix) – The adjacency matrix of the graph.
 clear_all_edges()#
Remove all edges from the graph
 delete_edges(edges)#
Remove a list of edges
 delete_nodes(nodes)#
Remove nodes (and associated edges) from the graph.
 property edge_attributes#
Access edge attributes.
 edge_id(edge)#
Return the ID a given edge or a list of edges in the graph. Raises an error if the edge is not in the graph or if one of the vertices in the edge is nonexistent.
 Parameters:
edge (2tuple or array of edges) – Edge descriptor (source, target).
 Returns:
index (int or array of ints) – Index of the given edge.
 edge_nb()#
Number of edges in the graph
 property edges_array#
Edges of the graph, sorted by order of creation, as an array of 2tuple.
 static from_file(filename, fmt='auto', separator=' ', secondary=';', attributes=None, attributes_types=None, notifier='@', ignore='#', from_string=False, name=None, directed=True, cleanup=False)[source]#
Import a saved graph from a file.
Changed in version 2.0: Added optional attributes_types and cleanup arguments.
 Parameters:
filename (str) – The path to the file.
fmt (str, optional (default: deduced from filename)) – The format used to save the graph. Supported formats are: “neighbour” (neighbour list), “ssp” (scipy.sparse), “edge_list” (list of all the edges in the graph, one edge per line, represented by a
source target
pair), “gml” (gml format, default if filename ends with ‘.gml’), “graphml” (graphml format, default if filename ends with ‘.graphml’ or ‘.xml’), “dot” (dot format, default if filename ends with ‘.dot’), “gt” (only when using graph_tool as library, detected if filename ends with ‘.gt’).separator (str, optional (default “ “)) – separator used to separate inputs in the case of custom formats (namely “neighbour” and “edge_list”)
secondary (str, optional (default: “;”)) – Secondary separator used to separate attributes in the case of custom formats.
attributes (list, optional (default: [])) – List of names for the attributes present in the file. If a notifier is present in the file, names will be deduced from it; otherwise the attributes will be numbered. For “edge_list”, attributes may also be present as additional columns after the source and the target.
attributes_types (dict, optional (default: str)) – Backup information if the type of the attributes is not specified in the file. Values must be callables (types or functions) that will take the argument value as a string input and convert it to the proper type.
notifier (str, optional (default: “@”)) – Symbol specifying the following as meaningfull information. Relevant information are formatted
@info_name=info_value
, whereinfo_name
is in (“attributes”, “directed”, “name”, “size”) and associatedinfo_value
are of type (list
,bool
,str
,int
). Additional notifiers are@type=SpatialGraph/Network/SpatialNetwork
, which must be followed by the relevant notifiers among@shape
,@population
, and@graph
.from_string (bool, optional (default: False)) – Load from a string instead of a file.
ignore (str, optional (default: “#”)) – Ignore lines starting with the ignore string.
name (str, optional (default: from file information or ‘LoadedGraph’)) – The name of the graph.
directed (bool, optional (default: from file information or True)) – Whether the graph is directed or not.
cleanup (bool, optional (default: False)) – If true, removes nodes before the first one that appears in the edges and after the last one and renumber the nodes from 0.
 Returns:
graph (
Graph
or subclass) – Loaded graph.
 classmethod from_library(library_graph, name='ImportedGraph', weighted=True, directed=True, **kwargs)[source]#
Create a
Graph
by wrapping a graph object from one of the supported libraries. Parameters:
library_graph (object) – Graph object from one of the supported libraries (graphtool, igraph, networkx).
name (str, optional (default: “ImportedGraph”))
**kwargs – Other standard arguments (see
__init__()
)
 classmethod from_matrix(matrix, weighted=True, directed=True, population=None, shape=None, positions=None, name=None, **kwargs)[source]#
Creates a
Graph
from ascipy.sparse
matrix or a dense matrix. Parameters:
matrix (
scipy.sparse
matrix ornumpy.ndarray
) – Adjacency matrix.weighted (bool, optional (default: True)) – Whether the graph edges have weight properties.
directed (bool, optional (default: True)) – Whether the graph is directed or undirected.
population (
NeuralPop
) – Population to associate to the newNetwork
.shape (
Shape
, optional (default: None)) – Shape to associate to the newSpatialGraph
.positions ((N, 2) array) – Positions, in a 2D space, of the N neurons.
name (str, optional) – Graph name.
