def _(self, data: PathCollection, order: Optional[int] = None) -> None:
# Check order
if order is not None:
self._order = order
if 0 <= self.order <= 1:
super().fit(data, order=self.order)
elif self.order > 1:
# --- START ---
nc = NodeCollection()
for node in data.nodes.values():
nc.add(node)
ec = EdgeCollection(nodes=nc)
for edge in data.edges.values():
ec.add(edge)
self._nodes = HigherOrderNodeCollection(nodes=nc, edges=ec)
# --- END ---
# get path data
paths = data
# generate first order representation of data
network = Network.from_paths(paths, frequencies=True)
self.calculate(network, paths)
else:
LOG.error('A Null Model with order %s is not supported',
self.order)
raise AttributeError
def test_NodeCollection_iter():
"""Test iter trough the node collection"""
nodes = NodeCollection()
nodes.add(['a', 'b', 'c', 'd'])
for node in nodes:
assert isinstance(node.uid, str)
for uid, node in nodes.items():
assert isinstance(uid, str)
assert isinstance(node, Node)
for uid in nodes.keys():
assert isinstance(uid, str)
for node in nodes.values():
assert isinstance(node, Node)
def test_NodeCollection():
"""Test node collection"""
nodes = NodeCollection()
assert len(nodes) == 0
a = Node('a')
nodes.add(a)
assert len(nodes) == 1
assert nodes['a'] == a
assert nodes[a] == a
assert 'a' in nodes
assert a in nodes
assert 'a' in nodes.uids
assert 'a' in nodes.keys()
assert a in nodes.values()
assert ('a', a) in nodes.items()
nodes.add(('b', 'c'))
assert len(nodes) == 3
# with pytest.raises(Exception):
# nodes.add('a')
# with pytest.raises(Exception):
# nodes.add(a)
d = Node('d', color='blue')
nodes.add(d)
assert nodes['d']['color'] == 'blue'
d['color'] = 'red'
assert nodes['d']['color'] == 'red'
nodes.add(['e', ('f', 'g'), ['h', 'i']])
assert len(nodes) == 9
nodes.remove(a)
assert len(nodes) == 8
assert a not in nodes
nodes.remove('b')
assert len(nodes) == 7
assert 'b' not in nodes
nodes.remove([('e', 'f', 'g'), 'h', ['i']])
assert len(nodes) == 2
def _(self, data: Network, order: Optional[int] = None) -> None:
# Check order
if order is not None:
self._order = order
if 0 <= self.order <= 1:
super().fit(data, order=self.order)
elif self.order > 1:
# TODO: create function to transfer base data from PathCollection object
# --- START ---
nc = NodeCollection()
for node in data.nodes.values():
nc.add(node)
ec = EdgeCollection(nodes=nc)
for edge in data.edges.values():
ec.add(edge)
self._nodes = HigherOrderNodeCollection(nodes=nc, edges=ec)
# --- END ---
# get network data
network = data
# generate a path representation of the data
paths = PathCollection(directed=network.directed,
nodes=network.nodes,
edges=network.edges)
for edge in data.edges:
paths.add(edge, frequency=edge.attributes.get('frequency', 1))
self.calculate(network, paths)
else:
LOG.error('A Null Model with order %s is not supported',
self.order)
raise AttributeError
def _(self,
data: PathCollection,
order: Optional[int] = None,
subpaths: bool = True) -> None:
if order is not None:
self._order = order
order = self.order
# TODO: create function to transfer base data from PathCollection object
# --- START ---
nc = NodeCollection()
for node in data.nodes.values():
nc.add(node)
ec = EdgeCollection(nodes=nc)
for edge in data.edges.values():
ec.add(edge)
self._nodes = HigherOrderNodeCollection(nodes=nc, edges=ec)
# --- END ---
# iterate over all paths
for path in data:
# get frequency of the observed path
# TODO: define keyword in config file
frequency = path.attributes.get('frequency', 1)
nodes: list = []
if order == 0:
for node in path.nodes:
if (node, ) not in self.nodes:
self.add_node(node, frequency=0.0)
#self.nodes[(node,)]['frequency'] += frequency
for node in path.nodes:
self.nodes[(node, )]['frequency'] += frequency
elif order == 1:
nodes.extend([tuple([n]) for n in path.nodes])
elif 1 < order <= len(path):
for subpath in self.window(path.edges, size=order - 1):
nodes.append(subpath)
elif order == len(path) + 1:
if tuple(path.edges) not in self.nodes:
self.nodes.add(tuple(path.edges))
else:
pass
_edges = []
for _v, _w in zip(nodes[:-1], nodes[1:]):
if _v not in self.nodes:
self.nodes.add(_v)
if _w not in self.nodes:
self.nodes.add(_w)
_nodes = (self.nodes[_v], self.nodes[_w])
if _nodes not in self.edges:
self.add_edge(*_nodes, possible=0, observed=0, frequency=0)
_edges.append(self.edges[_nodes])
for edge in _edges:
edge['frequency'] += frequency
if order == len(path):
edge['observed'] += frequency
else:
edge['possible'] += frequency
if order == 0:
frequencies = [n['frequency'] for n in self.nodes]
for node in self.nodes:
node['frequency'] = node['frequency'] / sum(frequencies)
if subpaths:
self._subpaths = SubPathCollection.from_paths(data,
max_length=order,
include_path=True)
def read_pathcollection(filename: str,
separator: str = ',',
frequency: bool = False,
directed: bool = True,
maxlines: int = None) -> PathCollection:
"""Read path in edgelist format
Reads data from a file containing multiple lines of *edges* of the form
"v,w,frequency,X" (where frequency is optional and X are arbitrary
additional columns). The default separating character ',' can be changed.
