NetworkxTransformFeatures

class caldera.transforms.networkx.NetworkxTransformFeatures(node_transform=None, edge_transform=None, global_transform=None)

Transform networkx feature data.

def only_self_loops(edges):
    for e1, e2, edata in edges:
        if e1 == e2:
            yield e1, e2, edata

transform = nx_transform(edge_transform=only_self_loops)
transform(graphs)

Alternatively, using the functional programming module:

from caldera.utils.functional import Functional

only_self_loops = Fn.filter_each(lambda x: x[0] == x[1])
transform = nx_transform(edge_transform=only_self_loops)
transform(graphs)

Parameters

• node_transform (Optional[Callable[[Generator[Tuple, None, None]], Generator[Tuple, None, None]]]) –

• edge_transform (Optional[Callable[[Generator[Tuple, None, None]], Generator[Tuple, None, None]]]) –

• global_transform (Optional[Callable[[Generator[Tuple, None, None]], Generator[Tuple, None, None]]]) –

• kwargs –

Returns

__init__(node_transform=None, edge_transform=None, global_transform=None)

Transform networkx feature data.

def only_self_loops(edges):
    for e1, e2, edata in edges:
        if e1 == e2:
            yield e1, e2, edata

transform = nx_transform(edge_transform=only_self_loops)
transform(graphs)

Alternatively, using the functional programming module:

from caldera.utils.functional import Functional

only_self_loops = Fn.filter_each(lambda x: x[0] == x[1])
transform = nx_transform(edge_transform=only_self_loops)
transform(graphs)

Parameters

• node_transform (Optional[Callable[[Generator[Tuple, None, None]], Generator[Tuple, None, None]]]) –

• edge_transform (Optional[Callable[[Generator[Tuple, None, None]], Generator[Tuple, None, None]]]) –

• global_transform (Optional[Callable[[Generator[Tuple, None, None]], Generator[Tuple, None, None]]]) –

• kwargs –

Returns

Methods

__init__([node_transform, edge_transform, …])	Transform networkx feature data.
generate(datalist)	param datalist
transform(g)

__init__(node_transform=None, edge_transform=None, global_transform=None)

Transform networkx feature data.

def only_self_loops(edges):
    for e1, e2, edata in edges:
        if e1 == e2:
            yield e1, e2, edata

transform = nx_transform(edge_transform=only_self_loops)
transform(graphs)

Alternatively, using the functional programming module:

from caldera.utils.functional import Functional

only_self_loops = Fn.filter_each(lambda x: x[0] == x[1])
transform = nx_transform(edge_transform=only_self_loops)
transform(graphs)

Parameters

• node_transform (Optional[Callable[[Generator[Tuple, None, None]], Generator[Tuple, None, None]]]) –

• edge_transform (Optional[Callable[[Generator[Tuple, None, None]], Generator[Tuple, None, None]]]) –

• global_transform (Optional[Callable[[Generator[Tuple, None, None]], Generator[Tuple, None, None]]]) –

• kwargs –

Returns