 Returns:
 get_attribute_type(attribute_name, attribute_class=None)[source]#
Return the type of an attribute (e.g. string, double, int).
 Parameters:
attribute_name (str) – Name of the attribute.
attribute_class (str, optional (default: both)) – Whether attribute_name is a “node” or an “edge” attribute.
 Returns:
type (str) – Type of the attribute.
 get_betweenness(btype='both', weights=None)[source]#
Returns the normalized betweenness centrality of the nodes and edges.
 Parameters:
g (
Graph
) – Graph to analyze.btype (str, optional (default ‘both’)) – The centrality that should be returned (either ‘node’, ‘edge’, or ‘both’). By default, both betweenness centralities are computed.
weights (bool or str, optional (default: binary edges)) – Whether edge weights should be considered; if
None
orFalse
then use binary edges; ifTrue
, uses the ‘weight’ edge attribute, otherwise uses any valid edge attribute required.
 Returns:
nb (
numpy.ndarray
) – The nodes’ betweenness if btype is ‘node’ or ‘both’eb (
numpy.ndarray
) – The edges’ betweenness if btype is ‘edge’ or ‘both’
See also
 get_degrees(mode='total', nodes=None, weights=None, edge_type='all')[source]#
Degree sequence of all the nodes.
Changed in version 2.0: Changed deg_type to mode, node_list to nodes, use_weights to weights, and edge_type to edge_type.
 Parameters:
mode (string, optional (default: “total”)) – Degree type (among ‘in’, ‘out’ or ‘total’).
nodes (list, optional (default: None)) – List of the nodes which degree should be returned
weights (bool or str, optional (default: binary edges)) – Whether edge weights should be considered; if
None
orFalse
then use binary edges; ifTrue
, uses the ‘weight’ edge attribute, otherwise uses any valid edge attribute required.edge_type (int or str, optional (default: all)) – Restrict to a given synaptic type (“excitatory”, 1, or “inhibitory”, 1), using either the “type” edge attribute for non
Network
or theinhibitory
nodes.
 Returns:
degrees (
numpy.array
).. warning :: – When using MPI with “nngt” (distributed) backend, returns only the degrees associated to local edges. “Complete” degrees are obtained by taking the sum of the results on all MPI processes.
 get_delays(edges=None)[source]#
Returns the delays of all or a subset of the edges.
Changed in version 1.0.1: Added the possibility to ask for a subset of edges.
 Parameters:
edges ((E, 2) array, optional (default: all edges)) – Edges for which the type should be returned.
 Returns:
the list of delays
 get_density(ignore_loops=True)[source]#
Density of the graph.
 Parameters:
ignore_loops (bool, optional (default: True)) – Whether selfloops should be considered.
Note
The density is computed via if ignore_loops is True, or via if it is False. E is the number of edges, N the number of nodes, and d is 0 if the graph is undirected and 1 if it is directed.
 get_edge_attributes(edges=None, name=None)[source]#
Attributes of the graph’s edges.
 Parameters:
edges (tuple or list of tuples, optional (default:
None
)) – Edge whose attribute should be displayed.name (str, optional (default:
None
)) – Name of the desired attribute.
 Returns:
Dict containing all graph’s attributes (synaptic weights, delays…)
by default. If edge is specified, returns only the values for these
edges. If name is specified, returns value of the attribute for each
edge.
Note
The attributes values are ordered as the edges in
edges_array()
if edges is None.
 get_edge_types(edges=None)[source]#
Return the type of all or a subset of the edges.
 Parameters:
edges ((E, 2) array, optional (default: all edges)) – Edges for which the type should be returned.
 Returns:
the list of types (1 for excitatory, 1 for inhibitory)
 get_edges(attribute=None, value=None, source_node=None, target_node=None)[source]#
Return the edges in the network fulfilling a given condition.
For undirected graphs, edges are always returned in the order where .
Warning
Contrary to
edges_array()
that returns edges ordered by creation time (i.e. corresponding to the order of the edge attribute array), this function does not enforce any specific edge order. This also means that, if order does not matter, it may be faster to callget_edges
that to calledges_array
. Parameters:
attribute (str, optional (default: all nodes)) – Whether the attribute of the returned edges should have a specific value.
value (object, optional (default : None)) – If an attribute name is passed, then only edges with attribute being equal to value will be returned.
source_node (int or list of ints, optional (default: all nodes)) – Retrict the edges to those stemming from source_node.
target_node (int or list of ints, optional (default: all nodes)) – Retrict the edges to those arriving at target_node.