Parameters
----------
filename : str
path to edgelist file
separator : str
character separating the nodes
frequency : bool
is a frequency given? if ``True`` it is the last element in the
edge (i.e. ``a,b,2``)
directed : bool
are the edges directed or undirected
maxlines : int
number of lines to read (useful to test large files).
None means the entire file is read
"""
from pathpy.core.path import Path, PathCollection
nodes: dict = {}
edges: dict = {}
paths: dict = {}
with open(filename, 'r') as csv:
for n, line in enumerate(csv):
fields = line.rstrip().split(separator)
assert len(fields) >= 1, 'Error: empty line: {0}'.format(line)
if frequency:
path = tuple(fields[:-1])
freq = float(fields[-1])
else:
path = tuple(fields)
freq = 1.0
for node in path:
if node not in nodes:
nodes[node] = Node(node)
if len(path) == 1 and path not in paths:
paths[path] = Path(nodes[path[0]], frequency=freq)
else:
edge_list = []
for u, v in zip(path[:-1], path[1:]):
if (u, v) not in edges:
edges[(u, v)] = Edge(nodes[u], nodes[v])
edge_list.append(edges[(u, v)])
if path not in paths:
paths[path] = Path(*edge_list, frequency=freq)
if maxlines is not None and n >= maxlines:
break
ncoll = NodeCollection()
for node in nodes.values():
ncoll.add(node)
ecoll = EdgeCollection(nodes=ncoll)
for edge in edges.values():
ecoll._add(edge)
_paths = PathCollection(directed=directed, nodes=ncoll, edges=ecoll)
for _path in paths.values():
_paths._add(_path)
return _paths
def read_file(cls,
filename: str,
separator: str = ',',
frequency: bool = False,
directed: bool = True,
maxlines: int = None) -> None:
"""
Read path in edgelist format
Reads data from a file containing multiple lines of *edges* of the
form "v,w,frequency,X" (where frequency is optional and X are
arbitrary additional columns). The default separating character ','
can be changed.
Parameters
----------
filename : str
path to edgelist file
separator : str
character separating the nodes
frequency : bool
is a frequency given? if ``True`` it is the last element in the
edge (i.e. ``a,b,2``)
directed : bool
are the edges directed or undirected
maxlines : int
number of lines to read (useful to test large files).
None means the entire file is read
"""
nodes = {}
edges = {}
paths = {}
with open(filename, 'r') as f:
for n, line in enumerate(f):
fields = line.rstrip().split(separator)
assert len(fields) >= 2, 'Error: malformed line: {0}'.format(
line)
if frequency:
path = tuple(fields[:-1])
f = int(fields[-1])
else:
path = tuple(fields)
f = 1
for node in path:
if node not in nodes:
nodes[node] = Node(node)
edge_list = []
for u, v in zip(path[:-1], path[1:]):
if (u, v) not in edges:
edges[(u, v)] = Edge(nodes[u],
nodes[v],
uid=u + '-' + v)
edge_list.append(edges[(u, v)])
if path not in paths:
paths[path] = Path(*edge_list, frequency=f)
if maxlines is not None and n >= maxlines:
break
nc = NodeCollection()
nc.add(*nodes.values())
ec = EdgeCollection(nodes=nc)
for edge in edges.values():
ec._add(edge)
p = PathCollection(nodes=nc, edges=ec)
for path in paths.values():
p._add(path)
return p
def test_NodeCollection():
"""Test node collection"""
nodes = NodeCollection()
assert len(nodes) == 0
a = Node('a')
nodes.add(a)
assert len(nodes) == 1
assert nodes['a'] == a
assert nodes[a] == a
assert 'a' in nodes
assert a in nodes
assert 'a' in nodes.uids
assert 'a' in nodes.keys()
assert a in nodes.values()
assert ('a', a) in nodes.items()
assert {'a': a} == nodes.dict
nodes.add('b', 'c')
print(nodes)
assert len(nodes) == 3
with pytest.raises(Exception):
nodes.add('a')
with pytest.raises(Exception):
nodes.add(a)
d = Node('d', color='blue')
nodes.add(d)
assert nodes['d']['color'] == 'blue'
d['color'] = 'red'
assert nodes['d']['color'] == 'red'
nodes.add('e', ('f', 'g'), ['h', 'i'])
assert len(nodes) == 9
nodes.remove(('e', 'f', 'g'), 'h', ['i'])
assert len(nodes) == 4