 Returns:
A list of edges (2tuples).
See also
 get_node_attributes(nodes=None, name=None)[source]#
Attributes of the graph’s edges.
Changed in version 1.0.1: Corrected default behavior and made it the same as
get_edge_attributes()
.New in version 0.9.
 Parameters:
nodes (list of ints, optional (default:
None
)) – Nodes whose attribute should be displayed.name (str, optional (default:
None
)) – Name of the desired attribute.
 Returns:
Dict containing all nodes attributes by default. If nodes is
specified, returns a
dict
containing only the attributes of thesenodes. If name is specified, returns a list containing the values of
the specific attribute for the required nodes (or all nodes if
unspecified).
See also
get_edge_attributes()
,new_node_attribute()
,set_node_attribute()
,new_edge_attributes()
,set_edge_attribute()
 get_nodes(attribute=None, value=None)[source]#
Return the nodes in the network fulfilling a given condition.
 Parameters:
attribute (str, optional (default: all nodes)) – Whether the attribute of the returned nodes should have a specific value.
value (object, optional (default : None)) – If an attribute name is passed, then only nodes with attribute being equal to value will be returned.
See also
 get_structure_graph()[source]#
Return a coarsegrained version of the graph containing one node per
nngt.Group
. Connections between groups are associated to the sum of all connection weights. If no structure is present, returns an empty Graph.
 get_weights(edges=None)[source]#
Returns the weights of all or a subset of the edges.
Changed in version 1.0.1: Added the possibility to ask for a subset of edges.
 Parameters:
edges ((E, 2) array, optional (default: all edges)) – Edges for which the type should be returned.
 Returns:
the list of weights
 property graph#
Returns the underlying library object.
Warning
Do not add or remove edges directly through this object.
 has_edge(edge)#
Whether edge is present in the graph.
New in version 2.0.
 is_connected(mode='strong')[source]#
Return whether the graph is connected.
 Parameters:
mode (str, optional (default: “strong”)) – Whether to test connectedness with directed (“strong”) or undirected (“weak”) connections.
References
[igconnected]
 is_network()[source]#
Whether the graph is a subclass of
Network
(i.e. if it has aNeuralPop
attribute).
 is_spatial()[source]#
Whether the graph is embedded in space (i.e. is a subclass of
SpatialGraph
).
 static make_network(graph, neural_pop, copy=False, **kwargs)[source]#
Turn a
Graph
object into aNetwork
, or aSpatialGraph
into aSpatialNetwork
. Parameters:
graph (
Graph
orSpatialGraph
) – Graph to convertneural_pop (
NeuralPop
) – Population to associate to the newNetwork
copy (bool, optional (default:
False
)) – Whether the operation should be made inplace on the object or if a new object should be returned.
Notes
Inplace operation that directly converts the original graph if copy is
False
, else returns the copiedGraph
turned into aNetwork
.
 static make_spatial(graph, shape=None, positions=None, copy=False)[source]#
Turn a
Graph
object into aSpatialGraph
, or aNetwork
into aSpatialNetwork
. Parameters:
graph (
Graph
orSpatialGraph
) – Graph to convert.shape (
Shape
, optional (default: None)) – Shape to associate to the newSpatialGraph
.positions ((N, 2) array) – Positions, in a 2D space, of the N neurons.
copy (bool, optional (default:
False
)) – Whether the operation should be made inplace on the object or if a new object should be returned.
Notes
Inplace operation that directly converts the original graph if copy is
False
, else returns the copiedGraph
turned into aSpatialGraph
. The shape argument can be skipped if positions are given; in that case, the neurons will be embedded in a rectangle that contains them all.
 property name#
Name of the graph.
 neighbours(node, mode='all')[source]#
Return the neighbours of node.
 Parameters:
node (int) – Index of the node of interest.
mode (string, optional (default: “all”)) – Type of neighbours that will be returned: “all” returns all the neighbours regardless of directionality, “in” returns the inneighbours (also called predecessors) and “out” retruns the outneighbours (or successors).
 Returns:
neighbours (set) – The neighbours of node.
 new_edge(source, target, attributes=None, ignore=False, self_loop=False)#
Adding a connection to the graph, with optional properties.
Changed in version 2.0: Added self_loop argument to enable adding selfloops.
 Parameters:
source (
int/node
) – Source node.target (
int/node
) – Target node.attributes (
dict
, optional (default:{}
)) – Dictionary containing optional edge properties. If the graph is weighted, defaults to{"weight": 1.}
, the unit weight for the connection (synaptic strength in NEST).ignore (bool, optional (default: False)) – If set to True, ignore attempts to add an existing edge and accept selfloops; otherwise an error is raised.
self_loop (bool, optional (default: False)) – Whether to allow selfloops or not.
 Returns:
The new connection or None if nothing was added.
 new_edge_attribute(name, value_type, values=None, val=None)[source]#
Create a new attribute for the edges.
 Parameters:
name (str) – The name of the new attribute.
value_type (str) – Type of the attribute, among ‘int’, ‘double’, ‘string’, or ‘object’
values (array, optional (default: None)) – Values with which the edge attribute should be initialized. (must have one entry per node in the graph)
val (int, float or str , optional (default: None)) – Identical value for all edges.
 new_edges(edge_list, attributes=None, check_duplicates=False, check_self_loops=True, check_existing=True, ignore_invalid=False)#
Add a list of edges to the graph.
Changed in version 2.0: Can perform all possible checks before adding new edges via the
check_duplicates
check_self_loops
, andcheck_existing
arguments. Parameters:
edge_list (list of 2tuples or np.array of shape (edge_nb, 2)) – List of the edges that should be added as tuples (source, target)
attributes (
dict
, optional (default:{}
)) – Dictionary containing optional edge properties. If the graph is weighted, defaults to{"weight": ones}
, whereones
is an array the same length as the edge_list containing a unit weight for each connection (synaptic strength in NEST).check_duplicates (bool, optional (default: False)) – Check for duplicate edges within edge_list.
check_self_loops (bool, optional (default: True)) – Check for selfloops.
check_existing (bool, optional (default: True)) – Check whether some of the edges in edge_list already exist in the graph or exist multiple times in edge_list (also performs check_duplicates).
ignore_invalid (bool, optional (default: False)) – Ignore invalid edges: they are not added to the graph and are silently dropped. Unless this is set to true, an error is raised whenever one of the three checks fails.
.. warning:: – Setting check_existing to False will lead to undefined behavior if existing edges are provided! Only use it (for speedup) if you are sure that you are indeed only adding new edges.
 Returns:
Returns new edges only.
 new_node(n=1, neuron_type=1, attributes=None, value_types=None, positions=None, groups=None)#
Adding a node to the graph, with optional properties.
 Parameters:
n (int, optional (default: 1)) – Number of nodes to add.
neuron_type (int, optional (default: 1)) – Type of neuron (1 for excitatory, 1 for inhibitory)
attributes (dict, optional (default: None)) – Dictionary containing the attributes of the nodes.
value_types (dict, optional (default: None)) – Dict of the attributes types, necessary only if the attributes do not exist yet.
positions (array of shape (n, 2), optional (default: None)) – Positions of the neurons. Valid only for
SpatialGraph
orSpatialNetwork
.groups (str, int, or list, optional (default: None)) –
NeuralGroup
to which the neurons belong. Valid only forNetwork
orSpatialNetwork
.
 Returns:
The node or a list of the nodes created.
 new_node_attribute(name, value_type, values=None, val=None)[source]#
Create a new attribute for the nodes.
 Parameters:
name (str) – The name of the new attribute.
value_type (str) – Type of the attribute, among ‘int’, ‘double’, ‘string’, or ‘object’
values (array, optional (default: None)) – Values with which the node attribute should be initialized. (must have one entry per node in the graph)
val (int, float or str , optional (default: None)) – Identical value for all nodes.
 property node_attributes#
Access node attributes.
 node_nb()#
Number of nodes in the graph
 set_delays(delay=None, elist=None, distribution=None, parameters=None, noise_scale=None)[source]#
Set the delay for spike propagation between neurons.
 Parameters:
delay (float or class:numpy.array, optional (default: None)) – Value or list of delays (for user defined delays).
elist (class:numpy.array, optional (default: None)) – List of the edges (for user defined delays).
distribution (class:string, optional (default: None)) – Type of distribution (choose among “constant”, “uniform”, “gaussian”, “lognormal”, “lin_corr”, “log_corr”).
parameters (dict, optional (default: {})) – Dictionary containing the properties of the delay distribution.
noise_scale (class:int, optional (default: None)) – Scale of the multiplicative Gaussian noise that should be applied on the delays.
 set_edge_attribute(attribute, values=None, val=None, value_type=None, edges=None)[source]#
Set attributes to the connections between neurons.
Warning
The special “type” attribute cannot be modified when using graphs that inherit from the
Network
class. This is because for biological networks, neurons make only one kind of synapse, which is determined by thenngt.NeuralGroup
they belong to. Parameters:
attribute (str) – The name of the attribute.
value_type (str) – Type of the attribute, among ‘int’, ‘double’, ‘string’
values (array, optional (default: None)) – Values with which the edge attribute should be initialized. (must have one entry per node in the graph)
val (int, float or str , optional (default: None)) – Identical value for all edges.
value_type (str, optional (default: None)) – Type of the attribute, among ‘int’, ‘double’, ‘string’. Only used if the attribute does not exist and must be created.
edges (list of edges or array of shape (E, 2), optional (default: all)) – Edges whose attributes should be set. Others will remain unchanged.
 set_node_attribute(attribute, values=None, val=None, value_type=None, nodes=None)[source]#
Set attributes to the connections between neurons.
 Parameters:
attribute (str) – The name of the attribute.
value_type (str) – Type of the attribute, among ‘int’, ‘double’, ‘string’
values (array, optional (default: None)) – Values with which the edge attribute should be initialized. (must have one entry per node in the graph)
val (int, float or str , optional (default: None)) – Identical value for all edges.
value_type (str, optional (default: None)) – Type of the attribute, among ‘int’, ‘double’, ‘string’. Only used if the attribute does not exist and must be created.
nodes (list of nodes, optional (default: all)) – Nodes whose attributes should be set. Others will remain unchanged.
 set_types(edge_type, nodes=None, fraction=None)[source]#
Set the synaptic/connection types.
Changed in version 2.0: Changed syn_type to edge_type.
Warning
The special “type” attribute cannot be modified when using graphs that inherit from the
Network
class. This is because for biological networks, neurons make only one kind of synapse, which is determined by thenngt.NeuralGroup
they belong to. Parameters:
edge_type (int, string, or array of ints) – Type of the connection among ‘excitatory’ (also 1) or ‘inhibitory’ (also 1).
nodes (int, float or list, optional (default: None)) – If nodes is an int, number of nodes of the required type that will be created in the graph (all connections from inhibitory nodes are inhibitory); if it is a float, ratio of edge_type nodes in the graph; if it is a list, ids of the edge_type nodes.
fraction (float, optional (default: None)) – Fraction of the selected edges that will be set as edge_type (if nodes is not None, it is the fraction of the specified nodes’ edges, otherwise it is the fraction of all edges in the graph).
 Returns:
t_list (
numpy.ndarray
) – List of the types in an order that matches the edges attribute of the graph.
 set_weights(weight=None, elist=None, distribution=None, parameters=None, noise_scale=None)[source]#
Set the synaptic weights.
 Parameters:
weight (float or class:numpy.array, optional (default: None)) – Value or list of the weights (for user defined weights).
elist (class:numpy.array, optional (default: None)) – List of the edges (for user defined weights).
distribution (class:string, optional (default: None)) – Type of distribution (choose among “constant”, “uniform”, “gaussian”, “lognormal”, “lin_corr”, “log_corr”).
parameters (dict, optional (default: {})) – Dictionary containing the properties of the weight distribution. Properties are as follow for the distributions
‘constant’: ‘value’
‘uniform’: ‘lower’, ‘upper’
‘gaussian’: ‘avg’, ‘std’
‘lognormal’: ‘position’, ‘scale’
noise_scale (class:int, optional (default: None)) – Scale of the multiplicative Gaussian noise that should be applied on the weights.
Note
If distribution and parameters are provided and the weights are set for the whole graph (elist is None), then the distribution properties will be kept as the new default for subsequent edges. That is, if new edges are created without specifying their weights, then these new weights will automatically be drawn from this previous distribution.
 property structure#
Object structuring the graph into specific groups.
Note
Points to
population
if the graph is aNetwork
.
 to_file(filename, fmt='auto', separator=' ', secondary=';', attributes=None, notifier='@')[source]#
Save graph to file; options detailed below.
See also
nngt.lib.save_to_file()
 to_undirected(combine_numeric_eattr='sum')[source]#
Convert the graph to its undirected variant.
Note
All nonnumeric edge attributes will be discarded from the returned undirected graph.
 Parameters:
combine_numeric_eattr (str, optional (default: “sum”)) – How to combine numeric attributes from reciprocal edges. Can be either:
“sum” (attributes are summed)
“min” (smallest value is kept)
“max” (largest value is kept)
“mean” (the average of both attributes is taken)
In addition, combine_numeric_eattr can be a dictionary with one entry for each edge attribute.
 property type#
Type of the graph.
 class nngt.SpatialGraph(*args, **kwargs)[source]#
The detailed class that inherits from
Graph
and implements additional properties to describe spatial graphs (i.e. graph where the structure is embedded in space.Initialize SpatialClass instance.
 Parameters:
nodes (int, optional (default: 0)) – Number of nodes in the graph.
name (string, optional (default: “Graph”)) – The name of this
Graph
instance.weighted (bool, optional (default: True)) – Whether the graph edges have weight properties.
directed (bool, optional (default: True)) – Whether the graph is directed or undirected.
shape (
Shape
, optional (default: None)) – Shape of the neurons’ environment (None leads to a square of side 1 cm)positions (
numpy.array
(N, 2), optional (default: None)) – Positions of the neurons; if not specified and nodes is not 0, then neurons will be reparted at random inside theShape
object of the instance.**kwargs (keyword arguments for
Graph
or) –Shape
if no shape was given.
 Returns:
self (
SpatialGraph
)
 get_positions(nodes=None)[source]#
Returns a copy of the nodes’ positions as a (N, 2) array.
 Parameters:
nodes (int or arraylike, optional (default: all nodes)) – List of the nodes for which the position should be returned.
 set_positions(positions, nodes=None)[source]#
Set the nodes’ positions as a (N, 2) array.
 Parameters:
positions (arraylike) – List of positions, of shape (N, 2).
nodes (int or arraylike, optional (default: all nodes)) – List of the nodes for which the position should be set.
 property shape#
The environment’s spatial structure.
 class nngt.Network(*args, **kwargs)[source]#
The detailed class that inherits from
Graph
and implements additional properties to describe various biological functions and interact with the NEST simulator.Initializes
Network
instance. Parameters:
nodes (int, optional (default: 0)) – Number of nodes in the graph.
name (string, optional (default: “Graph”)) – The name of this
Graph
instance.weighted (bool, optional (default: True)) – Whether the graph edges have weight properties.
directed (bool, optional (default: True)) – Whether the graph is directed or undirected.
copy_graph (
GraphObject
, optional (default: None)) – An optionalGraphObject
to serve as base.population (
nngt.NeuralPop
, (default: None)) – An object containing the neural groups and their properties: model(s) to use in NEST to simulate the neurons as well as their parameters.inh_weight_factor (float, optional (default: 1.)) – Factor to apply to inhibitory synapses, to compensate for example the strength difference due to timescales between excitatory and inhibitory synapses.
 Returns:
self (
Network
)
 classmethod exc_and_inhib(size, iratio=0.2, en_model='aeif_cond_alpha', en_param=None, in_model='aeif_cond_alpha', in_param=None, syn_spec=None, **kwargs)[source]#
Generate a network containing a population of two neural groups: inhibitory and excitatory neurons.
 Parameters:
size (int) – Number of neurons in the network.
i_ratio (double, optional (default: 0.2)) – Ratio of inhibitory neurons: .
en_model (string, optional (default: ‘aeif_cond_alpha’)) – Nest model for the excitatory neuron.
en_param (dict, optional (default: {})) – Dictionary of parameters for the the excitatory neuron.
in_model (string, optional (default: ‘aeif_cond_alpha’)) – Nest model for the inhibitory neuron.
in_param (dict, optional (default: {})) – Dictionary of parameters for the the inhibitory neuron.
syn_spec (dict, optional (default: static synapse)) – Dictionary containg a directed edge between groups as key and the associated synaptic parameters for the postsynaptic neurons (i.e. those of the second group) as value. If provided, all connections between groups will be set according to the values contained in syn_spec. Valid keys are:
(‘excitatory’, ‘excitatory’)
(‘excitatory’, ‘inhibitory’)
(‘inhibitory’, ‘excitatory’)
(‘inhibitory’, ‘inhibitory’)
 Returns:
net (
Network
or subclass) – Network of disconnected excitatory and inhibitory neurons.
See also
 classmethod from_gids(gids, get_connections=True, get_params=False, neuron_model='aeif_cond_alpha', neuron_param=None, syn_model='static_synapse', syn_param=None, **kwargs)[source]#
Generate a network from gids.
Warning
Unless get_connections and get_params is True, or if your population is homogeneous and you provide the required information, the information contained by the network and its population attribute will be erroneous! To prevent conflicts the
to_nest()
function is not available. If you know what you are doing, you should be able to find a workaround… Parameters:
gids (arraylike) – Ids of the neurons in NEST or simply user specified ids.
get_params (bool, optional (default: True)) – Whether the parameters should be obtained from NEST (can be very slow).
neuron_model (string, optional (default: None)) – Name of the NEST neural model to use when simulating the activity.
neuron_param (dict, optional (default: {})) – Dictionary containing the neural parameters; the default value will make NEST use the default parameters of the model.
syn_model (string, optional (default: ‘static_synapse’)) – NEST synaptic model to use when simulating the activity.
syn_param (dict, optional (default: {})) – Dictionary containing the synaptic parameters; the default value will make NEST use the default parameters of the model.
 Returns:
net (
Network
or subclass) – Uniform network of disconnected neurons.
 get_edge_types(edges=None)[source]#
Return the type of all or a subset of the edges. For all edges, the types are ordered according to the edges ids, i.e. in the same order as :property:`~nngt.Graph.edges_array`.
Changed in version 2.4: Updated it to make it compatible with the default
Graph
function, including the edges argument. Parameters:
edges ((E, 2) array, optional (default: all edges)) – Edges for which the type should be returned.
 Returns:
the list of types (1 for excitatory, 1 for inhibitory)
 get_neuron_type(neuron_ids)[source]#
Return the type of the neurons (+1 for excitatory, 1 for inhibitory).
 Parameters:
neuron_ids (int or tuple) – NEST gids.
 Returns:
ids (int or tuple) – Ids in the network. Same type as the requested gids type.
 id_from_nest_gid(gids)[source]#
Return the ids of the nodes in the
nngt.Network
instance from the corresponding NEST gids. Parameters:
gids (int or tuple) – NEST gids.
 Returns:
ids (int or tuple) – Ids in the network. Same type as the requested gids type.
 neuron_properties(idx_neuron)[source]#
Properties of a neuron in the graph.
 Parameters:
idx_neuron (int) – Index of a neuron in the graph.
 Returns:
dict of the neuron’s properties.
 set_types(edge_type, nodes=None, fraction=None)[source]#
Warning
This function is not available for
Network
subclasses.
 to_nest(send_only=None, weights=True)[source]#
Send the network to NEST.
See also
make_nest_network()
for parameters
 classmethod uniform(size, neuron_model='aeif_cond_alpha', neuron_param=None, syn_model='static_synapse', syn_param=None, **kwargs)[source]#
Generate a network containing only one type of neurons.
 Parameters:
size (int) – Number of neurons in the network.
neuron_model (string, optional (default: ‘aief_cond_alpha’)) – Name of the NEST neural model to use when simulating the activity.
neuron_param (dict, optional (default: {})) – Dictionary containing the neural parameters; the default value will make NEST use the default parameters of the model.
syn_model (string, optional (default: ‘static_synapse’)) – NEST synaptic model to use when simulating the activity.
syn_param (dict, optional (default: {})) – Dictionary containing the synaptic parameters; the default value will make NEST use the default parameters of the model.
 Returns:
net (
Network
or subclass) – Uniform network of disconnected neurons.
 class nngt.SpatialNetwork(*args, **kwargs)[source]#
Class that inherits from
Network
andSpatialGraph
to provide a detailed description of a real neural network in space, i.e. with positions and biological properties to interact with NEST.Initialize SpatialNetwork instance.
 Parameters:
name (string, optional (default: “Graph”)) – The name of this
Graph
instance.weighted (bool, optional (default: True)) – Whether the graph edges have weight properties.
directed (bool, optional (default: True)) – Whether the graph is directed or undirected.
shape (
Shape
, optional (default: None)) – Shape of the neurons’ environment (None leads to a square of side 1 cm)positions (
numpy.array
, optional (default: None)) – Positions of the neurons; if not specified and nodes != 0, then neurons will be reparted at random inside theShape
object of the instance.population (class:~nngt.NeuralPop, optional (default: None)) – Population from which the network will be built.
 Returns:
self (
SpatialNetwork
)